Eastern Mediterranean Health Journal La Revue De Santé De La Méditerranée Orientale

Contents

Letter from the Editor...... 241

Research articles AIDS awareness and attitudes among Yemeni young people living in high-risk areas...... 242 HIV/AIDS surveillance in Egypt: current status and future challenges ...... 251 Evidence-based approach to HIV/AIDS policy and research prioritization in the Islamic Republic of Iran...... 259 Changes in tobacco use among 13–15-year-olds between 1999 and 2007: findings from the Eastern Mediterranean Region...... 266 Assessing validity of the adapted Arabic Paediatric Asthma Quality of Life Questionnaire among Eastern Mediterranean Egyptian children with asthma...... 274 Corrections: 1. Hepatitis B and C viral infection: prevalence, knowledge, attitude and practice among Health Journal barbers and clients in Gharbia governorate, Egypt. 2. Nosocomial infections in a neonatal intensive care unit in south-western Saudi Arabia...... 280

Cigarette smoking, hypertension and diabetes mellitus as risk factors for erectile dysfunction in Volume 16 Number 3 La Revue de Santé de upper Egypt...... 281 la Méditerranée orientale Behavioural and clinical factors associated with depression among individuals with diabetes...... 286

Profile of diabetic ketoacidosis at a teaching hospital in Benghazi, Libyan Arab Jamahiriya...... 292 Efficacy of metoclopramide and dexamethasone for postoperative nausea and vomiting: a double-blind clinical trial...... 300

Chlamydia trachomatis and cervical intraepithelial neoplasia in married women in a March 2010 Middle Eastern community...... 304 Pregnancy outcome in women with antiphospholipid syndrome on low-dose aspirin and heparin: a retrospective study...... 308 Accuracy of ultrasound, clinical and maternal estimates of birth weight in term women...... 313 Road traffic fatalities in Qatar, Jordan and the UAE: estimates using regression analysis and the relationship with economic growth...... 318 Drug prescription habits in public and private health facilities in 2 provinces in South Africa...... 324 Gender-specific oral health attitudes and behaviour among dental students in Palestine...... 329 Amalgam use and waste management by Pakistani dentists: an environmental perspective...... 334 Case reports

Disseminated leishmaniasis caused by Leishmania tropica in HIV-positive patients in the Islamic Republic of Iran...... 340 Lupoid leishmaniasis due to Leishmania major with remaining large scars: report of 2 cases...... 344 Young footballers, Islamic Republic of Iran One in every 5 people in the world is an adolescent (aged 10–19 years); 85% of them live in A case of Behçet disease with pulmonary artery pseudoaneurysm: long term follow-up...... 346 developing countries. Nearly two-thirds of premature deaths and one-third of the total disease Ascaris lumbricoides infection: an unexpected cause of pancreatitis in a western Mediterranean country...... 350 burden in adults are associated with conditions or behaviours that began in youth, including tobacco use, lack of physical activity, unprotected sex or exposure to violence. Promoting healthy practices during adolescence ensures longer, more productive lives for the future.

املجلد السادس عرش / عدد Volume 16 / No. 3 3 آذار / مارس March / Mars 2010

Cover 3.indd 1 2/16/2010 11:26:44 AM Members of the WHO Regional Committee for the Eastern Mediterranean املجلة الصحية لرشق املتوسط Afghanistan . Bahrain . Djibouti . Egypt . Islamic Republic of Iran . Iraq . Jordan . Kuwait . Lebanon هى املجلة الرسمية التى تصدر عن املكتب اإلقليمى لرشق املتوسط بمنظمة الصحة العاملية. وهى منرب لتقديم Libyan Arab Jamahiriya . Morocco . Oman . Pakistan . Palestine . Qatar . Saudi Arabia . Somalia السياسات واملبادرات اجلديدة ىف اخلدمات الصحية والرتويج هلا، ولتبادل اآلراء واملفاهيم واملعطيات الوبائية Sudan . Syrian Arab Republic . Tunisia . United Arab Emirates . Republic of Yemen ونتائج األبحاث وغري ذلك من املعلومات، وخاصة ما يتعلق منها بإقليم رشق املتوسط. وهى موجهة إىل كل أعضاء املهن الصحية، والكليات الطبية وسائر املعاهد التعليمية، وكذا املنظامت غري احلكومية املعنية، واملراكز املتعاونة مع منظمة الصحة العاملية واألفراد املهتمني بالصحة ىف اإلقليم وخارجه.

EASTERN MEDITERRANEAN HEALTH JOURNAL البلدان أعضاء اللجنة اإلقليمية ملنظمة الصحة العاملية لرشق املتوسط IS the official health journal published by the Eastern Mediterranean Regional Office of the World Health Organization. It is a forum for the presentation and promotion of new policies and initiatives in health services; and for the exchange of ideas, concepts, epidemiological data, research findings and other information, with special reference to the Eastern Mediterranean Region...... األردن أفغانستان اإلمارات العربية املتحدة باكستان البحرين تونس اجلامهريية العربية الليبية مجهورية إيران اإلسالمية It addresses all members of the health profession, medical and other health educational institutes, interested NGOs, WHO Col- ...... اجلمهورية العربية السورية اجلمهورية اليمنية جيبويت السودان الصومال العراق عُ امن فلسطني قطر الكويت لبنان مرص laborating Centres and individuals within and outside the Region. . املغرب اململكة العربية السعودية LA REVUE DE SANTÉ DE LA MÉDITERRANÉE ORIENTALE EST une revue de santé officielle publiée par le Bureau régional de l’Organisation mondiale de la Santé pour la Méditerranée orientale. Elle offre une tribune pour la présentation et la promotion de nouvelles politiques et initiatives dans le domaine des ser‑vices de santé ainsi qu’à l’échange d’idées, de concepts, de données épidémiologiques, de résultats de recherches et d’autres Membres du Comité régional de l’OMS pour la Méditerranée orientale informations, se rapportant plus particulièrement à la Région de la Méditerranée orientale. Elle s’adresse à tous les professionnels de la santé, aux membres des instituts médicaux et autres instituts de formation médico-sanitaire, aux ONG, Centres collabora- Afghanistan . Arabie saoudite . Bahreïn . Djibouti . Égypte . Émirats arabes unis . République islamique d’Iran teurs de l’OMS et personnes concernés au sein et hors de la Région. Iraq . Jamahiriya arabe libyenne . Jordanie . Koweït . Liban . Maroc . Oman . Pakistan . Palestine . Qatar République arabe syrienne . Somalie . Soudan . Tunisie . République du Yémen

EMHJ is a trilingual, peer reviewed, open access journal and the full contents are freely available at its website: http://www/emro.who.int/emhj.htm

EMHJ is abstracted/indexed in the Index Medicus and MEDLINE (Medical Literature Analysis and Retrieval Systems on Line) and the ExtraMed-Full text on CD-ROM, the Cumulative Index to Nursing and Allied Health Literature (CINAHL), CAB International, Lexis Nexis, Scopus and the Index Medicus for the WHO Eastern Mediterranean Region (IMEMR). Correspondence

Editor-in-chief ©World Health Organization 2010 EMHJ All rights reserved WHO Regional Office for the Eastern Mediterranean P.O. Box 7608 Disclaimer The designations employed and the presentation of the material in this publication do not imply the expression of any opinion Nasr City, Cairo 11371 whatsoever on the part of the World Health Organization concerning the legal status of any country, territory, city or area or of Egypt its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on maps represent approximate border Tel: (+202) 2276 5000 lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products Fax: (+202) 2670 2492/(+202) 2670 2494 does not imply that they are endorsed or recommended by the World Health Organization in preference to others of a similar Email: [email protected]/[email protected] nature that are not mentioned. All reasonable precautions have been taken by the World Health Organization to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either express or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the World Health Organization be liable for damages arising from its use. The named authors alone are responsible for the views expressed in this publication.

Subscriptions and Distribution Enquiries regarding subscriptions and distribution of the print edition of EMHJ should ISSN 1020-3397 be addressed to: Printing and Marketing of Publications at: email: [email protected]; tel: (+202) 2276 5000; fax: (+202) 2670 2492 or 2670 2494

Permissions Cover designed by Diana Tawadros Requests for permission to reproduce or translate articles, whether for sale or Internal layout designed by Emad Marji and Diana Tawadros non-commercial distribution should be addressed to Printed by WHO Regional Office for the Eastern Mediterranean EMHJ at: [email protected]

Cover 6.indd 2 5/24/2010 11:46:47 AM Eastern Mediterranean La Revue de Santé de Health Journal la Méditerranée orientale

املجلد السادس عرش عدد Vol. 16 No. 3 • 2010 • 3

Contents Letter from the Editor...... 241 Research articles AIDS awareness and attitudes among Yemeni young people living in high-risk areas A.W. Al-Serouri, M. Anaam, B. Al-Iryani, A. Al Deram and S. Ramaroson...... 242 HIV/AIDS surveillance in Egypt: current status and future challenges S. Boutros and J. Skordis ...... 251 Evidence-based approach to HIV/AIDS policy and research prioritization in the Islamic Republic of Iran A. Feizzadeh, S. Nedjat, S. Asghari, A. Keshtkar, R. Heshmat, H. Setayesh and R. Majdzadeh...... 259 Changes in tobacco use among 13–15-year-olds between 1999 and 2007: findings from the Eastern Mediterranean Region F. El-Awa, C.W. Warren and N.R. Jones...... 266 Assessing validity of the adapted Arabic Paediatric Asthma Quality of Life Questionnaire among Egyptian children with asthma R. Abdel Hai, E. Taher and M. Abdel Fattah...... 274 Corrections: 1. Hepatitis B and C viral infection: prevalence, knowledge, attitude and practice among barbers and clients in Gharbia governorate, Egypt. 2. Nosocomial infections in a neonatal intensive care unit in south-western Saudi Arabia...... 280 Cigarette smoking, hypertension and diabetes mellitus as risk factors for erectile dysfunction in upper Egypt H. Zedan, A.A. Hareadei, A.A. Abd-Elsayed and E.M. Abdel-Maguid...... 281 Behavioural and clinical factors associated with depression among individuals with diabetes Z. Yekta, R. Pourali and R. Yavarian...... 286 Profile of diabetic ketoacidosis at a teaching hospital in Benghazi, Libyan Arab Jamahiriya R.R. Elmehdawi and H.M. Elmagerhei ...... 292 Efficacy of metoclopramide and dexamethasone for postoperative nausea and vomiting: a double-blind clinical trial M. Entezariasl, M. Khoshbaten, K. Isazadehfar and G. Akhavanakbari...... 300 Chlamydia trachomatis and cervical intraepithelial neoplasia in married women in a Middle Eastern community M. Valadan, F. Yarandi, Z. Eftekhar, S. Darvish, M.S. Fathollahi and A. Mirsalehian...... 304 Pregnancy outcome in women with antiphospholipid syndrome on low-dose aspirin and heparin: a retrospective study T. Naru, R.S. Khan and R. Ali...... 308 Accuracy of ultrasound, clinical and maternal estimates of birth weight in term women T. Ashrafganjooei, T. Naderi, B. Eshrati and N. Babapoor...... 313 Road traffic fatalities in Qatar, Jordan and the UAE: estimates using regression analysis and the relationship with economic growth A. Bener, S.J. Hussain, M.A. Al-Malki, M.M. Shotar, M.F. Al-Said and K.S. Jadaan...... 318 Drug prescription habits in public and private health facilities in 2 provinces in South Africa G. Mohlala, K. Peltzer, N. Phaswana-Mafuya and S. Ramlagan ...... 324 Gender-specific oral health attitudes and behaviour among dental students in Palestine E. Kateeb ...... 329 Amalgam use and waste management by Pakistani dentists: an environmental perspective R. Mumtaz, A. Ali Khan, N. Noor and S. Humayun ...... 334 Case reports Disseminated leishmaniasis caused by Leishmania tropica in HIV-positive patients in the Islamic Republic of Iran S. Jafari, M. Hajiabdolbaghi, M. Mohebali, H. Hajjaran and H. Hashemian ...... 340 Lupoid leishmaniasis due to Leishmania major with remaining large scars: report of 2 cases G. Sadeghian, H. Ziaei, L. Shirani-Bidabadi and M.A. Nilforoushzadeh...... 344 A case of Behçet disease with pulmonary artery pseudoaneurysm: long term follow-up M-H.S. Modaghegh, G. H. Kazemzadeh and M. H. Jokar...... 346 Ascaris lumbricoides infection: an unexpected cause of pancreatitis in a western Mediterranean country A. Galzerano, E. Sabatini and D. Durì...... 350 Belgacem Sabri MD MPA, MA (Econ), Editor-in-chief Muhammad Afzal MSc, MPhil, PhD, Executive Editor Editorial Board Mohammad Abdur Rab MBBS, DTM&H, MPH&TM, PhD Mohamed M. Ali BSc, MSc, PhD, DTMH Abdulla S. Assaedi MBBS, MPH Mounir Farag MD, DGS, DEmS, DPH Abdul Ghaffar MD, MPH, MHA, PhD Malekafzali Hossein MK, MPH, PhD Jaouad Mahjour MD, MPH Mamunur Rahman Malik MBBS, Dip (Health Economics), MSc, MPhil Kassem Sara MD

International Advisory Panel Dr S. Aboulazm. Professor of Orthodontics. Egypt Dr Abdul Rahman Al-Awadi BSc, MD, MPH, Honorary FRCM, Ireland Dr Law, Korea, Honorary FRCS & P, Glasgow, FRCP, Edinbugh. Kuwait Dr Fariba Al-Darazi RN, MSc, PhD. Bahrain Dr M. Al-Nozha, MD, FRCP, FACC, FESC. Professor of Medicine and Consultant Cardiologist. Saudi Arabia Dr Ala’din Alwan MD, FRCP, FFPHM. Iraq Dr F. Azizi. Professor of Internal Medicine and Endocrinology. Islamic Republic of Iran Dr K. Bagchi BSc, MD, PhD. India Professor K. Dawson BA, MD, PhD, FRCP, FRACP, FRCPCH, DObst, RCOG. New Zealand Professor Kaussay Dellagi MD. Tunisia Dr R. Dybkaer MD. Denmark Dr M. Aziz El-Matri. Professor of Medicine. Tunisia Professor F. El-Sabban BSc, MS, PhD. United States of America Dr A.H. El-Shaarawi MSc (Stat), PhD (Stat). Canada Professor N. Fikri-Benbrahim PhD (Pub health) (SocSci). Morocco Professor A.T. Florence BSc (Pharm), PhD, DSc, FRSC, FRPharmS, FRSE. United Kingdom Professor Cheherezade M.K. Ghazi BS (Nursing), MS (Nursing), DPH, MPA. Egypt Professor M.A. Ghoneim MD, MD (Hons). Egypt Dr J.A. Hashmi DTM&H, FRCP. Pakistan Professor J. Jervell MD, PhD. Norway Professor G.J. Johnson MA, MD, BChir, FRCS (C), FRCOphth, DCEH. United Kingdom Dr M. Kassas. Emeritus Professor of Plant Ecology. Egypt Professor M.M. Legnain MBBS, MRCOG, FRCOG. Libyan Arab Jamahiriya Professor El-Sheikh Mahgoub DipBact, PhD, MD, FRCPath. Sudan Professor A.M.A. Mandil MSc (Paediatr), MPH, DrPH. Egypt Professor A.B. Miller MB, FRCP. Canada Professor S.S. Najjar MD. Lebanon Dr Abubaker A. Qirbi BSc, MD (Edin), FRCPC (Can), FRCP FRCPath (UK). Republic of Yemen Professor O.S.E. Rasslan MD, PhD. Egypt Professor W.A. Reinké MBA, PhD. United States of America Professor I.A. Sallam, MD, Dip High Surgery Cairo, Honorary FRCS, PhD (Glasgow), LRCP, MRCS, FRCS (London), ECFMG. Egypt Dr C.Th.S. Sibinga FRCP (Edin), FRCPath. The Netherlands Mr Taoufik Zeribi Eng BSc, MSc. Tunisia Editors Fiona Curlet, Eva Abdin, Alison Bichard, Guy Penet

Graphics Suhaib Al Asbahi, Hany Mahrous, Diana Tawadros

Administration Nadia Abu-Saleh, Nagwan Behairy, Yasmine El Sakhawy املجلد السادس عرش املجلة الصحية لرشق املتوسط العدد الثالث

Letter from the Editor

On the occasion of the 30th anniversary of the Declaration of Alma-Ata, the Member States of the World Health Organization in the Eastern Mediterranean Region reaffirmed in the Qatar Declaration in November 2008 that the primary health care approach is the main strategy to achieve better health and well-being for the people of the Region. The Member States reiterated their commitment to primary health care and its values and principles of universal access and coverage, based on equity and social justice, community involvement and health as a basic human right. As part of the implementation of the Qatar Declaration, which called for partnership with academia and civil society organizations to support public health and primary health care revival, the Regional Office for the Eastern Mediterranean convened a meeting to gather academicians and researchers in public health together with policy-makers in Beirut in December 2009. In the Beirut meeting, the role of academia in support of public health covering training, research and practice was highlighted, the areas of potential support in the Region were deliberated upon, and the limited use of research findings in policy analysis and development was discussed. A particular focus of the meeting was sharing of the experience of networking among academic institutions in the Region. In this context the Maghreban Network on Health Systems and Health Economics provided a good model, underscoring the contribution of academic institutions to training of health professionals on the use of analytical tools to generate evidence on health care financing and burden of disease analysis. One of the major outcomes of the Beirut meeting was a consensus among the participants to establish a Regional Network of Academic Institutions in support of public health and revival of primary health care in the Region. It was recommended that the Regional Office support the Network in order to facilitate its growth through its promotion inside and outside the Region as well as through collaborative activities for training, research and practice. Partnership of WHO EMRO with some academic and research institutions has contributed in the past to capacity development in the Region as well as institutional strengthening of ministries of health at national and subnational levels. The proposed Regional Network will help strengthen collaboration between academia and research institutions for their joint input to policy-making for health care in the Region.

رسالة من املحرر يفالذكرى الثالثني إلعالن أملا أتا، أعادت الدول األعضاء يف إقليم رشق املتوسط ملنظمة الصحة العاملية التأكيد يف إعالن قطر يف ترشين الثاين/نوفمرب 2008عىل أن أسلوب الرعاية الصحية األولية هو االسرتاتيجية الرئيسية لتحقيق درجة أفضل من الصحة واملعافاة للناس يف اإلقليم. وقد أعاد البلدان األعضاء التأكيد عىل التزامهم بالرعاية الصحية األولية وبقيمتها ومبادئها املتمثلة باإلتاحة والتغطية الشاملة، واالستناد إىل العدالة االجتامعية واإلنصاف، وإسهام املجتمع، والنظر إىل الصحة عىل أهنا أحد احلقوق األساسية لإلنسان. وكجزء من تنفيذ إعالن قطر الذي دعا إىل الرشاكة مع األوساط األكاديمية، ومعمنظامت املجتمع املدين لدعم الصحة العمومية، وإلحياء الرعاية الصحية األولية، فقد عقد املكتب اإلقليمي لرشق املتوسط اجتامعاً استهدف اجلمع بني األكاديميني والباحثني يف ميادين الصحة العمومية معاً،إىل جانب أصحاب القرار السيايس، وذلك يف بريوت يف شهر كانون األول/ديسمرب عام 2009. ويف اجتامع بريوت تم توضيح دور األوساط األكاديمية يف تقديم الدعم للتدريب وللبحوث وللمامرسات يف الصحة العمومية؛ وتداول املجتمعون حول املجاالت التي حيتمل أن تتلقى الدعم يف اإلقليم، إىل جانب التداول حول االنتفاع املحدود النطاق ملا تسفر عنه البحوث من موجودات ونتائج يف حتليل وتطوير السياسات. وقدركّز االجتامع تركيزاً خاصاً عىل تبادل اخلربات املكتسبة يف إعداد الشبكات التي جتمع املؤسسات األكاديمية يف اإلقليم. ويف هذا املجال قدمت الشبكة املغاربية للنُظُ م الصحية واالقتصاديات الصحية مثاالً جيداً، معاألخذ باالعتبار ما تقدمه املؤسسات األكاديمية من إسهامات لتدريب أرباب املهن الطبية عىل استخدام أدوات التحليل يف جتميع البيانات حول متويل الرعاية الصحية وحول حتليل عبء املرض. ومناحلصائل الرئيسية التي متخَّ ض عنها اجتامع بريوت اتفاق اآلراء لدى املسامهني به عىل إنشاء الشبكة اإلقليمية للمؤسسات األكاديمية، لتقديم الدعم للصحة العمومية، وإلحياء الرعاية الصحية األولية يف اإلقليم. وقد أوىص املجتمعون املكتب اإلقليمي بتقديم الدعم هلذه الشبكة، لكي يصبح نموها وتطورها أمراً ميسوراً بتعزيزها داخل وخارج اإلقليم، وبالقيام بأنشطة تعاونية للتدريب والبحوث واملامرسات. وقدسبقأن سامهت الرشاكة بني املكتب اإلقليمي ملنظمة الصحة العاملية لرشق املتوسط مع بعض املؤسسات األكاديمية والبحثية يف املايض، يف تطوير القدرات يف اإلقليم، إىل جانب تعزيز القوى املؤسسية لوزارات الصحة، عىل الصعيد الوطني ودون الوطني. وستساعد الشبكة اإلقليمية املقرتحة يف تعزيز التعاون بني األوساط األكاديمية ومؤسسات البحوث، للحصول عىل نتائج مشرتكة، تقدم ألصحاب القرار السيايس، وتساعدهم يف اختاذ القرارات حول الرعاية الصحية يف اإلقليم.

241 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

AIDS awareness and attitudes among Yemeni young people living in high-risk areas A.W. Al-Serouri,1 M. Anaam,2 B. Al-Iryani,3 A. Al Deram4 and S. Ramaroson3

الوعي واملواقف من اإليدز بني الشباب اليمنيني يف املناطق املعرضة للخطورة العالية عبد الواحد الرسوري، مهيوب أنعم، بثينة اإلرياين، أروى الدرام، سلوف رامرسون

اخلالصـة:ثمة قلق حول احتامل انتشار العدوى بفريوس َالع َوز املناعي البرشي بني املجموعات املستضعفة َّواملعرضة خلطورة عالية، رغم معدل االنتشار املنخفض هلا. وقد أجرى الباحثون دراسة جمتمعية عام 2005حول الوعي باإليدز واملواقف من اإليدز لدى 601 ًشابا ترتاوح أعامرهم بني 15 – 24سنة وهم معرضون خلطر مرتفع يف املناطق املجاورة لعدن. واتضح من الدراسة أن الشبان يفتقدون املعلومات املالئمة عن اإليدز والعدوى 28 3 46 89 بفريوس َالع َز واملناعي البرشي، فرغم أن % منهم قد سمع به فإن % منهم استطاع ذكر طرق لإلصابة بالعدوى به، و % منهم استطاع ذكر 3 طرق ُّلتجنبالعدوى به. وباملقابل كانت املفاهيم اخلاطئة عن أنامط الرساية منترشة بني هؤالء الشباب فمعظمهم كان يعتقد أهنم ال يواجهون أي خطر أو يواجهون ً خطرا ً. ضئيالوكان لدهيم مواقف عدم ُّحتمل للمرىض املصابني باإليدز. وكام كان نصف الشباب يعرفون بوجود البغاء واللواطة يف املناطق التي يعيشون فيها.

ABSTRACT Despite the low rate of infection in Yemen, there are concerns about the possible spread of HIV among high-risk and vulnerable groups. A community-based study was made in 2005 of AIDS awareness and attitudes among 601 young people aged 15–24 years from low-income, high-risk neighbourhoods in Aden. Young people lacked proper information about HIV/AIDS. Although 89% had heard of AIDS, fewer (46%) could name 3 ways of transmission or 3 ways to avoid infection (28%). Misconceptions about modes of transmissions were prevalent and many young people believed that they faced little or no risk. There were intolerant attitudes towards AIDS patients. About half the young people knew that prostitution and homosexuality existed in their area.

Sensibilisation au sida et attitudes des jeunes Yéménites vivant dans des quartiers à haut risque

RÉSUMÉ Malgré le faible taux d’infection au Yémen, la propagation possible du VIH parmi les groupes à haut risque et vulnérables suscite des préoccupations. Une étude communautaire, sur la sensibilisation au sida et les attitudes parmi 601 jeunes âgés de 15 à 24 ans issus des quartiers à haut risque d’Aden, caractérisés par de faibles revenus, a été conduite en 2005. Les jeunes gens manquaient d’informations correctes sur le VIH/sida. Bien que 89 % d’entre eux aient entendu parler du sida, ils étaient nettement moins nombreux (46 %) à pouvoir citer trois modes de transmission ou trois moyens d’éviter l’infection (28 %). Les idées reçues sur le mode de transmission étaient fréquentes et de nombreux jeunes étaient persuadés de courir très peu de risques, voire aucun. Des attitudes intolérantes vis-à-vis des malades du sida ont été observées. Environ la moitié des jeunes savaient que la prostitution et l’homosexualité existaient dans leur quartier.

1Department of Community Medicine, Faculty of Medicine and Health Sciences; 2Faculty of Education, University of Sana’a, Sana’a, Yemen (Correspondence to A.W. Al-Serouri: [email protected]). 3United Nations Children’s Fund, Sana’a, Yemen. 4SOUL for the Development of Women and Children, Sana’a, Yemen. Received: 05/07/07; accepted: 10/12/07

242 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Introduction HIV/AIDS and self-protection from an alpha error of 0.05 (2-tailed) (i.e. infection and their attitudes towards acceptable 95% confidence limit 0.4 to Young people aged 15–24 years remain people living with HIV and AIDS. The 0.6)—was estimated as 100 for each at the centre of the global HIV/AIDS findings will provide baseline indicators of the 4 communities. Due to potential pandemic in terms of transmission and to help design and evaluate a package nonresponse, a much larger sample size vulnerability [1]. Nevertheless, young of integrated interventions supported of 601 was studied. people present an opportunity to halt by the United Nations Children’s Fund The sampling unit (the household) the spread of the epidemic [2]; in coun- (UNICEF). was selected by systematic sampling tries where the spread of HIV/AIDS is methods. Initial mapping by the study subsiding or declining, such as Thailand team helped to determine the sampling and Uganda, it is primarily because Methods frame. According to the area size, each young men and women are being given area was divided into 5–10 sections The present study, conducted from the knowledge, tools and services to and each section was divided into 5 to March to September 2005, was a adopt safe behaviours [2]. The United 10 blocks with each block comprising community-based, cross-sectional Nations has resolved that by 2010 at 25–40 households. First, 2 to 4 sections study of young people aged 15–24 years least 95% of young men and women were selected randomly (by random living in 4 districts in the city of Aden, aged 15–24 years should have access the economic capital and the main port tables) from each area and then similar- to the information, education and serv- of Yemen. ly 4–7 blocks from each section. Finally, ices necessary to develop the life skills 20–25 households were selected from required to reduce their vulnerability to Study area and sample each block by a systematic sampling HIV infection [3]. The 4 areas were Abdul Qaui and Al method where the sampling interval Yemen is still considered a country Memdarah (inhabited mainly by Yem- was calculated by dividing the total with a low rate of HIV infection, with eni citizens), Al Basateen (inhabited number of houses in the block by the an accumulated number of registered mainly by Somali refugees) and Al Se- number of houses in the sample. The cases by the year 2004 of 1549 and saban (inhabited by Yemeni margin- number of the first house in the sample an estimated number of 15 000 [4]. alized citizens known as al-akhdam). was selected from simple random tables Nevertheless, recent situation analyses Al-akhdam, who live in isolation from and within the sampling interval. The cited conditions that could facilitate the rest of society, are stereotyped as next house was determined by adding the spread of HIV among high-risk and dishonourable, immoral, dependent the sampling interval to the previous vulnerable groups as well as the popula- and dirty. They exist at the bottom of house number in the sample and so on. tion in general [5,6]. These include: the class system of Yemeni society and Only households containing males high rates of illiteracy, low quality and are forbidden to socialize or intermarry and/or females aged between 15 and limited access to health care services, with any other classes. Although a grow- 24 years were eligible for inclusion in the high prevalence of sexually transmitted ing number of al-akhdam children have sample. If a household had no members infections, population movements to started going to school, large numbers in the target age group, the interviewer and from countries with high HIV infec- continue to be unenrolled or drop out continued sampling as previously de- tion rates and inadequate public aware- soon after joining because of financial scribed until the required sample size ness. The HIV/AIDS national strategy, constraints [9]. These 4 low-income was achieved. A maximum of 2 eligible endorsed in 2002, sets the framework neighbourhoods are recognized as high persons were interviewed from each for effective and coordinated interven- risk due to the large numbers of com- household. During pilot studies we tions for the prevention and control of mercial sex workers living there and found that for cultural reasons the HIV/AIDS [7]. because they represent transit hubs for youngest females were less likely to be Research data on young people’s migrant populations coming from the allowed to participate by the head of knowledge about how HIV is transmit- Horn of Africa, where the prevalence the household. Even when permitted, ted and their perspectives on whether of HIV/AIDS is higher than in Yemen we found that the females lagged far HIV is an issue that is directly relevant [10]. behind their male peers in exposure to them are scarce in Yemen [8]. This The minimum sample size—keep- to knowledge in general and to HIV/ paper presents the findings of a house- ing the expected parameter (propor- AIDS in particular and rarely respond- hold survey that aimed to establish a tion of young people with satisfactory ed to questions. They were also more clear understanding of young people’s knowledge) at 0.5 and with an accept- likely to withdraw during the interview. knowledge about the transmission of able deviation of 0.1 on each side at Therefore, the study team was obliged

243 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

to use the following selection criteria conducted in private in a quiet room Results when there were more than 2 eligible or area in the household. Respondents individuals at a household: if more than were informed of the scope of the study A total of 601 young people aged 15–24 2 eligible persons but only 1 female, years participated in the study. Table 1 and were assured of full confidential- interview that female and the youngest shows the demographic characteristics male; if more than 2 eligible persons ity. After initial rapport-building, the of the study sample. but only 1 male, interview that male and questions were asked in a personal face- the oldest female; if more than 2 eligible to-face interview. Although about 15% General awareness about HIV infection/AIDS persons at a household and all were of the eligible young people refused males, interview the oldest and young- Basic awareness—as measured by re- to participate at the beginnning , this est males; if more than 2 eligible persons sponses to the question whether they at a household and all were females, dropped to only 6% after explanation. had ever heard about AIDS—was interview the 2 oldest females; if more Data analysis generally high (89%) (Table 2). Such than 2 eligible persons at a household knowledge was significantly related and of both sexes, interview the oldest All collected data were coded, entered to sex, as significantly more males had female and the youngest male. and analysed using SPSS, version 13.0. heard about AIDS than had females Questionnaire The questionnaire was adapted from Table 1 Characteristics of the 601 young people enrolled in the study the core interview schedule on AIDS Characteristic No. % knowledge, attitudes, belief and practices Sex of the Social and Behavioural Research Female 342 56.9 Unit of the World Health Organization Male 259 43.1 (WHO) Global Programme on AIDS. Education The Arabic version of this questionnaire 137 22.8 was prepared by the WHO Regional Illiterate 50 8.3 Office for the Eastern Mediterranean Read and write and was previously tested in the HIV/ Basic 247 41.1 AIDS Situation and Needs Assessment Secondary 140 23.3 Study in Yemen in 2001; it proved to be Diploma 9 1.5 acceptable and reliable in the Yemeni University 18 3.0 context [5]. Additional questions were Marital status extracted from the UNICEF Multiple Single 453 76.0 Indicator Cluster Survey question- Married 136 22.8 naire [11]. The questions were a mix of Widowed 1 0.2 closed- and open-ended questions that Divorced 6 1.0 covered sociodemographic characteris- Occupation tics, knowledge, attitudes and practices. Student 195 32.4 The draft questionnaire was piloted on Housewife 171 28.5 25 young people and was refined ac- Unemployed 145 24.1 cordingly. The subjects of the pilot study Salaried employee 32 5.3 were not included in the final analysis. Skilled labourer 21 3.5 The questionnaire was administered Non-skilled labourer 23 3.8 by qualified male and female fieldwork- Other 14 2.3 ers. They were trained in administration Social category of the questionnaire by the first author, Yemeni citizen 373 62.0 who also supervised the entire conduct Yemeni citizen: marginalizeda 156 26.0 of the study. Male respondents were Refugee 66 11.0 interviewed by male fieldworkers and Yemeni citizen: returneeb 6 1.0 female respondents by female fieldwork- aAl-akhdam. ers. As far as possible, interviews were bLiving/working aboard and evacuated after Gulf war.

244 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 2 General awareness of HIV/AIDS and knowledge about transmission and methods of prevention among young people in Aden (n = 601) Item Yes No Don’t know No. % No. % No. % Had heard about AIDS 530 89 68 11 – – Knew at least 3 correct modes of AIDS transmission 239 45 291 55 – – Knew at least 3 correct modes of AIDS prevention 148 28 382 72 – – AIDS causative agent is a virus 197 37 64 12 269 51 AIDS patient may have no symptoms or signs 186 35 248 47 96 18 AIDS is an infectious disease 488 92 16 3 27 5 AIDS is a curable disease 88 17 349 65 94 18 Condoms are an effective preventive measure 152 29 148 28 232 43

n = total sample; data are missing for some items.

(94% versus 84%) (χ2 = 13.8, P < 0.001) some believed that HIV/AIDS could reluctant to take the test than males (8% [odds ratio (OR) = 2.1, 95% confidence be acquired through shaking hands versus 15% would refuse) (P < 0. 001). interval (CI): 1.3–3.3]. Illiteracy was and touching (28%). There was only Refugees expressed more reluctance to another significant determinant of basic a small difference in misconceptions take the HIV test than citizens or mar- awareness about AIDS (24% versus of the modes of transmission between ginalized groups (18%, 12% and 10% 8%) (χ2 = 28.8, P < 0.001) (OR = 3.2, males and females. However, knowl- respectively would refuse) (P < 0.01). 95% CI: 2.1–4.9). edge of possible modes of transmission Half of the respondents who would However, when HIV/AIDS aware- and misconceptions were significantly be willing to take the HIV test (49%) ness was measured by knowledge of at influenced by social category; a higher knew that the test could be taken at a least 3 correct modes of AIDS transmis- percentage of those from marginalized hospital, 24% said it could be taken in sion and 3 correct methods of AIDS groups (36%) stated that AIDS could any laboratory, 3% said it could only be prevention, the proportions answering be transmitted by shaking hands and taken at a specialized AIDS centre, while correctly dropped to 45% and 28% re- touching compared with 25% and 26% 1% believed that the test could only be spectively. Refugees and marginalized among refugees and Yemeni citizens taken abroad and 13% did not know people were less likely to know 3 cor- respectively (P < 0.05). how or where to take an HIV test. Respondents thought that the gov- rect modes of transmission or preven- Attitudes towards AIDS tion compared with Yemeni citizens ernment should take a harsher approach (33%, 41% and 49% respectively) (P < Of the respondents, 94% stated that to AIDS patients; 13% stated that the 0.05). Such knowledge also increased as AIDS was a serious problem and that it government should kill AIDS patients, education level increased (no education could potentially threaten the country’s while 18% of the respondents felt that 76% versus highly educated 100%) (P future (Table 4). Around three-quarters AIDS patients should be imprisoned < 0.001). Only 29% of the respondents of the sample would be willing to look and 78% thought that the government knew that condoms were an effective after relatives who contracted HIV/ should isolate and quarantine AIDS preventive measure, with a significant AIDS. Nevertheless, there was a com- patients in hospitals away from the rest sex difference (43% of males versus 16% mon attitude that AIDS patients needed of the community. of females) (P < 0.0001). to be isolated and should receive special When asked what role the commu- care in special health settings and spe- nity should play towards people living Knowledge about modes of cialized staff. A few respondents (1%) with AIDS, a small percentage (1%) transmission believed there was no need to care for suggested that the community should A large majority of young adults had AIDS patients and that they should be report suspected AIDS/HIV cases to correct knowledge about the major killed. the police. About half of respondents modes of transmission of HIV, such as As for voluntary testing and coun- suggested that the community should sexual intercourse, sharps/instruments, selling, 86% of respondents were will- not allow HIV/AIDS patients to be blood transfusions and homosexual ing to be tested for HIV/AIDS and integrated within society, while the contact (Table 3). Nevertheless, there 12% would refuse (2% didn’t know) remaining half suggested that the com- were still important misconceptions as (Table 4). Females, however, were less munity should provide assistance and

245 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Table 3 Knowledge about specific modes of transmission of AIDS among young people in Aden n( = 601) Mode of transmission Yes No Don’t know No. % No. % No. % Mode of transmission (correct) Extramarital sex 506 95 11 2 14 3 Having sex with an infected person 508 95 7 1 18 4 Injections 499 94 23 4 9 2 Blood transfusions 486 92 17 3 27 5 Sharp blades/instruments 477 89 30 6 26 5 Male-to-male sex 444 84 31 6 55 10 Mother-to-child transmission: During pregnancy 404 76 46 9 81 15 During labour 328 62 62 12 141 26 During lactation 343 64 75 14 113 21 Mode of transmission (misconceptions) Mosquitoes/insects 352 67 79 15 97 18 Deep kissing 299 56 132 25 102 19 Contact with clothes 263 50 157 30 110 21 Drinking/eating 246 46 73 14 212 40 Swimming pool 221 41 134 25 178 33 Contact with bathroom items 212 40 189 36 129 24 Touching people 152 28 317 60 63 12

n = total sample; data are missing for some items.

be sympathetic towards them. Other with AIDS. Additionally, 74% stated patients (12%). Other less commonly suggestions were that infected individu- that a teacher who is infected with AIDS mentioned actions were checking air- als should divorce their partners (18%), should not continue teaching. ports and testing foreigners for HIV/ report to the nearest hospital (6%) or A majority of the respondents AIDS (7%), ensuring blood safety (4%) that the community should provide (72%) strongly believed that the gov- and imposing prevention and control financial support to AIDS patients for ernment should take action to prevent measures (12%). medicines (2%). the spread of AIDS in the country. The Perceptions of risk As far as attitude towards contact most commonly chosen actions were with HIV/AIDS patients, 77% of re- awareness raising (27%), prevention As many as 28% of respondents believed spondents stated that they were not of prostitution (24%), isolation of pa- that there was no possibility whatsoever willing to buy groceries from a person tients (19%) and imprisoning AIDS that they themselves could get infected

Table 4 Attitudes towards AIDS and people living with HIV/AIDS among young people in Aden (n = 601) Item Yes No Don’t know No. % No. % No. % AIDS is one of the most dangerous and important diseases facing Yemen 482 94 11 2 22 4 Willing to look after relatives with AIDS 415 78 101 19 16 3 AIDS patients need to be isolated and cared for at special AIDS hospitala 415 78 – – – – Willing to buy groceries from a person with AIDS 111 21 410 77 11 2 Teacher infected with AIDS should continue teaching 122 23 394 74 16 3 Willing to be tested for AIDS 458 86 62 12 13 2

a”No” and “Don’t know” categories were not applicable. n = total sample; data are missing for some items.

246 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 5 Perceptions of risk of infection with HIV/AIDS and behavioural changes to avoid infection among young people in Aden (n = 601) Item Yes No Don’t know No. % No. % No. % Possibility of getting infected him/herself 230 62 149 28 51 10 Which people are at high risk of AIDS?a Those who have extramarital sex 352 63 – – – – Homosexuals 12 2 – – – – Receivers of infected blood 4 1 – – – – Users of contaminated syringes or surgical 4 1 – – – – instruments Foreigners, especially Somalis 61 12 – – – – Which people are at low risk of AIDS?a – – – – Those who follow their religion 149 28 – – – – Those who avoid extramarital sex 128 24 – – – – Those with a single partner 71 13 – – – – Those aware about transmission methods 20 4 – – – – Changed behaviour after knowing about AIDS 178 33 323 61 32 6 Behavioural changes (correct)a Avoiding extramarital sexual relations 62 35 – – – – Not sharing blades 44 25 – – – – Keeping away from homosexuality 34 19 – – – – Not sharing needles 22 12 – – – – Checking blood before transfusion 12 7 – – – – Behavioural changes (incorrect)a Not sharing other’s belongings 19 33 – – – – Avoiding contact with patients 12 7 – – – – Avoiding crowds in khat sessions 3 2 – – – – Avoiding swimming pools 2 1 – – – – Avoiding open foods 2 1 – – – –

aRespondents answered query directly. “No” and “Don’t know” categories were not applicable. n = total sample; data are missing for some items.

(Table 5). When asked which people e.g. by avoiding extramarital sexual rela- males had heard about prostitution were at high risk of acquiring AIDS, the tions (35%) and not sharing shaving in their area compared with only one- most common group mentioned was blades/razors (25%). Some erroneous third of females (73% versus 34%) (P < those who had extramarital sex (63%). behavioural changes were mentioned, 0.001). Furthermore, there were signifi- Other important risk groups, e.g. homo- e.g. not sharing other people’s clothes cant differences between geographical sexuals and receivers of infected blood and food utensils (33%) and avoiding areas, with a higher percentage of the were mentioned by fewer respondents contact with HIV/AIDS patients (7%). respondents in Al Basateen (mainly (2% and 1% respectively), but 12% be- Occurrence of AIDS cases and inhabited by refugees) and Al Sesaban lieved that foreigners, especially Somalis, risky behaviours in the study (mainly inhabited by marginalized were at high risk. Respondents believed area citizens) aware of the existence of pros- that the people who were at low risk More than half of the respondents titution (69% versus 57%) compared were those who were religious (28%) (59%) knew that there were potential with 41% in Abdul Qaui and 47% in and those who avoided extramarital sex cases of AIDS in Aden. Al Memdarah (inhabited mainly by (24%). A great majority (86%) stated that Yemeni citizens) (P < 0.01). Only one-third of the young people prostitution existed in society in general, Of the respondents, 84% thought mentioned that they had changed their and 52% had heard that prostitution that homosexual practices existed behaviour after knowing about AIDS, existed in their area. Three-quarters of among society in general, while 47%

247 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

had heard that such practices existed Table 6 Sources of information about HIV/AIDS and discussions with others in their area. Male respondents were among young people in Aden (n = 601) more aware of the presence of homo- Item Yes No sexual relations in the study area than No. % No. % were females (62% versus 34%) (P < Current sources of information about 0.001). Residents from Al Basateen and AIDS 499 83 – – Al Sesaban were more aware of such Television 288 48 – – relations (55% each) compared with Friends/relatives residents from Abdul Qaui (38%) and Radio 234 39 – – a Al Memdarah (42%) (P < 0.01). Newspapers 180 30 – – Schoolteachers 175 29 – – Sources of information Social gatherings 138 23 – – Television was the main source of Health workers 72 12 – – information about AIDS (83%), fol- Religious leaders 66 11 – – lowed by information from relatives/ Posters/pamphlets 24 4 friends (48%) (Table 6). More males Preferred source of information about used radio and newspapers as sources AIDS 532 89 – – of information than did females (73% Television versus 34%) (P < 0.001). Receiving in- Radio 361 60 – – formation about AIDS from relatives/ Newspapersa 249 42 – – friends was also more common among Schoolteachers 219 36 – – males than females (60% versus 38%, Health workers 187 31 – – χ2 = 27.0, P < 0.001) (OR 1.5, 95% CI: Religious leaders 170 28 – – 1.3–1.8). Health workers were less men- Friends 89 15 – – tioned as a source of AIDS information Family 72 12 – – as were religious leaders and posters/ Discussed HIV/AIDS pamphlets. With friends 285 54 246 46 Only 34% had discussed HIV/ With family members 182 34 348 66 AIDS-related matters with family mem- aLiterate respondents only. bers, while 54% had discussed such n = total sample; data are missing for some items. matters with friends. Females were less likely to talk to their friends about Before discussing the findings of this which may have led to higher levels of AIDS than males (28% versus 61%, P study we would like to mention some awareness than expected. Last but not < 0.001). limitations. Discussions on sexual mat- least, the final stage of sampling that ters and issues related to HIV/AIDS involved putting criteria for selection Discussion are taboo in Yemen and therefore some of interviewees when there were more difficulties were faced during the inter- than 2 eligible persons, led to inclusion Young people’s ability to protect them- viewing process. We were unable to ask of a higher parentage of females in the selves from HIV ultimately depends some direct questions about personal older age group. on their own safe behaviour, which to behaviours which would have been Nevertheless, the findings of this a large extent, but not exclusively, de- important to know. In some cases, and study are still important as it is the first pends on their knowledge about how despite repeated assurance by the data published study about young people’s HIV is transmitted and their perspec- collectors, mothers did not allow the awareness and attitude in high-risk areas tives on whether HIV is an issue that is interviewers to interview their children in Yemen. The study shows that young directly relevant to them. [12]. Baseline alone, especially females, and their pres- people lacked proper information knowledge, attitudes and practices stud- ence may have influenced the response about HIV/AIDS. Although 89% had ies are very useful tools prior to any to certain questions. The fieldwork was heard of AIDS, only 46% could name interventions, to assess the extent to also conducted after some health edu- 3 ways of transmission and only 28% which an individual or a community are cation activities had been conducted by could name 3 ways to avoid infection. in a position to adopt risk-free behav- national and international nongovern- A major United Nations study found iours [13]. mental organizations in the study areas “an alarming lack of knowledge about

248 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

HIV/AIDS” among young people and Denial about the risk of AIDS is spread of HIV/AIDS. The evidence for that they did not have the proper knowl- widespread even when there is knowl- this is growing; in countries with severe edge to protect themselves. In Ukraine, edge about the disease. Studies show epidemics, young people with higher for example, although 99% of girls had that many young people who are levels of education are more likely to heard of AIDS, only 9% could name 3 sexually active and know the modes use condoms and less likely to engage ways to avoid infection [2]. of transmission of the virus report that in casual sex than their less-educated Misconceptions about how HIV/ they face little or no risk of becoming peers [18]. We need to educate children AIDS is transmitted reveal that young HIV-positive themselves; for example, and young people about HIV/AIDS people are not getting access to the right according to UNICEF, as many as 87% and teach them skills that translate this information. Our study showed that of 15–19-year-olds did not believe knowledge into action. Therefore this misconceptions about modes of trans- themselves to be at risk [2]. The find- opportunity should not be missed and missions, e.g. touching, eating or drink- ings of our study showed that 28% of the schools should be used as a hub for such prevention lessons. ing, contact with clothes, kissing and young people interviewed believed that insect bites, were widely prevalent. This there was no possibility that they could Last but not least, the finding that issue was also addressed by previous get infected. A situation analysis among about half of the young people knew Yemen AIDS patients found that no research among the general population that prostitution and homosexual- patient believed that he/she could be a in Yemen [14] and neighbouring Arab ity existed in their area, albeit with a victim of HIV/AIDS [5]. Respondents’ countries [15]. According to UNICEF significant gender and geographical good awareness of links between the risk “surveys from 60 countries indicated difference, is concerning and needs of AIDS and extramarital sex—and to a that more than 50% of young people future attention. It supports previous lesser extent homosexuality—is positive anecdotal reports about these areas. Al aged 15 to 24 years had serious mis- and needs to be encouraged, especially Basateen (inhabited by refugees) and conceptions about how HIV/AIDS is as the study findings show that such Al Sesaban (inhabited by marginalized transmitted” [2]. Therefore, providing risky behaviours may be common in the populations) are characterized as high information about HIV/AIDS trans- study areas. However, as many of the risk due to the large numbers of commission that emphasizes the lack of sci- participants stressed that they did not mercial sex workers living there, and entific evidence for these beliefs should need to make any behavioural changes they represent transit hubs for migrant be a priority for any future information, themselves—as they were conservative populations coming from the Horn of education and communication cam- and not doing “bad” things—the fact Africa, where HIV/AIDS prevalence is paigns about HIV/AIDS. needs to be emphasized that avoiding higher than in Yemen [9]. In this study there were negative at- extramarital sex or homosexuality do Based on these findings, certain titudes among the participants towards not confer 100% protection from HIV/ recommendations can be made. As the AIDS patients. There was a common AIDS and that it still could be transmit- prevention of the spread of HIV infec- opinion that AIDS patients needed to ted by other means (e.g. through health tion is a priority in Yemen [8] and one be isolated (if not jailed or even killed) care settings). of its Millennium Development Goals and that, even when given medical care, This study confirms previous [4], the findings of this study should they should receive specialized care in research findings from Yemen that help policy-makers and health care special health settings. About three- television is the current and probably professionals to develop a culturally quarters of the respondents stated that the future leading source knowledge sensitive and needs-specific educational they were not willing to buy grocer- about HIV/AIDS [13]. The fact that programme for Yemeni young people. ies from a person who had AIDS and television is one of the few sources of Settings such as Yemen which have that a teacher who was infected with information that is used equally by both low HIV/AIDS prevalence must be AIDS should not continue teaching. males and females is also important prepared to act concurrently at 3 key Such attitudinal problems were found as it can be used to minimize the sig- levels of interventions—advocacy, in- among young people from other similar nificant gender gap in knowledge that formation and knowledge—in order developing countries, for example, the was underlined by this study. The fact to keep the prevalence rates low. An Islamic Republic of Iran [16] and India that schoolteachers were mentioned as enabling environment within which [17]. Such serious attitudinal problems a current as well as a future preferred these interventions can take place is es- and widespread misconceptions caused source of knowledge about HIV/AIDS sential. Advocacy to raise the awareness by lack of education about AIDS need is promising. Education is one of our of young people regarding the issues to be addressed. most important weapons against the of HIV/AIDS and reduce the stigma

249 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

associated with the disease is vital for More research on young people’s Acknowledgements achieving this. Youth-friendly services, HIV/AIDS knowledge, attitudes and especially for marginalized and refugee beliefs needs to be done in Yemen. We would like to sincerely thank the groups, that provide adequate repro- Similar investigations should be re- fieldworkers and supervisors. We are ductive health care information are es- grateful to all the young people who co- peated to provide insights about how sential. Such services have the ability to operated and participated in making this attract young people, meet their needs behaviour and knowledge are altered in study a reality. The work was supported comfortably and responsively, and suc- the study areas in relation to any future by a grant from HIV/AIDS UNICEF ceed in retaining these young clients for HIV/AIDS information and education and implemented by SOUL for the De- continuing care. campaigns. velopment of Women and Children.

References

1. HIV/AIDS and work: global estimates, impact on children and irinnews.org/report.aspx?reportid=25764 12, accessed 3 Au- youth, and response. Geneva, International Labour Organiza- gust 2009). tion, 2006 11. Multiple indicator cluster surveys—round 3. New York, United 2. Young people and HIV/AIDS: opportunity in crisis. New York, Nations Children’s Fund, 2005 (http://www.childinfo.org/ United Nations Children’s Fund, 2002 (http://www.unicef. org/publications/index_4447.html, accessed 3 August 2009). mics3.html, accessed 3 August 2009). 3. UNGASS AIDS 2006 review. Youth summit message. United Na- 12. Epidemiological fact sheets on HIV/AIDS and sexually transmitted tional Children’s Fund [online document] (http://www.unicef. infections. Yemen, Joint United Nations Programme on HIV/ org/voy/media/UNGASS_2006_Review_Youth_Msg.pdf, ac- AIDS, 2004. cessed 31 August 2009). 13. National AIDS programmes: a guide to monitoring and evalua- 4. MDG: needs assessment. Health and population. Sana’a, Yemen, tion. Geneva, Joint United Nations Programme on HIV/AIDS/ Health and Population Core Team, Ministry of Public Health and Population, 2005 World Health Organization/MEASURE Evaluation Project, 2000 (UNAIDS/00.17). 5. HIV/AIDS situation and needs assessment report. Sana’a, Yemen, United Nations Development Programme, 2001. 14. Al-Serouri AW et al. Knowledge, attitudes and beliefs about 6. Busulwa R. HIV/AIDS situation analysis study, conducted in Ho- HIV/AIDS in Sana’a, Republic of Yemen. Eastern Mediterranean deidah, Taiz, Aden and Hadhramut, Republic of Yemen. Sana’a, health journal, 2002, 8(6):706–15. Yemen, United Nations Development Programme/World 15. Al-Owish R et al. Knowledge, attitudes, beliefs and practice Health Organization/National AIDS Program, Ministry of Health and Population, 2003. about HIV/AIDS in Kuwait. AIDS education and prevention, 1999, 11(2):163–73. 7. National strategic framework for the control and prevention of HIV/AIDS in the Republic of Yemen. Sana’a, Yemen, Ministry of 16. Tavoosi A et al. Knowledge and attitude towards HIV/AIDS Public Health and Population, 2002. among Iranian students. BMC public health, 2004, 4:17. 8. Al-Rabee’ A. Adolescent and youth reproductive health in Yemen: 17. Agrawal HK et al. Knowledge of and attitude to HIV/AIDS of status, policies, programs, and issues. Washington DC, Futures senior secondary school pupils and trainee teachers in Udupi Group International, 2003. District, Karnataka, India. Annals of tropical paediatrics, 1999, 9. Al Ahmadi A, Beatty S. Participatory socioeconomic needs survey 19:143–9. of the Sana’a urban settlements dwellers with special reference to women. Yemen, Oxfam, 1997. 18. School education to prevent AIDS and sexually transmitted dis- 10. Yemen: stigma, ignorance hampering fight against HIV-AIDS. eases. Geneva, World Health Organization, 1992 (AIDS Series IRIN HIV/AIDS (PlusNews) [online press release] (http://www. No 10).

250 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

HIV/AIDS surveillance in Egypt: current status and future challenges S. Boutros1 and J. Skordis1

ُّترصد اإليدز والعدوى بفريوس العوز املناعي البرشي يف مرص: الوضع الراهن والتحديات املستقبلية سارة بطرس، جولني سكورديس 0.1 اخلالصـة:إن معدل انتشار اإليدز والعدوى بفريوس َالع َوز املناعي البرشي يف مرص )أقل من %( وهو أخفض من املعايـري العاملية. وثمة تساؤل حول دقة هذا الرقم مع خماوف بازدياد وشيك فيه، إىل جانب ِّبينات تشري إال أنه رغم الثقافة املحافظة يف مرص، فإن السلوك الذي ِّيعرض ألخطار عالية ينترش أكثر مما ّيبلغ عنه ًعادة، وأن التغ ُّيريف السياق االقتصادي واالجتامعي يف مرص دائم التغري. ويستقيص الباحثون يف هذه الورقة بعض التحديات الفريدة يف هذا البلد والتي تواجه التعاطي مع وباء اإليدز والعدوى بفريوس َالع َز واملناعي البرشي، وذلك من خالل حتليل نظام ُّالتـرصد احلايل لإليدز والعدوى بفريوس العوز املناعي البرشي. ويستنتج الباحثون أن بعض العوائق، مثل بعض املعاييـر الثقافية وبعض القوانني، وافتقار السكان للمعلومات حول اإليدز والعدوى بفريوس َالع َوز املناعي البرشي والنظام الصحي البريوقراطي، تعيق تطوير وتنفيذ نظم الرتصد َّالفعالة.

ABSTRACT By international standards, HIV/AIDS prevalence is low in Egypt (< 0.1%). However, questions about the accuracy of this figure are coupled with fears of an imminent increase in prevalence, with evidence suggesting that, despite Egypt’s conservative culture, high-risk behaviour is more widespread than commonly reported and the country’s changing socioeconomic context is perpetuating this trend. Through an analysis of the current HIV/AIDS surveillance system in Egypt, this paper explores some of the unique challenges this country faces in dealing with the HIV/AIDS epidemic. It concludes that constraints, such as Egypt’s cultural norms and laws, the population’s lack of knowledge about HIV/AIDS and the bureaucratic health system, hinder the development and implementation of effective surveillance systems.

Surveillance du VIH/sida en Égypte : situation actuelle et défis à venir

RÉSUMÉ Comparée aux normes internationales, la prévalence du VIH/sida est faible en Égypte (<0,1 %). Cependant, les interrogations sur la précision de ce chiffre sont associées à la crainte d’une augmentation imminente de la prévalence, avec des éléments qui laissent penser qu’en dépit d’une culture conservatrice en Égypte, les comportements à haut risque sont plus fréquents que communément rapporté et le contexte socio-économique qui tend à changer dans le pays accentue cette propension. Par le biais d’une analyse du système de surveillance actuel du VIH/sida en Égypte, le présent article examine certains des défis que ce pays doit relever face à l’épidémie de VIH/sida. Il conclut que les contraintes, telles les normes culturelles et la législation égyptiennes, le manque de connaissance de la population à propos du VIH/sida et le système de santé bureaucratique empêchent le développement et la mise en place des systèmes de surveillance efficaces.

1Centre for International Health and Development, University College London, London, United Kingdom (Correspondence to S. Boutros: [email protected]). Received: 17/10/07; accepted: 31/01/08

251 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Introduction collection and analysis. As explained be- place and its findings, as well as on the low, secondary data from peer-reviewed obstacles encountered for surveillance HIV/AIDS prevalence rates are low in papers and the grey literature were sup- and for tackling the epidemic in general. Egypt at less than 0.1% in the general plemented with primary data from key The diverse nature of the key informants population [1,2]. However, United Na- informant interviews. prohibited any attempt to standardize tions agencies UNAIDS and UNICEF Although the majority of available the discussions around a more formal and those working within the Egyptian resources were grey literature and web- guide. National AIDS Programme (NAP) sites, it was possible to identify through The data were later analysed using a fear a significant increase in this tra- extensive research a number of peer- modified grounded theory approach, a ditionally conservative society [3,4], reviewed papers on the topic. Databases coding/editing method that extracted where adultery is punishable by law [3] searched included: Medline, PubMed, emerging themes from the data [10]. and homosexuals are imprisoned for Eldis, JSTOR, Web of Science, MetaLib, The purpose of this analysis was to “habitual debauchery” [5]. The chang- BioMed Central, Google Scholar and a group ideas and themes together to ing socioeconomic context with the number of relevant journals, includ- allow comparison. This highlighted ongoing economic crisis [6] has in part ing the Lancet, British medical journal, commonalities as well as contradictions led to a delay in the age of marriage, AIDS and Eastern Mediterranean health between the different key informants. which in turn has contributed to an journal. Key search terms included The findings were triangulated with the increase in risky behaviour [1]. Indeed, “HIV” or “AIDS” in combination with secondary data to identify points of con- in a country where sex is assumed only “Egypt”, “HIV and Middle East” as well vergence and divergence and expand to happen between a husband and wife, as “HIV and Arab”. Articles were then the total realm of understanding. The as preached by the Muslim majority and selected by scanning through the titles data from the interviews are therefore the Coptic Christian minority [7], it is and abstracts for relevance to the topic. not quoted routinely throughout this difficult to openly discuss sexuality and Furthermore, the reference lists of the paper. Instead, the primary data are used high-risk behaviour. articles extracted were checked for any only where they either critically inform These cultural constraints, among additional relevant research papers. or contradict key issues raised by the others, are likely to challenge the way Data from key informant interviews literature review. Egypt deals with its emerging HIV were used primarily to expand on, or to The analytical framework through epidemic. More importantly, they may illustrate, discussion points raised by the which the data collected were ex- limit Egypt’s ability to collect data on the literature review described above. As amined draws on the World Health epidemic and its determinants, despite such, data from the key informant inter- Organization (WHO) guidelines for the importance of surveillance systems views were used only where they added second-generation HIV surveillance in guiding monitoring and evaluating to the general analysis and improved the [11]. Through integration of biological HIV-related policy and interventions, as completeness of the research, consist- and behavioural data, this surveillance well as in strengthening commitment, ent with the method of triangulation system aims “to concentrate resources mobilizing communities and lobbying [9]. The informants interviewed were where they will yield information that for resource allocation [8]. selected from the government as well is most useful in reducing the spread Through an analysis of Egypt’s sur- as from key nongovernmental organi- of HIV and in providing care of those veillance systems, this paper aims to zations (NGOs) working in the field affected”. It is therefore also tailored to give an insight into some of the unique including UNAIDS and UNICEF. accommodate the different stages of challenges faced by this country in deal- In total, 5 interviews were conducted. a country’s epidemic. Egypt has been ing with HIV/AIDS. Each interview lasted approximately 1 consistently recognized as a low-level hour and all were directly transcribed epidemic country [2,3]. However, the both during and immediately after the lack of prevalence data raises the pos- Methods interviews. The interviews were all con- sibility that a concentrated epidemic ducted in Arabic (the mother-tongue may already exist in high-risk groups Data collection and analytical of the first author). They were largely [4]. Some evidence for this hypothesis framework unstructured; broad introductory, open comes from 2 recent studies, demon- To overcome the limitations created questions were used to initiate the dis- strating a 6.2% and 8% infection rate in by the lack of published data on HIV/ cussion. These included questions on men who have sex with men (MSM) AIDS in Egypt and improve confidence the current state of the HIV epidemic and intravenous drug users (IVDU) in the findings, this paper made use in Egypt, on its likely future trajectory, respectively [4,12]. As such, Egypt’s of the triangulation method of data on the kind of surveillance already in surveillance system will be analysed in

252 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

accordance with WHO guidelines for Findings clients, the wives of MSM (73% of concentrated epidemics. MSM are married in Egypt) [4] and Biological surveillance the sexual partners of IVDU [15] as Analysis of the second-generation likely groups. Biological surveillance surveillance systems was supplemented Biological surveillance is essential for monitoring the level of HIV infection in Egypt should thus focus firstly on by an overview of appropriate so- over time and analysing trends to MSM, CSW and IVDU and secondly cioeconomic data, as these are proven predict possible future trajectories of on CSW clients and the sexual partners of MSM and IVDU. indicators of HIV risk [8]. Age, sex, the epidemic [13]. For concentrated educational and marital status as well as epidemics, the WHO recommends car- HIV serosurveillance rying out HIV serosurveillance in at-risk migration and gender issues are crucial In 2007, the NGO Family Health In- to our understanding of the country’s sub-populations, bridging populations ternational (FHI) reported the results epidemic [11]. At present, however, and the general population, as well as of its first round of biological and be- screening of donated blood, HIV/AIDS very few socioeconomic data are col- havioural surveillance of at-risk groups case reporting, surveillance of sexually lected under the second-generation [15]. Despite small sample sizes and a transmitted infections (STIs) and other low number of participating women, surveillance system. This paper argues biological markers of risk [11]. But first, the findings were consistent with other that inclusion of these data will be a the relevant populations must be identi- isolated studies [16,17], i.e. a low preva- useful complement to biological and fied and accessed [11]. lence of HIV infection (< 1%) among all behavioural surveillance. In Egypt, commercial sex workers at-risk groups, except MSM, who had an Figure 1 illustrates the variety of (CSW), MSM and IVDU constitute infection rate of 6.2% [15]. This suggests data that will be analysed to assess the at-risk populations [14]. Vulnerable a concentrated epidemic among MSM, groups include women, youth, prison- although it remains to be confirmed by epidemic, the appropriateness of the ers, street children, uniformed services consistent and repeated surveillance. current surveillance system and the and refugees [14]. Bridging populations Among bridging populations and challenges faced by Egypt in dealing have not been officially identified, al- the general population, little serosur- with the HIV epidemic. though isolated reports suggest CSW veillance is being conducted. HIV

Socioeconomic data • Economic • Sociocultural • Gender

Analysis of HIV/AIDS epidemic Analysis of the suveilance system Analysis of constraints and challenges

Biological surveillance Behavioural surveillance • HIV serosurveillance • At-risk groups • Case reporting • Bridging populations • Blood banks • General population • STIs, hepatitis B & C prevalence

Figure 1 Framework of analysis

253 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

screening is taking place in a number of HIV/AIDS case reporting products [3] and 5 outbreaks have oc- sentinel surveillance sites including 12 Case detection takes place mostly in curred in renal dialysis units due to the tuberculosis clinics and chest hospitals, government central testing laboratories, reuse of syringes and dialysis equip- 3 sexually transmitted disease clinics sentinel surveillance sites, private labo- ment [25,26]. This number is much and 5 antenatal clinics as well as among ratories [3] and newly opened, anony- higher than the 5% average worldwide tourism workers and sailors presenting mous voluntary counselling and testing for such HIV infections [3]. Increased for work permits [3]. Although these centres [20]. Positive cases are reported media attention prompted the Ministry surveillance sites do not target bridg- directly to the government’s NAP, al- of Health and Population (MOHP) to launch an infection control programme ing populations and only marginally though this is not the case for private in 2003 [27]. However, remaining chal- represent the general population, they laboratories [3] whose cooperation in lenges include a lack of equipment and constitute a positive step towards ex- HIV case reporting remains unreliable according to one key informant. Cur- training, the lack of a coordinating body panding Egypt’s surveillance efforts. within the MOHP, and the highly bu- rent case reports, although far lower than Appropriate systematic screening of reaucratic and under-resourced nature UNAIDS estimates (5 000–13 000) at-risk groups may be frustrated by the of the Egyptian health care system in [2], do point to an emerging epidemic social stigma attached to these groups general [1]. with an increase in the number of newly and the illegality of their behaviours reported cases [4]. [3]. Indeed, the General Penalties STI surveillance and other markers of HIV risk Laws in Egypt, as in most countries, Although this may be due to an increase in the number of people being criminalize CSW and IVDU [3]. MSM In the absence of comprehensive HIV tested and to improved surveillance, are similarly criminalized through surveillance, STI surveillance is con- there is a general belief, as revealed in laws regulating “inappropriate” social sidered a useful proxy for risk factor all the interviews, that HIV prevalence prevalence [11]. Furthermore, ulcera- conduct and/or “insult to Islam” [18]. is highly underestimated and is on the tive STI infection increases the risk of Indeed, MSM have been arrested and increase. This may be due to weak re- HIV infection and WHO now consid- imprisoned on a number of occasions, porting, under-utilization of voluntary ers STI patients a high-risk group for despite condemnation by the NGO counselling and testing centres, a wide- HIV infection [11]. Hepatitis B and C Amnesty International [5]. These laws spread reluctance to test due to fear and also serve as proxies as they have similar combine with the highly conservative stigma as discussed above [4], as well as modes of transmission [11]. Egyptian culture to further marginalize a general lack of knowledge about the In Egypt, national STI surveillance these groups, making them difficult to disease among health workers and the is virtually nonexistent although some access [18]. Additionally, HIV-infected general population [21–24]. One key studies were identified and are detailed people are usually seen as immoral and informant explained that most testing in Table 1 [28–31]. These results dem- promiscuous [19], making surveillance is currently “passive” (i.e. not related onstrate that STI prevalence is signifi- more difficult to achieve. Added to this to perceived risk) with the majority of cantly higher than current estimates of are rumours about HIV-positive people those getting tested being young males HIV prevalence. Similarly, there is no being imprisoned, as reported by one seeking “virus-free” certificates for work comprehensive surveillance for hepati- key informant, which further prevents permits abroad [3]. Therefore, current tis B and C, although more research is people from using testing services. case reporting, although indicative of an being conducted on their epidemiology emerging epidemic, is not yet represent- [3], possibly due to the longer epidemic NGOs such as FHI may have bet- ative of the true state of the epidemic. history of hepatitis in Egypt [32] as well ter access to at-risk groups because as the lack of stigma associated with it. they are not directly under government Blood-bank surveillance As for STIs, prevalence data for hepa- authority, but access is particularly dif- By law, all donated blood must be titis B and C point to the presence of ficult for the NAP with its government screened for HIV and cases must be high-risk behaviours that could similarly attachment 3[ ]. As such, despite recent reported to the NAP along with the spread HIV among the general popula- improvements in Egypt’s HIV surveil- donor’s identification information [3]. tion [33,34]. lance effort, many continue to voice However, there have been a number It is unclear whether these data have their concern that the little surveillance of reported cases of infections due to been analysed within the second-gener- that is taking place is rarely representa- blood transfusions in Egypt. In fact, 24% ation surveillance system. Nevertheless, tive and is leading to a systematic under- of known HIV cases are attributed to this clearly needs to be incorporated into estimation of the epidemic [3]. the use of infected blood and/or blood HIV data projections and strengthened

254 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

as a step towards improving overall sur- [3]. Despite the wide availability of veillance. condoms in Egypt [37], reasons cited for non-use included belief that Behavioural surveillance 95 1.5 n/a 14.7 36.8 it would lower sexual satisfaction, 30.9 Behavioural surveillance is increasingly problems with condom use and a

recognized as an essential component lack of knowledge about condoms Patients attending of any HIV/AIDS surveillance system and their existence [3]. Another po- clinic, Cairo main STI [35]. In low-level epidemics it “can be tential barrier to condom use may used to identify who is at-risk of infec- be the popular belief that condom 0 8.6 0.8 n/a 509 18.3 tion and which behaviours commonly buyers are automatically assumed Giza put them at-risk”, while in concentrated

to be homosexual, a highly stigma- women, Rural epidemics surveillance should explore tized notion in this society [38]. the links between the at-risk and general Among CSW, one study of populations [11]. non-institutionalized female CSW Despite a lack of research into sex- described high numbers of partners 0 2.8 2.8 2.8 n/a ual and drug-taking behaviour in Egypt and low levels of condom use [28]; 108

[3], the behaviours of at-risk groups, 14% exhibited scars of intravenous clinic, Cairo especially IVDU, have received some

injection (suggesting drug use) and FP attending Women isolated study and were included in the signs of physical abuse suggesting FHI surveillance initiative mentioned exposure to multiple risks [28]. earlier [3,15]. Between 6% and 13% Studies on institutionalized female of the general population use recrea- 0 1.3 0.7 2.0 n/a

CSWs also revealed exposure to 607 tional drugs and about 16% to 41% of HIV risk [3]. Despite the known these use injected drugs [36]. About ANC, Cairo

existence of male CSW, no behav- attending Women half of IVDU have ever shared or re- ioural studies have been carried out used syringes [3]. Despite the wide in this group [3]. availability and relatively low price of No behavioural surveillance is 1.3 2.7 2.7 injecting equipment at pharmacies, 29% 0.7 150 n/a of IVDU in one study were unable to currently taking place in the general Cairo Drug users, obtain a syringe from pharmacies when population, although some data ex- needed. Reasons stated include being ist for vulnerable groups, particularly refused purchase at pharmacies and the Egypt’s youth [14]. Data from vari- 1.3 7.5 80 8.8 8.8 pharmacy being closed at the desired ous studies show an early average age n/a of sexual initiation, multiple sexual

time [3]. As for sexual relations, of the MSM, Cairo 74% who were sexually active in one partners, premarital and extramari- study, 15% reported having had more tal sex (including commercial sex), than 3 sexual partners in the previous injecting drug use and low levels 52 7.7 7.7 5.8 n/a month. In addition, of those who were of condom use [39]. Additionally, 19.2 sexually active, 50% had previously en- orfi marriages, where young people Cairo CSW, gaged in commercial sex as a provider obtain a clandestine, temporary, or client and 58% of respondents had marriage certificate allowing them never used a condom [3]. to engage in sexual relations, are 0 0.3 n/a n/a n/a For MSM, among whom 26% of all on the increase [7]. Although only 740

HIV cases occur in Egypt [14], only 2 11% of HIV cases are currently re- south Sinai studies could be identified, both con- ported among Egypt’s youth, this workers, Tourism ducted by the NAP [3]. They revealed a group may become a large pool of low rate of condom use, multiple sexual HIV-infected persons if no action is partners and receptive and total pen- taken, as they represent 20% of the etrative sexual acts. Regular condom population [3] and are still highly use was non-existent among a sample unaware of the disease [39–41]. Syphilis HIV Chlamydia Gonorrhoea of 80 MSM and was reported as 19% in Indeed “misinformation and lack of Trichomoniasis STI rate (%) rate STI Sample size (no.) Variable CSW = commercial sex workers; MSM = men who have withCSW = commercial FP = family planning; n/a not tested. men; ANC = antenatal care; another sample of 73 homeless MSM information simply increase sexual infections population of transmission 28–31 ] ) among different sexually [sources: (STIs in Egypt 1 Prevalence different groups Table

255 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

confusion and vulnerability” among If these data are not integrated in the of immoral Western society” [47,48], it young people [40]. epidemiological analysis, it may leave is clear from the above analysis that this The data above demonstrate that, policy-makers with the illusion that culture represents more of a challenge apart from a number of isolated surveil- societal and economic contexts cannot to, than a protection from, risk at this lance studies and the FHI surveillance impact the course of the epidemic [13]. stage of the epidemic. initiative, there is as yet no systemic This section will therefore give a wide Egypt, like many countries in the surveillance currently in place among overview of Egypt’s socioeconomic Eastern Mediterranean Region, is at-risk populations. Additionally, behav- context and will demonstrate links with additionally characterized by gender iours that may link the general popula- existing HIV risk behaviours. This analy- inequalities [42]. For example, stud- tion to at-risk groups have not yet been sis will additionally serve to highlight a ies show that most women are unable explicitly studied. Possible constraints number of unique challenges that Egypt to request that their husbands wear a to achieving adequate surveillance will have to face to deal appropriately condom [49]. As a result of their lim- may include the difficulty in accessing with its emerging epidemic. ited access to information, women’s at-risk groups as they are illegal and Economically, Egypt has been in knowledge of HIV is also weak [50]. stigmatized, as discussed earlier. Ad- crisis for a number of years with devalu- Their lack of financial autonomy and ditionally, the conservative nature of ation of the Egyptian pound by more overall unawareness of available re- Egyptian society may lead to reluctance than 40% between 2000 and 2001 [6]. sources constitute significant barriers to discuss sexuality and behaviours and This has led to a sharp deterioration of to accessing health care [1]. However, therefore limits the ability to conduct living standards [43]. Men have “suf- as illustrated by data from the NAP, the representative behavioural surveillance. fered” particularly from this crisis as the female to male ratio of HIV infection Indeed, perhaps the strongest barrier to financial security of the male is a prereq- stands at 1:4 [16]. Although this may effective behavioural surveillance is the uisite for marriage in Egyptian society be explained by the much higher rate “strong reluctance to discuss sexuality” [44]. Although some suggestion of an of men testing, this ratio may also result [42]. However, in the context of Egypt, improvement has begun to emerge, from the particularly strong cultural it should also be mentioned that if be- the prevailing situation has resulted in prohibitions against extramarital sex havioural surveillance focused solely on a general delay in the age of marriage applied to women. There are few data at-risk groups this could have negative [1]. In a country where sex is only “al- on the socioeconomic reasons underly- impacts on the public and the media as lowed” in the context of marriage, this ing the lower prevalence of HIV among it would lead to further stigmatization of delay has increased frustration among Egyptian women. Without these data, those groups as well as a decreased risk young people. This may explain many changes in women’s positions in society perception among the wider population, behavioural changes, particularly the and the resultant changes in behaviour who do not engage in those risk behav- increase in premarital sex and orfi mar- cannot be appropriately monitored. iours [15]. It is therefore crucial in this riage discussed earlier. The above data have provided some context that behavioural surveillance Socially, conservative norms and insight into the complex constraints also covers the general population. This laws prevent sexual education, as it is Egypt faces in dealing with its HIV is not yet systematically conducted. still believed that the theory may en- epidemic. It is therefore paramount that courage the practice of high-risk be- Socioeconomic surveillance such data be collected and analysed as haviour [45]. However, Egypt may be an integral part of surveillance systems Socioeconomic surveillance indicates overestimating the protective effect of to both understand and limit the con- the evolution of risk factors [11]. WHO its social and cultural conservatism and tinuing spread of the epidemic. advises that basic socioeconomic data may therefore be giving low priority to such as age and sex be collected at all HIV/AIDS education [46]. Indeed, sentinel sites [11]. However, this paper political support has been hard to ob- Discussion argues that more in-depth and wider tain, especially when dealing with at-risk socioeconomic information is crucial groups, as reported by one key inform- Egypt’s HIV epidemic has remained to understanding and dealing with an ant. Admitting the presence of high-risk low by international standards. How- emerging epidemic. Prevalence rates are behaviour means national leaders have ever, as discussed here, the country risks largely determined by behaviour that is to acknowledge that their citizens do developing a much larger epidemic. directly and indirectly affected by socio- not always conform to cultural and re- Consistently low case detection and the economic factors [13]. High unemploy- ligious ideals. Although some may still relative lack of behavioural surveillance ment, migration and illiteracy increase a believe that these conservative norms may provide a false sense of invulner- country’s vulnerability to HIV [11]. have protected Egypt from this “plague ability [46]. This is reinforced by an

256 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

over-confidence in the adherence to integrate socioeconomic data into the appropriate surveillance can ensure a conservative social cultural norms that surveillance system. timely and appropriate response to this prohibit sex outside marriage [46]. This Four key challenges explain many of emerging epidemic. could result in HIV remaining a low these gaps in Egypt’s surveillance system. government priority. First, at-risk groups are difficult to ac- That said, Egypt has a window of cess due to their social marginalization. opportunity to stop HIV becoming a MSM are arguably the most affected by Acknowledgements generalized epidemic. However, this this problem. Secondly, a severe lack of can only be achieved if appropriate and knowledge among health workers and We would like to thank all the key timely surveillance systems are put in the general population makes it less informants from UNICEF Cairo, UN- place. As illustrated above, a significant likely that people will seek a potential number of gaps were found in the cur- HIV diagnosis. Thirdly, the conservative AIDS Cairo and the Egyptian National rent surveillance system, despite efforts nature of Egyptian society, including a AIDS Programme for their time and general reluctance to discuss sexuality, by the NAP and other national and dedication. We would also like to thank international organizations. Gaps in- limits the ability to conduct representa- clude a lack of biological surveillance tive behavioural surveillance. Finally, Nabil and Sue Raphael, as well as Nadia among at-risk groups, weak surveillance structural factors within the NAP itself Taher for their invaluable comments among women, failure to systematically and the wider health system, ranging and suggestions for the final draft of this and universally apply the regulations on from understaffing, bureaucracy, limited infection control and blood-bank safety, budget allocation and fragmentation paper. Finally, we would like to thank a lack of comprehensive data on STIs, of efforts, may be constraining surveil- the reviewers for their very useful input. a lack of systematic and representative lance efforts. Further research and behavioural surveillance (especially for practical strategies are urgently needed Any remaining omissions or errors are, the general population), and a failure to to overcome these challenges so that of course, our own.

References

1. Abdel-Rasik M. Analysis of public policy on HIV/AIDS. Cairo, 10. Miller WL, Crabtree BF. Doing qualitative research, 2nd ed. Ministry of Health and Population and UNICEF, 2005. Thousand Oaks, California, Sage, 1999. 2. Country and regional responses to AIDS: Egypt. Cairo, Joint 11. UNAIDS/WHO Working Group on Global HIV/AIDS and United Nations Programme on HIV/AIDS, 2007 (http://www. STI Surveillance. Guidelines for second generation HIV surveil- unaids.org/en/CountryResponses/Countries/egypt.asp, ac- lance. Geneva, Joint United Nations Programme on HIV/AIDS, cessed 11 August 2009). 2000. 3. El-Sayed N, Aon M, Mortagy I. Assessment of the HIV/AIDS situ- 12. HIV/AIDS Profile: Egypt. Washington DC, United States Bureau ation and response in Egypt. Cairo, National AIDS Programme, of the Census, Population Division, 2000. Ministry of Health and Population and the Expanded Theme Group on HIV/AIDS, 2004. 13. Diaz T et al. New strategies for HIV surveillance in resource- constrained settings: an overview. AIDS, 2005, 19 (Suppl. 4. Aon M. HIV situation and response in Egypt. Cairo, Joint United 2):1–8. Nations Programme on HIV/AIDS, 2007. 14. Global Fund proposal form: sixth call for proposals. Cairo, Egyp- 5. Egypt: private internet conversation used as evidence. The tian National Country Coordinating Mechanism to Fight HIV/ Wire, 2003, April, Vol. 33, No.3. Amnesty International AIDS, TB and Malaria, 2006. [online document] (http://www.amnesty.org/en/library/ asset/NWS21/003/2003/en/780c1825-d710-11dd-b0cc- 15. HIV/AIDS biological behavioural surveillance survey summary 1f0860013475/nws210032003en.html, accessed 11 August report. Cairo, Ministry of Health and Population, Family Health 2009) International and Joint United Nations Programme on HIV/ 6. Raphaeli N. Economic crisis in Egypt. Middle East Media Re- AIDS, 2006. search Institute, 2002 (Economic Studies No. 20) [online arti- 16. AIDS/HIV surveillance report for the period 1/4/2006 to cle] (http://memri.org/bin/articles.cgi?Page=subjects&Area= 30/6/2006. Cairo, National AIDS Program, 2006. economic&ID=EA2002, accessed 11 August 2009). 17. Watts DM et al. Prevalence of HIV infection and AIDS in Egypt 7. DeJong J, El-Khoury G. Reproductive health of Arab young over four years of surveillance (1986–1990). Journal of tropical people. British medical journal, 2006, 333:849–51. medicine and hygiene, 1993, 96:113–7. 8. Rehle T et al. Second-generation surveillance: better data for 18. Srikameswaran A. Egypt’s fearful gays shy from HIV testing. Inter- decision-making. Bulletin of the World Health Organization, national reporting project, Paul H. Nitze School of Advanced 2004, 82(Suppl. 2):121–7. International Studies of the Johns Hopkins University [online 9. Adami MF, Kiger A. The use of triangulation for completeness article] (http://www.internationalreportingproject.org/sto- purposes. Nurse researcher, 2005, 12(4):19–29. ries/detail/575/, accessed 11 August 2009).

257 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

19. Shahine G. Facing up to AIDS. Al-Ahram weekly, 2004, 10 34. Quinti I et al. Seroprevalence of HIV and HCV infections in Al- March 2004, 679 (Issue No. 4) (http://weekly.ahram.org.eg/ exandria, Egypt. Zentralblatt bacteriology, 1995, 283(2):239–44. print/2004/680/eg2.htm, accessed 11 August 2009). 35. Brown T. Behavioural surveillance: current perspectives, and 20. Ingram S. New HIV/AIDS counselling and testing centres open in its role in catalyzing action. Journal of acquired immunodefi- Egypt. United Nations Children’s Fund [online article], 2005 ciency syndrome, 2003, 32(Suppl. 1):S12–7. (http://www.unicef.org/aids/egypt_28921.html, accessed 11 36. UNOCD. Rapid assessment of trends and patterns of drug abuse August 2009). in Egypt. Cairo, United Nations Office on Drugs and Crime and 21. Megeid AA et al. Knowledge and attitudes about reproductive Ministry of Health, 2001. health and HIV/AIDS among family planning clients. Eastern 37. Kandela P. Arab nations: attitudes to AIDS. Lancet, 1993, Mediterranean health journal, 1996, 2(Suppl. 3):459–69. 341:884–5. 22. Faris R, Shouman A. Study of the knowledge and attitude of 38. Srikameswaran A. AIDS fight hurt by Egyptian men’s shunning Egyptian health care workers towards occupational HIV infection. Journal of the Egyptian Public Health Association, 1994, of condoms. International reporting project, Paul H. Nitze 69(Suppl. 1/2):115–28. School of Advanced International Studies of the Johns Hop- kins University [online article] (http://www.internationalre- 23. Sallam SA et al. Continuing medical education needs regarding portingproject.org/stories/detail/577/, accessed 11 August AIDS among Egyptian physicians in Alexandria, Egypt and in 2009). the Asir region, Saudi Arabia. AIDS care, 1995, 7(Suppl. 1):49. 39. Aon M. Young people and HIV/AIDS. A qualitative study in Cairo 24. Sallam SA et al. HIV/AIDS knowledge, perception, and risk and Qena. Cairo, Joint United Nations Programme on HIV/ behaviour among patients suspected to have STDs in Alexan- AIDS, 2004. dria, Egypt. American journal of epidemiology, 2001, 153(Suppl. 11):190. 40. Farghaly AG, Kamel MM. The study of the opinion and level of knowledge about AIDS problem among secondary school stu- 25. Hassan NF et al. HIV infection in renal dialysis patients in Egypt. AIDS, 1994, 8(Suppl. 6):853. dents and teachers in Alexandria. Journal of the Egyptian Public Health Association, 1991, 66(Suppl. 1/2):209–25. 26. El Sayed NM et al. Epidemic transmission of HIV in renal dialysis centres in Egypt. Journal of infectious diseases, 2000, 181:91–7. 41. Ginedy M, El-Sayed N, Darwish A. Knowledge and attitudes of teenage students in relation to sexual issues. Eastern Mediter- 27. Talaat M et al. Evolution of infection control in Egypt: achieve- ranean health journal, 1998, 4(Suppl. 1):76–80. ments and challenges. American journal of infection control, 2006, 34(Suppl. 4):193–200. 42. Obermeyer CM. HIV in the Middle East. British medical journal, 2006, 333:851–4. 28. AbdelSattar A et al. Evaluation of selected reproductive health infections in various Egyptian population groups in Greater Cairo. 43. Country report: Egypt. Egypt at a glance 2007–08. London, Cairo, Ministry of Health and Population, IMPACT/Family Economist Intelligence Unit, 2006. Health International and Joint United Nations Programme on 44. John A. The socio-cultural context of population programs. HIV/AIDS, 2002. Bochum, Germany, Institute of Development Research and 29. Ali F et al. Prevalence of certain sexually transmitted diseases Development Policy, 1997 (http://www.ruhr-uni-bochum.de/ in Egypt. Journal of the Egyptian Public Health Association, 1996, iee/publ/mat/pdf/mat_161.pdf, accessed 11 August 2009). 71(Suppl. 5/6):553–75. 45. Luyendenijk J. “Egypt doesn’t have AIDS, so it doesn’t need 30. El-Sayed NM et al. Seroprevalence survey of Egyptian tourism AIDS education”. Sexual health exchange, 2001, 2:10–1. workers for HBV, HCV, HIV and Treponema pallidum infec- 46. Jenkins C. Overview of the HIV/AIDS situation in the MENA re- tions: association of HCV infections with specific regions of gion. Washington DC, World Bank, 2001. Egypt. American journal of tropical medicine and hygiene, 1996, 55(Suppl. 2):179–84. 47. Lenton C. Will Egypt escape the AIDS epidemic? Lancet, 1997, 349:1005. 31. Khattab H, Younis N, Zurrayk. Women, reproduction and health in rural Egypt. The Giza study. Cairo, American University in 48. Soliman S. Egypt’s fundamentalists say condoms immoral. AIDS Cairo Press, 1990. analysis Africa, 1995, 5(Suppl. 3):14–5. 32. Mohamed MK et al. Viral hepatitis C infection among Egyp- 49. Afifi AA. Factors influencing participation for prevention of tians, the magnitude of the problem: epidemiological and HIV/STDs in an urban community in Cairo-Egypt. In: Proceed- laboratory approach. Journal of the Egyptian Public Health As- ings of the International Conference on AIDS, 7–12 July, 2002, sociation, 1996, 71(1/2):79–111. 14 (abstract number F12160) (http://gateway.nlm.nih.gov/ 33. El Ghazzawi E et al. Intravenous drug addiction: a high risk MeetingAbstracts/ma?f=102254008.html, accessed 11 August group for infection with human immunodeficiency virus, 2009). hepatitis viruses, cytomegalovirus and bacterial infections in 50. Sallam SA et al. Reproductive tract infections among married Alexandria, Egypt. Journal of the Egyptian Public Health Associa- women in Upper Egypt. Eastern Mediterranean health journal, tion, 1995, 70(1/2):127–50. 2001, 7(Suppl. 1/2):139–46.

258 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Evidence-based approach to HIV/AIDS policy and research prioritization in the Islamic Republic of Iran A. Feizzadeh,1,2 S. Nedjat,2,3 S. Asghari,2 A. Keshtkar,4 R. Heshmat,2 H. Setayesh1 and R. Majdzadeh2,3

أسلوب ُم ْس َند ِّبالبينات لوضع أولويات للبحوث والسياسات حول اإليدز والعدوى بفريوس َالع َوز املناعي البرشي يف مجهورية إيران اإلسالمية عيل فيض زاده، سحرناز نجات، شبنم أصغري، عباس كشتكار، رامني حشمت، محيد رضا ستايش، رضا جمد زاده

اخلالصـة: ُّمتس احلاجة إىل أسلوب ُم ْس َند ِّبالبينات أثناء صياغة اخلطة االستـراتيجية الوطنية الثانية للوقاية من اإليدز والعدوى بفريوس َالع َوز املناعي البرشي يف مجهورية إيران اإلسالمية. وتصف هذه الورقة مراجعة منهجية ِّللبينات املحلية حول حمددات رساية اإليدز والعدوى بفريوس 3 َالع َوز املناعي البرشي وذلك ضمن فئات هي: معارف ضئيلة ومواقف سلبية عن رساية فريوس العوز املناعي البرشي، وتعاطي املخدرات باحلقن، 53 93 والتساهل يف املامرسات اجلنسية. ومن بني ًتقريرا قام الباحثون بمراجعتها كان هناك ًتقريرا ّي تلـبمعايـري اإلدراج بالدراسة. وكانت هناك ندرة يف املعلومات حول معدل انتشار ومدى التأثريات للمحددات الثالثة للخطر عىل الصعيد الوطني واإلقليمي. وقد كان لعدم التجانس بني الدراسات حتى ضمن املجموعات السكانية الفرعية أمهية ُي ْع َت ُّ دهبا ًإحصائيا. واستنتج الباحثون أن احلاجة ّ ةماس لقاعدة حمسنة للبحوث ولتبادل أفضل للمعلومات بني بلدان إقليم رشق املتوسط.

ABSTRACT In formulating the second national strategic plan for prevention of HIV/AIDS in the Islamic Republic of Iran a more evidence-based approach was needed. This paper reports on a systematic review of the local evidence about the determinants of HIV/AIDS transmission in 3 categories: poor knowledge and negative attitudes about HIV transmission; injection drug use; and sexual promiscuity. Of 93 reports reviewed, 53 met the inclusion criteria. Information about the prevalence and magnitude of effect for the 3 risk determinants at the national and regional level was scarce. Heterogeneity between studies, even in the same sub-population, was significant. An improved research base and better sharing of information are needed within countries of the Eastern Mediterranean Region.

Approche fondée sur des bases factuelles de la politique relative au VIH/sida et établissement des priorités de la recherche en République islamique d’Iran

RÉSUMÉ Lors de l’élaboration du second plan national stratégique pour la prévention du VIH/sida en République islamique d’Iran, une approche davantage fondée sur des données factuelles s’est avérée nécessaire. Cet article présente une étude systématique des données locales sur les déterminants de la transmission du VIH/sida dans trois domaines : méconnaissance et attitudes négatives vis-à-vis de la transmission du VIH, consommation de drogues par voie intraveineuse et promiscuité sexuelle. Sur 93 rapports étudiés, 53 d’entre eux satisfont les critères d’inclusion. Les informations relatives à la prévalence et à l’amplitude de l’effet des trois déterminants du risque aux niveaux national et régional étaient peu nombreuses. L’hétérogénéité des études, y compris dans la même sous-population, était importante. Une meilleure base de recherche et un meilleur partage des informations sont nécessaires dans les pays de la Région de la Méditerranée orientale.

1Joint United Nations Programme on HIV/AIDS, Tehran, Islamic Republic of Iran. 2School of Public Health; 3Knowledge Utilization Research Centre, Tehran University of Medical Sciences, Tehran, Islamic Republic of Iran (Correspondence to R. Majdzadeh: [email protected] or S. Nedjat: [email protected]). 4School of Medicine, Golestan University of Medical Sciences, Golestan, Islamic Republic of Iran. Received: 27/11/07; accepted: 04/03/08

259 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Introduction to heterosexual transmission. We there- Iranian nongovernmental organiza- fore chose to investigate the following tions (NGOs), research centres and the The Joint United Nations Programme risk factors for HIV transmission: poor media as well as international NGOs on HIV/AIDS (UNAIDS) estimates knowledge and negative attitudes about and UN agencies, including the World the number of people living with HIV transmission of HIV; IDU; and sexual Health Organization. in the Islamic Republic of Iran at 66 000. promiscuity. At the time, there was no The keywords used for electronic Thus, while HIV prevalence among published review from the Islamic Re- and manual searches were as fol- women aged 15–49 years is well below public of Iran about the prevalence and lows: knowledge and attitude (health the 1% threshold which defines a gen- magnitude of effect of any of these fac- education, health knowledge, attitudes, eralized epidemic, prevalence estimates tors. The available evidence would then practice, perception) towards high-risk of more than 5% among injection drug be used to determine policy priorities behaviours including sexual behaviour users (IDUs) mean that the country is for the second national strategic plan, (unsafe sex, homosexuality and hetero- facing a concentrated HIV epidemic while the gap between the evidence sexuality, risk behaviour, prostitution, [1]. needed and that available would help to needle sharing, intravenous substance The Iranian National AIDS determine research priorities. and drug abuse, sexually transmitted Committee harmonizes the national infections), in combination with the response to HIV/AIDS through a keywords HIV and AIDS. strategic framework. The first national Methods The inclusion criteria for informa- strategic plan (years 2002–06) was tion were: no earlier than 1994; a defined mainly based on consensus among We used a systematic review approach population (in prevalence studies) and policy-makers and academics and was to find the prevalence and magnitude a well-defined exposure and outcome not fully evidence-based. The plan was of effect of each of the 3 risk determi- (in analytic studies). nants. based on 11 main strategies, including For each risk factor measure we used education and information, IDU, harm Systematic review a data extraction form with items includ- reduction and prevention of mother- ing: study population, date and location All domestic and foreign electronic and to-child transmission, the latter related of study, estimated measure and study nonelectronic databases which were ac- more to case-finding and treatment. quality indicators. The quality of the cessible and websites of key centres and A good strategic plan for prevention reports was assessed by considering the institutes were searched. We consulted of HIV/AIDS should cover all at-risk study population and design, examining conference abstracts, theses, journals and high-risk populations and allocate biases and confounding factors. (Iranian and foreign), the references of resources based on current and pro- Data from the forms were entered jected priorities. These broadly-defined articles and the “grey literature”. Experts were questioned about the relevant into 3 different evidence tables for the 3 populations should be translated into search themes. specific groups in each society based literature too. on the local context. It was therefore The following databases were Data analysis decided during the mid-term review searched: IranMedex (http://www. Based on the possibility of different sub- of the first strategic plan that a more iranmedex.com, a private company da- populations, we conducted sub-group evidence-based approach was needed tabank of articles published in Iranian analysis to reduce the heterogeneity in formulating the second strategic plan biomedical journals); Irandoc (http:// within each sub-group. After excluding (years 2007–10). For any risk factor for www.irandoc.ac.ir, a database of post- the remaining qualitative heterogene- a disease, knowing its factual prevalence, graduate theses and dissertations); ity, quantitative heterogeneity graphi- its counterfactual prevalence and the Scientific Information Database (http:// cal and statistical methods, i.e. forest magnitude of its effect can help estimate www.sid.ir; a database of articles pub- and radial plots and chi-squared [3] the avoidable burden of that disease lished in Iranian scientific research jour- statistical tests were utilized. All analy- [2]. nals); Cochrane Library; Medline; and ses were performed using Stata SE, The current research was a system- Embase. The following Iranian public version 8.0. atic review of the evidence under 3 institutions were consulted: Ministry main search themes. At the time of this of Health and Medical Education, study in 2005 the debate about HIV in State Welfare Organization, Minis- Results the Islamic Republic of Iran hinged on try of Culture and Islamic Guidance, whether the epidemic was shifting from Centre for Disease Control and Pris- The findings are reported and tabulated predominately IDU-based transmission ons Organization. We also consulted under the 3 search themes.

260 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Poor knowledge and the public sector and were mainly based differences, as well as epidemiological negative attitudes about HIV on unpublished registry data. differences 5[ ]. Another finding was the transmission As there was significant qualitative considerable heterogeneity between There were 35 reports about poor heterogeneity among reports, even studies, even in the same sub-population, knowledge and negative attitudes to within each stratum, we did not use which made it impossible to estimate HIV; 26 of them met the inclusion meta-analysis methods. There was no the prevalence of risk determinants. criteria and were entered in the study report on effect size for IDU as a risk The shortage of information needed (Table 1). The reports were divided into for HIV infection. However, analysis of to estimate the magnitudes of effects 6 groups based on the study population the crude results of one study yielded an and prevalence made it impossible to (health care workers, school and col- odds ratio of 3 for having HIV infection make even an educated guess about the lege students, runaway girls and female among IDUs compared with noninjec- avoidable burden of any of the 3 risk commercial sex workers, barbers and tion drug users [4]. determinants. The heterogeneity was hairdressers, and general population); statistically significant, as evident from the summary measure was calculated Sexual promiscuity significant chi-squared tests. Therefore, separately for each group. Because of the cultural taboos sur- there appears to be a qualitative source for the heterogeneity, which cannot be Although we aimed to estimate the rounding extramarital sex in any form explained by simple random variation summary for each subgroup, the vari- in the Islamic Republic of Iran, reports on sexual promiscuity and commercial in findings that could be treated by ran- ety of responses did not allow this. For dom-effects meta-analysis techniques. example poor knowledge varied from sex work were scarce and mostly meth- The observed heterogeneity may 1.6% to 90% of the population. Hence, odologically unacceptable. However, also be attributed to other factors: the in all cases except one (in which chi- we included all the reports we identified relatively long period covered by the squared = 0.08) the chi-squared value as they provided the only available information (Table 3). It should be noted study and possible time-related changes for heterogeneity was significant in all that the definitions of commercial sex in the risk profile of the country; the cases. We also intended to use funnel work, promiscuous sexual behaviour considerable differences in the qual- plots for subgroups with a sufficient and breach of Islamic law by women ity of studies; and inconsistency in the number of studies, but the large hetero- overlapped loosely in the reports. There operational definition of risk determi- geneity within each subgroup (due to was no report on men who have sex nants in different studies. There was also differences in study quality, a possible with men. evidence of delays in the publication of time trend and inconsistent definitions many reports and papers. The results of poor knowledge and attitude) made Due to the profound heterogeneity of the reports, we decided not cannot therefore be compared even in it impossible to differentiate between to conduct any meta-analysis or use similar target populations. heterogeneity and possible publication scenario-based estimations. There were One of the limitations of this study bias, in fact any source of informa- 2 unofficial reports providing estimates was that we were not certain of having tion bias. As there was no information of the number of commercial sex work- found all the existing evidence. Many of regarding the magnitude of effect for ers in the Islamic Republic of Iran, one the reports cited in this study were ini- this risk determinant on HIV transmis- estimating it as 60 000 and another as tially produced only for internal circula- sion, it was impossible to estimate the 50 000. tion within an organization. In addition, avoidable burden of risk related to poor evidence on sexual relations between knowledge and negative attitudes. men, which could be important, was non-existent. This can be explained Injection drug use Discussion by the very strong cultural taboos sur- We identified 48 reports in the category A key finding of this study was the scar- rounding homosexuality in Muslim of IDU, of which 18 met the inclusion city of information, at both the national countries and the corresponding dearth criteria. The reports were divided into and regional level, regarding the magni- of opportunities to investigate the issue. 2 groups: IDU in prison and other cor- tude of effect for the 3 risk determinants The systematic review method helps rectional facilities and IDU in the com- studied. While it would be possible to identify information and evidence gaps munity, including drop-in centres and utilize measures from studies in other that should be filled if planning and outpatient treatment facilities (Table countries, such estimates might not policy-making are to be truly evidence- 2). Many of the reports in this category be applicable for policy-making in the based. In many developing countries, could be classified as “grey literature”, Islamic Republic of Iran because of fun- including ours, the infrastructure for obtained from key resource people in damental social, cultural and economic generating and sharing evidence is

261 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Table 1 Prevalence of poor knowledge and negative attitudes about transmission of HIV among different groups of people in the Islamic Republic of Iran Study group/location Year of Sample Poor Negative Statistics [reference] study size knowledge attitudes χ2 for P-value I2-value heterogeneity % % % Health care workers Kerman [6] 1994 105 34.3 – 160.2 < 0.001 98 Shiraz [7] 1997 47 74.4 – Yazd [8] 1997 168 4.2 – Karaj [9] 2002 217 3.7 0.5 High-school and college students Hamadan [10] 1996 1272 32.2 – 4501 < 0.001 66 Yazd [11] 1999 1850 58.2 14.4 Tehran [12] 2000 646 46.1 – Babol [13] 2000 81 20.3 48.1 Babol [13] 2000 69 48.1 56.5 Astaraa 2002 163 11.3 – Tehran [14] 2002 4641 22.3 – Tehranb 2004 424 0.0 31.0 Shirazc 2004 288 9.4 – Prisoners Zanjan [15] 1998 284 54.7 – 10 457 < 0.001 94 Dezfuld 2002 205 15.1 – Countrywidee 2003 2251 28.5 21.5 Runaway girls and female commercial sex workers Kermanshaha 2002 50 33.3 – 5.05 0.08 94 Tehran [16] 2003 110 38.9 59.3 Shiraz [17] 2004 106 25.0 – Barbers and hairdressers Gorgan [18] 2000 150 56.2 43.4 144 < 0.001 92 Boroujerd [19] 2003 200 13.0 – Sari [20] 2003 199 5.5 – General population West Azerbaijanf 1996 504 44.0 – 12 000 < 0.001 73 Sanandaj [21] 1998 852 33.7 – Kermanshah [22] 2000 2300 90.0 – Tehran [23] 2001 400 1.6 22.3 Astaraa 2002 100 24.2 – Saravana 2002 100 32.9 – Eslamshahra 2002 140 12.7 – Tehran [24] 2003 1172 13.7 – Shiraz [25] 2004 1200 11.2 –

aMalakafzali H. A multidisciplinary prevention/management program for HIV/AIDS in high risk areas of Iran, unpublished report, 2002. bAria P et al. Knowledge, attitude, and practice of youngsters about AIDS in 17th municipality district of Tehran, unpublished report 2004. cMirahmadizadeh A, Hemati A. Effect of health education on knowledge, attitude, and practice in Shiraz population, unpublished report, 2004. dKazemi M et al. Knowledge of Dezful Fajr addict prisoners toward AIDS, unpublished report, 2002. eMoradi Lakeh M, Afshar P. Knowledge, attitude, and practical skills of prisoners about health priorities in prisons, unpublished report, 2003. fShariatzadeh M et al. Knowledge attitude and skills towards AIDS in West Azerbaijan, unpublished report, 1996.

262 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 2 Prevalence of injection drug use (IDU) among different groups of people inside and outside prisons, and prevalence of HIV among drug users in the Islamic Republic of Iran Study group/location [reference] Year of Sample IDU HIV-positive study size Non-IDU IDU No. % % % Inside prison Shiraz, rehabilitation campa 1999 1061 44.0 0.76 1.2 Countrywide, rehabilitation campsb 2000 – – 1.14 – Countrywide, drug-related prisoners [26] 2000 – – 3.3 – Countrywide, prisons [26] 2002 2799 23.1 – 4.0 Tehran, prisonc 2003 370 50.0 – 24.0 Countrywide, prisonsd 2003 2251 9.8 – – Countrywide, prisons [27] 2003 2437 – – 4.0 Outside prison Countrywide [28] 1998 1500 21.2 – – Tehran [29] 2000 – > 5.0 – – Tehran [30] 2001 65 35.7 – – Tehran [4] 2002 200 14.0 5.0 15.0 Countrywide [31]e 2002 – 24.0 – – Kermanshahf 2001 384 – – 16.1 2003 – 4000–9000 – – Countrywide [32]e females (No.) Countrywideg 2003 – 242 000 (No.) Kermanshah, triangular clinich,i 2002 697 31.0 25.0 (among all drug users)

aPrisons Organization, unpublished report. bBolhari J. Drug abuse in prisons in Iran, unpublished report, 2002. cFarhoudi B, Afshar P. HIV–TB co-infection and risk factors in injection drug users in a prison in Iran, unpublished report, 2003. dMoradi Lakeh M, Afshar P. Knowledge, attitude, and practical skills of prisoners about health priorities in prisons, unpublished report, 2003. eThis reference is a review not primary research. fAlaie K. Sociodemographic factors in HIV/AIDS drug dependents in Iran, unpublished report, 2001. gVazirian M. A review of demand reduction programs in Iran: recommendations for strategic development plans, unpublished report, 2003. hKermanshah provincial health centre, personal communication. iTriangular clinics are facilities providing services dealing with injection drug use, sexually transmitted infections and HIV.

poorly developed and the studies that [33,34], the Islamic Republic of Iran has systematic reviews, practice guidelines are conducted do not follow standard also taken steps toward reducing the and knowledge translation [35]. research guidelines or protocols. In gap between knowledge and practice However, the great heterogeneity line with global recommendations for by development of educational capac- of the studies identified in our system- utilizing knowledge in health decisions ity-building programmes in the field of atic search means that they could not

Table 3 Prevalence of HIV among people with high-risk sexual practices in the Islamic Republic of Iran Location Study group Year of study HIV-positive (%) Kermanshaha Women in prison for misdemeanours against Islamic laws 2001–03 2.4 Kermanshahb HIV test volunteers who had unsafe sex 2001–02 13.0 Kemranshac CSW – 3.0 Kohkilouyec Homeless CSW – 11.0 Charmahalc Homeless CSW 2000 14.0 Countrywideb STI clinic patients (male and female) – 0.02 Unnamed districtb HIV test volunteers with no IDU history – 9.0

aKermanshah provincial health centre, personal communication. bMinistry of Health, personal communication. cPrisons Organization, personal communication. CSW = commercial sex workers; STI = sexually transmitted infections; IDU = injection drug use.

263 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

be used as evidence to guide policy and HIV incidence and prevalence trends in AIDS Programme (Ministry of Health planning. Accordingly, we recommend different populations within the Islamic and Medical Education), and Dr Ali- that the relevant government bodies Republic of Iran needs to be analysed reza Mirahmadizadeh. We would also take steps to establish the necessary in- alongside routine reporting data within like to thank Dr Kamran Yazdani for frastructure for improving the research a comprehensive, unified system under reviewing the manuscript draft and Dr base in HIV/AIDS. the leadership and support of the Na- Alireza Vassigh for reviewing the final We also recommend the develop- tional AIDS Committee. manuscript. ment of a virtual network of all the pro- The study was done with full finan- ducers and users of HIV/AIDS-related cial support from the United Nations information at country level as well as a Acknowledgements strengthening of existing regional links HIV/AIDS Theme Group (UNAIDS). to facilitate the sharing of knowledge be- The authors wish to acknowledge with Hamidreza Setayesh is now the UN- tween countries of the Eastern Mediter- thanks all the individuals who provided AIDS Country Officer but at the time ranean Region in order to plug the gaps in them with their reports, especially Dr of grant confirmation and most parts of the evidence. Monitoring and evaluation Behnam Farhoudi, Dr Sepideh Sigari, the study he was UNICEF Health Of- of current interventions on HIV/AIDS and Dr Parviz Afshar from the Prisons ficer. None of the other authors has any prevention deserve more attention. Organization, Dr Hengameh Namdari conflict of interest either with the grant- Finally, information about risk be- from Kermanashah health centre, Dr ing agency or the information sources haviours for HIV transmission as well as Kianoosh Kamali from the National used for the review.

References

1. 2006 Report on the global AIDS epidemic. Geneva, Joint United 12. Gachkar L, Amini M. [A study into knowledge and perception Nations Programme on HIV/AIDS, 2004. of female high-school students on AIDS, Tehran, 2000.] Pajou- 2. Greenland S. Causality theory for policy uses of epidemio- handeh quarterly research journal, 2001, 23:287–91 [in Farsi]. logical research. In: Murray CJL et al. eds. Summary measures 13. Omidvar SH. [Knowledge level and attitude of midwifery and of population health: concepts, ethics, measurement and applica- nursing students regarding AIDS, Babol, 2000.] Journal of tions. Geneva, World Health Organization, 2002:291–301. Babol University of Medical Sciences, 2003, 2:37–41 [in Farsi]. 3. Higgins JPT et al. Measuring inconsistency in meta-analyses. 14. Tavoosi A et al. Knowledge and attitude towards HIV/AIDS British medical journal, 2003, 327:557–60. among Iranian students. BMC public health, 2004, 4:17–23. 4. Zamani S et al. Prevalence of and factors associated with HIV-1 15. Vakili MM. [A study of the addicted men’s knowledge and at- infection among drug users visiting treatment centers in Te- titude measurement in Zanjan central prison about AIDS.] Jour- hran, Iran. AIDS, 2005, 19:709–16. nal of Zanjan University of Medical Sciences and Health Services, 5. Rothman KJ, Greenland S, Timothy L. Modern epidemiology, 3rd 1998, 23:17–23 [in Farsi]. ed. Philadelphia, Lippincott–Raven, 2008. 16. Ardalan A et al. [Survival sex, an outcome of running away for 6. Aminipour E, Ameri G. [Knowledge of school nurses about girls.] Social welfare quarterly, 2002, 2:187–219 [in Farsi]. AIDS.] Journal of Sabzevar University of Medical Sciences, 1997, 17. Honravar B, Negin H. [Knowledge, attitude, and practice of 1:53–8 [in Farsi]. female engaged in temporary marriage in Shiraz toward AIDS, 7. Shojaiezadeh D. [Evaluation of AIDS health education program 2003–4.] In: Proceedings of the Second National Congress on Epi- on knowledge, attitude, and practice of experimental assistant demiology, Zahedan, Islamic Republic of Iran, 2004 [in Farsi]. dentists in Shiraz.] Journal of Tehran Faculty of Medicine, 1997, 18. Mahmoudi G, Hosseini SA. [Knowledge, attitude and perform- 6:116–20 [in Farsi]. ance of barbers about AIDS prevention.] Journal of Gorgan 8. Zarifnasab M et al. [Influence of education on knowledge of University of Medical Sciences, 2000, 5:32–6 [in Farsi]. working nurses concerning AIDS.] Journal of Shahid Sadoughi 19. Amini M, Shojaizadeh D. Knowledge and attitude of hairdressers University of Medical Sciences and Health Services, 1999, 2:57–62 towards AIDS in Broujerd [thesis]. Tehran, Islamic Republic of [in Farsi]. Iran, Tehran University of Medical Sciences, School of Public 9. Sotoudeh M, Eftekhar H, Gouya M. Knowledge, attitude, and Health, 2003. practice in health workers of a health center in Karaj about uni- 20. Naghibi SA, Asadpour A. [Knowledge of barbers towards HBV versal precaution [MPH thesis]. Tehran, Islamic Republic of and AIDS in Sari 2003.] In: Proceedings of the Second National Iran, Tehran University of Medical Sciences, School of Public Congress on Epidemiology, Zahedan, Islamic Republic of Iran, Health, 2002. 2004 [in Farsi]. 10. Fathi Y, Tavakol M. [Students’ knowledge in Hamadan Univer- 21. Memar N. [Knowledge of people over 15 years old about sity of Medical Sciences about AIDS in 1995–96.] Scientific Jour- prophylaxis and transferring of AIDS and Hepatitis B in nal of Hamadan University of Medical Sciences Health Services, Sanandaj.] Journal of the Medical Faculty of Guilan University of 1996, 7:40–6 [in Farsi]. Medical Sciences, 2001, 37:13–8 [in Farsi]. 11. Karimi M et al. [Attitude and knowledge of high school students 22. Darabi F. Study of the knowledge of people about AIDS in Ker- in Yazd towards AIDS.] Journal of Shahid Sadoughi University of manshah, 2001. Behboud, Journal of Kermanshah University of Medical Sciences Health Services, 2000, 4:5–10 [in Farsi]. Medical Sciences, 2000, 3:64–70.

264 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

23. Asghari S, Majlesi F. Knowledge and attitude of marriage candi- 29. Rahimi Movaghar A et al. Assessment of shared injection equip- dates towards HIV/AIDS in a health center in Tehran [MPH thesis]. ment among IDUs: research report. Tehran, Islamic Republic of Tehran, Islamic Republic of Iran, Tehran University of Medical Iran, United Nations Office on Drugs and Crime, 2000. Sciences, School of Public Health, 2001. 30. Samiei M, Rafiei H, Taheri Nakhost HR. [Psychosocial profile 24. Montazeri A. AIDS knowledge and attitudes in Iran: results of Iranian adolescent addicts.] Journal of rehabilitation, 2002, from a population-based survey in Tehran. Patient education 10:56–62 [in Farsi]. and counselling, 2005, 57:199–203. 31. Mokri A. Brief overview of the status of drug abuse in Iran. Ar- 25. Mirahmadizadeh AR et al. [Prevalence of HIV infection among chives of Iranian medicine, 2002, 5:184–90. intravenous drug addicts in Shiraz, 1998.] Journal of Gorgan University of Medical Sciences, 2001, 8:39–42 [in Farsi]. 32. Rahimi Movaghar A. [Drug abuse prevalence and model in women.] Social welfare quarterly, 2004, 3:203–26 [in Farsi]. 26. Centre for Disease Control sentinel site reports. Tehran, Islamic Republic of Iran, Ministry of Health and Medical Education, 33. World report on knowledge for better health. Strengthening health 2002. systems. Geneva, World Health Organization, 2004. 27. Centre for Disease Control sentinel site reports. Tehran, Islamic 34. Global ministerial forum on research for health. Strengthening re- Republic of Iran, Ministry of Health and Medical Education, search for health, development, and equity. Bamako, Mali, 17–19 2003. November, 2008 (http://www.bamako2008.org, accessed 20 28. Rahimi Movaghar A, Mohammad K, Razaghi EM. [Trend of drug October 2009). abuse situation in Iran: a three-decade survey.] Hakim research 35. Majdzadeh R et al. Iran’s approach to knowledge translation. journal, 2002, 3:171–81 [in Farsi]. Iranian journal of public health, 2009, 38(Suppl. 1):58–60.

HIV prevention and care among injecting drug users in the Islamic Republic of Iran: a review of best practice

This document reviews the implementation of harm reduction programmes, care and support services for people living with HIV/AIDS, and the attempts to reach and deliver services to hidden populations of injecting drug users in the Islamic Republic of Iran. It is intended for programme managers and policy-makers in all countries in order to share the Iranian experience as a model for a comprehensive approach to introducing and scaling up harm reduction. In addition to identifying harm reduction programme elements and their implementation, the document can also be a useful resource for advocacy.

A number of recommendations are also provided in order to enhance the Iranian response to HIV among injecting drug users. These recommendations can be used by policy-makers and programme managers in the Islamic Republic of Iran to improve and accelerate the positive impact of the services. They may also serve as useful programmatic tips for programme developers currently setting up harm reduction services in other countries of the Region. The full text of this document is freely available at: http://www.emro.who.int/asd/pdf/hiv_review_iran.pdf

265 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Changes in tobacco use among 13–15-year-olds between 1999 and 2007: findings from the Eastern Mediterranean Region F. El-Awa,1 C.W. Warren2 and N.R. Jones2

التغريات يف تعاطي التبغ بني املراهقني بأعامر 13 – 15 ًعاما يف املدة 1999 – : 2007موجودات من إقليم رشق املتوسط فاطمة ّالعوا، شارز وارن، ناثن جونز يركز هذااخلالصـة: التقرير عىل التغريات التي تراكمت بمرور الزمن يف تعاطي التبغ بني املراهقني يف بلدان إقليم رشق املتوسط، فقد ُأجري املسح العاملي حول تعاطي الشباب للتبغ يف كل موقع من اإلقليم مرتني عىل األقل يف املدة 1999 – . 2007وتشري النتائج إىل أنه لدى الطالب الذين ترتاوح أعامرهم بني 13 و15 ًعامايكون تعاطي التبغ أحد املشكالت الصحية العمومية، كام لوحظ ازدياد يف تعاطي الشيشة، ويف احتامل البدء بالتدخني ممن مل يسبقهلم تعاطيه، واحتامل ازدياد تعاطي التبغ لدى الشابات الصغريات يف معظم بلدان اإلقليم. ويمكن جلميع بلدان اإلقليم التي سامهت يف املسح العاملي حول تعاطي الشباب للتبغ االستفادة من نتائجه يف ما لدهيم من برامج ملكافحة التبغ ويف إعداد جدول األعامل والسياسات لذلك.

ABSTRACT This report focuses on change over time in tobacco use among adolescents in countries included in the Eastern Mediterranean Region (EMR) of the World Health Organization. The Global Youth Tobacco Survey (GYTS) was conducted in each site at least twice between 1999 and 2007. Results indicate that for students aged 13–15 years tobacco use is a major public health problem. Increase in the use of water pipe, the likely initiation of smoking by never smokers, and a potential increase in tobacco use among young girls was found in most of the EMR sites. The results from the GYTS can be used by all of the EMR countries involved to set their tobacco control programme and policy agenda.

Changement au niveau de la consommation de tabac chez les 13-15 ans entre 1999 et 2007 : constatations pour la Région de la Méditerranée orientale

RÉSUMÉ Le présent article concerne les changements survenus durant une certaine période au niveau de la consommation de tabac chez les adolescents dans des pays de la Région OMS de la Méditerranée orientale. L’enquête mondiale sur le tabagisme chez les jeunes a été conduite au moins deux fois sur chaque site entre 1999 et 2007. Les résultats montrent que pour les adolescents âgés de 13 à 15 ans, le tabagisme constitue un problème de santé publique majeur. Une augmentation de l’utilisation des pipes à eau, l’initiation probable au tabagisme de personnes n’ayantt jamais fumé, et une hausse potentielle du tabagisme chez les jeunes filles ont été observées sur la plupart des sites de la Région de la Méditerranée orientale. Les résultats de cette enquête peuvent être utilisés par l’ensemble des pays de la Région concernés afin de mettre en place un programme de lutte antitabac et d’orienter leurs politiques en la matière.

1Tobacco Free Initiative, Division of Health Promotion and Protection, World Health Organization Regional Office for the Eastern Mediterranean, Cairo, Egypt. 2Office on Smoking and Health, Centers for Disease Control and Prevention, Atlanta, Georgia, United States of America (Correspondence to C.W. Warren: [email protected]). Received: 12/11/07; accepted: 07/02/08

266 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Introduction resources available and safety concerns, and 2007), Tunisia (2001 and 2007) the scope of the GYTS can be national and the United Arab Emirates (UAE) Tobacco use is one of the major pre- or regional or focused on specific urban (2002 and 2005). These 13 countries ventable causes of premature death and or rural areas. In addition, many samples and 2 administrative regions will be disease in the world. A disproportionate are designed to yield information that referred to throughout the remainder of share of the global tobacco burden falls is representative of country geographi- this report as sites. on developing countries where 84% cal regions, but can be pooled to yield Data collection instrument of the 1.3 billion current smokers live nationally representative estimates. [1]. The Global Youth Tobacco Survey After the sampling frames are de- The core GYTS instrument includes (GYTS), part of the Global Tobacco fined, the GYTS research coordinator 54 questions covering 7 categories: Surveillance System (GTSS) initi- provides CDC with school enrolment tobacco use, knowledge and attitudes ated by the World Health Organization information and the samples are drawn regarding tobacco, second-hand smoke (WHO) and the Centers for Disease using a standard protocol and soft- exposure, pro- and anti-tobacco media Control and Prevention (CDC) in ware developed by CDC. The GYTS and advertising exposure, desire for 1999, was developed to monitor youth standard sampling methodology uses cessation, access and availability to ob- tobacco use, attitudes and exposure a 2-stage cluster sample design that tain tobacco, and having been taught to tobacco smoke, and has been com- produces samples of students in grades in school about the harmful effects of pleted by over 2 million students in 151 associated with students of 13–15 years tobacco use. Research coordinators are countries [2]. A key goal of GTSS is for of age. Each sampling frame includes all encouraged to develop and add ques- countries to repeat the GYTS every 4 schools (usually public and private) in a tions to the core questionnaire to gather years. geographically defined area containing information important to their country Our report presents findings from any of the identified grades. At the first or WHO Region. The final country GYTS conducted more than one time stage, the probability of schools being questionnaires are translated into in 13 countries and the areas adminis- selected is proportional to the number local languages as needed and back- tered by the Palestinian Authority in the of students enrolled in the specified translated to check for accuracy. GYTS Gaza Strip and West Bank in WHO’s grades. At the second sampling stage, country research coordinators conduct Eastern Mediterranean Region (EMR) classes within the selected schools are focus groups of students aged 13–15 between 1999 and 2007. The report randomly selected. All students in the years to further test the accuracy of the focuses on changes in the proportion selected classes attending school the day translation and student comprehension of students who currently smoke ciga- the survey is administered are eligible of the questions. rettes, currently use other tobacco prod- to participate. Student participation is Variable definitions ucts (such as water pipe, locally known voluntary and anonymous using a self- as shisha or nargileh), and never smokers administered questionnaire. The GYTS This report describes changes in several who are susceptible to initiate cigarette sample design produces representative, important tobacco indicators between smoking in the next year. independent, cross-sectional estimates 1999 and 2007, including current for each sampling frame. cigarette smoking, use of other tobacco products and susceptibility of never Methods GYTS participants smokers to initiate cigarette smoking. This report includes data from 13 EMR • Current cigarette smokers are defined Sampling countries and the areas administered as students who answered “one or The GYTS is a school-based survey that by the Palestinian Authority. Repeat more days” to the question, “During collects data from students aged 13–15 GYTS surveys were conducted in Egypt the past 30 days (1 month), on how years using a standardized methodology (2001 and 2005), Jordan (1999, 2003 many days did you smoke cigarettes?” for constructing the sample frame, se- and 2007), Kuwait (2001 and 2005), • Current users of other tobacco prod- lecting schools and classes, and process- Lebanon (2001 and 2005), Libyan Arab ucts are defined as students who an- ing data. Within each country, the scope Jamahiriya (2003 and 2007), the Gaza swered “yes” to the question, “During of the GYTS is defined through con- Strip (2000 and 2005), the West Bank the past 30 days (1 month), have sultation between the country GYTS (2000 and 2005), Morocco (2001 and you ever used any form of tobacco research coordinator, WHO Tobacco 2006), Oman (2002 and 2007), Qatar products other than cigarettes, e.g. Free Initiative (TFI) regional advisers (2004 and 2007), Somalia–Somaliland water pipe, chewing tobacco, snuff, and CDC technical advisers. Depending (2004 and 2007), Sudan (2001 and dip (dipping tobacco), cigars, cigaril- on the data requirements of the country, 2005), Syrian Arab Republic (2002 los, little cigars, pipe?”

267 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

• Never smokers who are susceptible Results Bank, 2005; Oman, 2002 and 2007; to start cigarette smoking in the next Syrian Arab Republic, 2002 and 2007; year are defined as 100% minus the Response rates Tunisia, 2001 and 2007; and UAE, percentage of never smokers who Sample sizes and response rates for the 2002 and 2005 (Table 1). The student answered “definitely not” to the ques- surveys by country are presented in response rate ranged from 83.7% (Syr- tion, “If one of your best friends offered ian Arab Republic, 2007) to over 99% Table 1. you a cigarette, would you smoke it?” in Lebanon (2005) and Libyan Arab and also answered “definitely not” The school response rates ranged Jamahiriya (2003). The overall re- to the question, “At any time in the from 90.2% (Egypt, 2005) to 100.0% sponse rate ranged from 77.0% (Egypt, next 12 months do you think you will (Egypt, 2001; Jordan, 2003 and 2007; 2005) to 98.3% (Syrian Arab Republic, smoke a cigarette?” the Gaza Strip, 2000 and 2005; the West 2002). The number of students who Susceptibility to cigarette smoking defined in the above-mentioned way has been strongly associated with sub- Table 1 Response rates in the Global Youth Tobacco Surveys (GYTS) in countries in sequent experimentation with cigarettes the Eastern Mediterranean Region of the World Health Organization, 1999–2007 among non-smoking students [3]. In Site Year survey School Student Overall Number of multivariate analyses, the susceptibility completed response response response students to initiate cigarette smoking measure rate (%) rate (%) rate (%) who completed was a stronger predictor of experimenta- the GYTS tion than the presence of cigarette smok- Egypt 2001 100.0 96.3 96.3 3 792 ers among family and best friends [3]. 2005 90.2 85.4 77.0 4 196 Response rate definitions Jordan 1999 91.0 92.2 83.9 3 912 2003 100.0 89.1 89.1 6 313 The school response rate is defined 2007 100.0 91.6 91.6 2 250 as: number of participating schools/ Kuwait 2001 100.0 94.8 94.8 6 330 number of selected schools. The stu- 2005 100.0 88.7 88.7 3 935 dent response rate is defined as: number Lebanon 2001 98.0 98.3 96.4 4 951 of students who completed the survey/ number of students enrolled in the class. 2005 98.0 99.2 95.2 3 314 Libyan Arab 2003 98.0 99.0 97.0 1 850 The overall response rate is defined as: Jamahiriya school response rate × student response 2007 100.0 94.1 94.1 2 028 rate. Gaza Strip 2000 100.0 95.8 95.8 2 906 2005 100.0 94.5 94.5 2 109 Data analysis West Bank 2000 98.7 94.8 93.5 8 374 A weighting factor is applied to each stu- 2005 100.0 95.6 95.6 2 182 dent record to adjust for non-response Morocco 2001 98.0 94.8 92.9 3 147 (by school, class and student) and vari- 2006 98.0 93.5 91.6 3 186 ation in the probability of selection at Oman 2002 100.0 96.9 96.9 1 962 the school and class levels. A final adjust- 2007 100.0 96.8 96.8 2 297 ment sums the weights by grade and Qatar 2004 92.0 91.9 84.5 3 240 sex to the population of schoolchildren 2007 96.0 90.9 87.3 1 434 in the selected grades in each sample Somalia–Somaliland 2004 88.0 94.2 82.9 1 563 site. SUDAAN, a software package for 2007 96.0 90.2 86.6 1 998 statistical analysis of correlated data, was Sudan 2001 94.0 94.2 88.5 2 783 used to calculate weighted prevalence 2005 92.0 93.2 85.8 4 277 estimates and standard errors (SE) Syrian Arab Republic 2002 100.0 98.3 98.3 4 531 of the estimates; the 95% confidence 2007 100.0 83.7 83.7 2 039 intervals (CI) were calculated from Tunisia 2001 100.0 94.1 94.1 4 282 the SEs [4]. In this paper differences in 2007 100.0 92.4 92.4 2 155 proportions are considered statistically United Arab 2002 100.0 95.1 95.1 4 178 significant at the P < 0.05 level if the 95% Emirates 2005 100.0 93.1 93.1 16 447 CIs do not overlap.

268 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

participated varied due to the level of (Table 2). Boys were significantly more Susceptibility of never stratification included in each sample likely than girls to use other tobacco smokers to initiate cigarette smoking in the next year design. For example, Jordan conducted products in every site, except Egypt, a single national sample in 1999, but Libyan Arab Jamahiriya, Morocco, During the initial surveys, susceptibil- stratified the country into 3 regions in Qatar, Somalia–Somaliland, Sudan and ity to initiate cigarette smoking among 2003; Lebanon stratified in urban and Syrian Arab Republic, which had no never smokers was highest in Somalia– rural samples in both 2001 and 2005; gender difference. Somaliland (25.0%), Sudan (22.3%), Tunisia (19.8%) and Libyan Arab and the West Bank was stratified into 3 Between the initial surveys and regions in 2000 but not in 2005. Jamahiriya (19.5%) and lowest in the the repeat surveys, the level of use of Gaza Strip (6.6%) (Table 2). Boys were other tobacco products significantly Current cigarette smoking significantly more likely than girls to be decreased in Egypt and Jordan, signifi- At the time of the initial surveys, cur- susceptible to initiate cigarette smoking cantly increased in Tunisia and UAE, rent cigarette smoking was highest in in Lebanon, the West Bank, Morocco, Somalia–Somaliland (18.6%), Jordan but did not change significantly in the Tunisia and UAE; in all other sites, there (16.6%) and the West Bank (14.2%), other sites. Also at the time of the repeat was no gender difference. surveys, the lack of statistical difference and lowest in Morocco (2.6%), Libyan Between the initial surveys and the between boys and girls in use of other Arab Jamahiriya (4.1%) and Egypt repeat surveys, susceptibility to initiate (4.2%) (Table 2). Boys were signifi- tobacco products remained in Libyan cigarette smoking significantly increased cantly more likely than girls to smoke Arab Jamahiriya, Morocco, Qatar, in Jordan, the Gaza Strip and the West cigarettes in every site except Egypt and Somalia–Somaliland and Sudan, while Bank, significantly decreased in Sudan, Somalia–Somaliland, which had no and boys continued to have significantly but did not change significantly in the gender differences. higher levels than girls in Kuwait, Leba- other sites. Also at the time of the repeat Between the initial survey and the non, the West Bank, Tunisia and UAE. surveys, the lack of statistical difference repeat survey, the level of smoking However, in Egypt and Syrian Arab between boys and girls in susceptibility decreased significantly in Jordan and Republic the relationship changed from to initiate cigarette smoking remained in Somalia–Somaliland, increased sig- no gender difference to boys having Jordan, Kuwait, Libyan Arab Jamahiriya, nificantly in the Syrian Arab Republic, significantly higher levels of use of other the Gaza Strip, Oman, Qatar, Somalia– but did not change significantly in the tobacco products than girls. In addition, Somaliland, Sudan, and Syrian Arab other sites. Also, at the time of the repeat in Jordan, the Gaza Strip and Oman the Republic, and boys continued to show surveys, gender differences in smok- relationship changed from boys hav- significantly higher susceptibility than ing behaviour did not change in most ing significantly higher levels of use of girls in Tunisia and UAE. However, in of the sites. Boys continued to have other tobacco products than girls to no Egypt the relationship changed from significantly higher levels of smoking gender difference. no gender difference to boys being sig- than girls in Kuwait, Lebanon, Libyan Current use of other tobacco nificantly more susceptible than girls, Arab Jamahiriya, the Gaza Strip, the products was significantly higher than and in Lebanon and the West Bank the West Bank, Qatar, Sudan, Syrian Arab cigarette smoking in Egypt, Lebanon, relationship changed from boys being Republic, Tunisia, and UAE. In Jordan, Morocco, Qatar, Syrian Arab Republic significantly more susceptible than girls Morocco and Oman there was a change to no gender difference. and UAE at both the initial and repeat from boys having significantly higher surveys. Current use of other tobacco levels of current smoking at the initial products changed from being signifi- survey to no gender difference at the Discussion time of the repeat survey; and in Egypt cantly higher than cigarette smoking to no gender difference at the initial survey no difference in Kuwait, Libyan Arab Results from this study document changed to boys having significantly Jamahiriya and Sudan. In the Gaza Strip, that tobacco use among students aged higher rate of smoking than girls at the the West Bank, Somalia–Somaliland 13–15 years is a significant public health repeat survey. and Tunisia, the relationship changed problem in EMR. Three major prob- from no difference to the use of other lems have been identified. Current other tobacco use tobacco products being significantly First, use of other tobacco products During the initial surveys, current use higher than cigarette smoking. In Jordan (especially shisha or nargileh) is a ma- of other tobacco products was highest the relationship changed for cigarette jor concern in all the sites, especially in Lebanon (38.6%) and lowest in Tu- smoking being higher than use of other Lebanon where almost 60% of the ado- nisia (7.2%) and the Gaza Strip (7.8%) tobacco products to no difference. lescents are current users. At the time of

269 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale NA Girls 7.5 (5.8–9.5) 7.5 7.5 (5.7–9.8) 7.5 5.1 (3.5–7.4) 5.1 11.5 (7.4–17.4) 11.5 11.7 9.6–14.2) 11.7 17.8 (14.0–22.5) 17.8 15.9 (13.8–18.3) 15.9 (11.1–21.8) 15.7 15.0 (11.5–19.3) 15.0 21.4 (10.9–37.6) 21.4 12.9 (11.1–15.0) 14.1 (10.9–18.0) 14.1 14.9 (12.0–18.4) 19.7 (16.9–22.9) 19.7 16.8 (13.7–20.5) 16.9 (14.5–19.6) 18.2 (16.0–20.7) 10.4 (7.8–13.9) 10.4 10.7 (6.6–16.7) 10.8 (6.9–16.5) 22.2 (15.4 -30.8) 22.2 (15.4 20.3 (17.2–23.9) NA Boys (95% CI) a % 8.6 (4.3–16.4) 17.1 (14.7–19.8) 17.1 15.2 (10.5–21.4) 15.2 15.1 (12.3–18.3) 15.1 15.5 (12.0–19.8) 15.5 21.8 (17.3–27.1) 21.8 21.5 (18.8–24.4) 21.5 21.1 (15.9–27.5) 21.1 14.2 (10.6–18.7) 14.8 (12.7–17.2) 19.0 (12.5–27.8) 19.0 19.5 (16.2–23.2) 19.5 (15.9–22.9) 19.2 19.1 (16.4–22.1) 19.1 16.1 (12.3–20.8) 16.1 23.0 (13.0–37.3) 23.0 -33.8) (18.1 25.1 23.5 (18.2–29.7) 23.5 25.5 (20.7–31.1) 25.5 23.9 (18.4–30.4) 23.9 22.1 (18.0–26.9) 22.1 22.3 (18.7–26.5) NA Total Never smokers susceptibleNever smokers to initiate smoking 11.3 (9.7–13.1) 11.3 6.6 (4.1–10.3) 13.2 (11.3–15.4) 13.2 12.5 (9.3–16.5) 17.3 (15.3–19.4) 17.3 (13.6–17.5) 15.5 10.0 (8.3–11.9) 12.6 (9.4–16.7) 15.0 (11.5–19.3) 15.0 24.1 (17.8–31.8) 24.1 13.1 (10.4–16.4) 13.1 18.3 (15.7–21.3) 15.5 (13.0–18.5) 15.5 19.5 (15.4–24.3) 19.5 20.7 (17.6–24.1) 18.5 (15.2–22.3) 16.9 (14.9–19.3) 21.5 (18.9–24.4) 21.5 25.0 (15.5 -37.8) (15.5 25.0 20.2 (17.0–23.7) 20.6 (17.4–24.2) 18.7 (16.8–20.6) Girls 7.5 (5.1–10.8) 7.5 7.1 (5.7–8.8) 7.1 7.6 (6.0–9.7) 7.6 5.6 (4.1–7.7) 5.6 3.6 (2.1–6.0) 3.6 9.8 (5.2–17.6) 9.8 8.1 (6.3–10.4) 8.1 6.7 (4.7–9.3) 6.9 (5.5–8.7) 6.7 (5.1–8.6) 11.7 (9.9–13.9) 11.7 15.0 (7.3–28.4) 15.0 12.6 (9.3–16.8) 12.0 (9.9–14.5) 12.9 (11.3–14.7) 12.0 (8.4–16.7) 12.7 (9.8–16.3) 18.2 (16.3–20.3) 33.9 (30.2–37.8) 33.9 35.7 (32.5–39.1) 35.7 10.6 (8.1–13.7) 10.0 (8.7–11.6) 10.0 (6.9–14.3) Boys (95% CI) a % 8.6 (5.2–14.0) 11.7 (8.1–16.4) 11.7 17.4 (15.0–20.1) 17.4 15.0 (7.4–28.1) 15.0 15.9 (13.1–19.1) 15.9 21.4 (19.0–24.0) 21.4 12.7 (10.2–15.8) 12.4 (8.4–17.8) 12.4 12.3 (9.5–15.8) 12.8 (9.3–17.2) 14.2 (10.8–18.4) 14.5 (11.4–18.3) 19.4 (15.7–23.8) 19.4 19.1 (17.1–21.4) 19.1 16.9 (12.8–22.0) 18.3 (14.2–23.4) 10.5 (8.2–13.4) 10.3 (7.8–13.5) 10.4 (9.0–11.9) 10.4 23.7 (20.8–26.8) 23.7 45.0 (39.9–50.2) 45.0 44.7 (40.8–48.6) 20.8 (17.2–24.8) Current otherCurrent use tobacco Total Total 7.2 (5.4–9.5) 7.2 7.8 (5.5–10.9) 7.8 9.8 (7.4–12.8) 9.8 9.5 (8.3–10.9) 9.5 (7.4–12.0) 9.5 9.0 (7.5–10.8) 9.0 9.2 (8.1–10.5) 9.2 11.2 (9.1–13.6) 11.2 11.7 (9.1–14.9) 11.7 13.7 (12.0–15.7) 13.7 15.6 (13.1–18.6) 15.6 15.3 (12.5–18.7) 15.3 12.5 (10.1–15.4) 14.5 (12.3–16.9) 14.4 (11.4–18.0) 14.4 18.7 (11.1–29.6) 16.2 (14.7–17.9) 16.6 (14.7–18.7) 16.7 (14.2–19.5) 10.1 (8.1–12.4) 10.1 38.6 (35.3–42.0) 20.0 (18.2–21.8) 40.0 (37.8–42.3) Girls 1.2 (0.3–4.1) 1.2 1.6 (0.8–3.0) 1.6 1.4 (0.9–2.3) 1.4 1.0 (0.4–2.2) 1.0 7.1 (4.9–10.3) 7.1 5.3 (3.9–7.2) 5.3 5.6 (4.2–7.5) 5.6 3.4 (2.2–5.3) 3.4 3.0 (1.6–5.4) 3.0 2.3 (1.0–5.1) 2.8 (1.7–4.8) 9.9 (7.8–12.6) 9.9 2.1 (1.1–3.9) 2.1 8.7 (5.8–12.8) 4.5 (1.6–11.8) 4.9 (3.9–6.2) 4.5 (3.0–6.9) 4.0 (2.1–7.6) 4.7 (3.8–6.0) 0.8 (0.3–2.4) 0.9 (0.3–2.5) 12.6 (10.1–15.7) 14.8 (7.4–27.5) Boys (95% CI) a % 7.3 (4.5–11.6) 7.3 (4.9–11.9) 7.7 5.9 (4.4–7.9) 5.9 3.5 (1.8–6.6) 3.5 3.9 (2.9–5.3) 3.9 3.9 (2.7–5.5) 3.9 9.7 (6.3–14.6) 9.7 8.6 (3.6–19.3) 4.9 (3.2–7.4) 4.3 (2.9–6.4) 11.8 (8.5–16.3) 11.8 17.7 (14.2–21.7) 17.7 13.4 (9.5–18.7) 13.4 13.2 (9.9–17.5) 13.2 (6.1–27.4) 13.6 15.1 (9.8–22.4) 15.1 21.4 (17.9–25.4) 21.4 27.6 (21.3–35.1) 27.6 14.8 (12.8–17.0) 10.4 (7.6–14.1) 10.4 10.7 (8.8–13.0) 22.0 (18.2–26.4) 24.7 (21.5–28.3) Current cigarette smoking cigarette Current Total 7.5 (6.1–9.2) 7.5 5.8 (4.0–8.4) 5.8 3.5 (2.7–4.6) 3.5 2.3 (1.1–4.8) 9.0 (6.4–12.5) 9.0 2.6 (1.9–3.4) 8.6 (6.8–10.8) 6.5 (4.7–8.9) 6.4 (5.3–7.8) 6.4 6.1 (3.0–12.2) 6.1 6.6 (3.9–10.9) 4.1 (2.6–6.5) 4.1 4.6 (2.9–7.2) 4.0 (2.7–5.8) 4.2 (2.8–6.1) 17.7 (14.7–21.3) 17.7 14.2 (11.2–17.7) 18.6 (10.9–29.8) 16.6 (13.8–19.8) 18.0 (12.5–25.3) 10.0 (8.5–11.7) 10.8 (7.7–15.1) 10.3 (7.9–13.3) Year 1999 2001 2001 2001 2001 2007 2007 2007 2007 2007 2005 2005 2005 2005 2005 2003 2003 2002 2004 2004 2006 2000 2000 Somaliland Jamahiriya Somalia- Libyan Arab Lebanon Kuwait Egypt Country Morocco Jordan Qatar Oman Gaza Strip West Bank West Current smoking, 2 Current other current susceptible use, and never smokers tobacco Table to initiate smoking, 1999–2007 EMR GYTS, by country and sex,

270 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

the repeat survey, current use of other tobacco products was significantly higher than cigarette smoking in Egypt, Gaza Strip, West Bank, Lebanon, Morocco, Qatar, Somalia–Somaliland, Girls Syrian Arab Republic, Tunisia and UAE; and there was no dif- 7.5 (5.4–10.4) 7.5 8.3 (6.2–11.1) 11.1 (9.6–12.8) 11.1 15.3 (13.1–17.8) 15.3 15.5 (11.8–20.3) 15.5 13.2 (8.4–20.3) 13.2 12.5 (9.0–17.0)

20.9 (16.9–25.6) ference in Jordan, Kuwait, Libyan Arab Jamahiriya and Sudan. Thus, the tobacco control effort in EMR must include intense efforts directed toward reducing the influence of other tobacco products. Boys (95% CI) a Second, susceptibility to initiate cigarette smoking among % 17.8 (13.8–22.7) 17.8 13.3 (11.1–15.9) 13.3 14.3 (13.1–15.7) 14.4 (12.4–16.7) 14.4 10.3 (7.4–14.0) 24.2 (19.8–29.3) 26.6 (23.4–30.0) 26.7 (21.9–32.1) never smokers increased significantly in Jordan, Gaza Strip and West Bank; decreased significantly in Sudan; but did not change significantly in the other sites. At the time of the repeat survey, there was no difference in the level of susceptibility for Total

Never smokers susceptibleNever smokers to initiate smoking boys and girls, except in Egypt, Tunisia and UAE. These find- 9.4 (7.3–12.0) 9.4 13.9 (10.5–18.3) 13.9 12.5 (11.4–13.7) 14.7 (12.0–17.8) 19.8 (17.7–22.2) 19.8 19.9 (16.5–23.9) 19.9 10.2 (8.3–12.5) 22.3 (19.3–25.6) ings suggest cigarette smoking could increase dramatically in these populations in the near future. Third, tobacco use among young girls appears to be under-

Girls going major changes. The most recent data from the GYTS 7.8 (5.8–10.4) 7.8 3.1 (2.4–4.1) 3.1

9.3 (7.0–12.2) 9.3 on the prevalence of smoking among girls is as high, or higher, 15.3 (12.4–18.9) 15.3 14.5 (11.0–18.9) 10.2 (8.1–12.9) 10.4 (8.0–13.5) 10.4 24.7 (21.9–27.7) than the prevalence of smoking among adult women in 9 of the 11 countries where comparisons can be made (Table 3). In addition, the boy:girl ratio is less than the male:female ratio

Boys in 10 of the 11 countries. These findings indicate tobacco use (95% CI) a

% among females may be increasing in EMR. Thus, tobacco use 11.3 (9.2–13.9) 11.3 11.0 (7.8–15.4) 11.0 17.2 (13.9–21.0) 17.2 19.9 (16.1–24.3) 19.9 (16.1–22.3) 19.0 19.2 (16.0–22.9) 19.2 32.7 (30.4–35.1) 29.7 (25.1–34.7) 29.7 among females, especially young girls, should be a priority in

Current otherCurrent use tobacco the EMR countries. For decades the tobacco industry has targeted females and

Total Total continues to expand this market [5,6]. The tobacco industry 7.2 (5.9–8.7) 7.2

17.6 (15.0–20.6) 17.6 targets women through advertisements showing smoking 13.9 (11.6–16.5) 13.9 15.0 (13.0–17.3) 15.0 13.5 (11.2–16.2) 13.5 10.2 (8.0–12.9) 22.6 (19.1–26.4) 28.8 (26.7–30.9) associated with independence, stylishness, weight control, sophistication and power [7]. The industry markets Virginia Slims, Capri, Misty and Camel No. 9 directly to women using

Girls feminine images. In addition, gender-neutral brands such as 1.6 (0.8–3.1) 1.6 1.9 (1.1–3.5) 1.9 5.9 (4.3–8.2) 5.9 3.6 (2.9–4.4) 3.6 3.0 (2.1–4.4) 3.0 3.1 (1.8–5.4) 3.1 2.2 (1.5–3.4) 2.1 (1.4–3.2) 2.1 Marlboro are marketed to women using imagery of independence and “fun-loving”. Also, gender norms constantly change. GYTS data showed susceptibility to initiate smoking among never smokers was significantly more prevalent than current Boys

(95% CI) cigarette smoking for girls in all sites. This might be an indica- a % 8.1 (5.7–11.4) 8.1 11.7 (8.9–15.4) 11.7 17.6 (14.7–21.0) 17.6 15.1 (12.3–18.4) 15.1 12.1 (10.3–14.1) 12.1

19.1 (14.6–24.7) 19.1 tion that EMR cultural traditions and social influences may be 10.8 (6.3–17.8) 10.2 (6.6–15.5) changing thus making smoking among women and young girls

Current cigarette smoking cigarette Current more acceptable both at home and in public [8]. These findings suggest the tobacco control programmes in EMR countries

Total should focus on cessation among adult men and on prevention 8.3 (6.6–10.4) 8.0 (6.6–9.7) 6.8 (5.2–8.9) 6.1 (3.8–9.8) 6.1 6.0 (3.6–10.0) 6.3 (4.8–8.1) 12.3 (9.3–16.1)

10.1 (8.5–11.8) 10.1 among women and youth (of both sexes). All of the EMR sites can use the results from the GYTS to inform politicians and policy-makers about the tobacco Year 2001 2001 2007 2007 2005 2005 2002 2002 problems in their populations. The information can be used as baseline data for decision-making concerning National Tobacco Control Action Plans. In addition, Egypt, Jordan, Ku- wait, Lebanon, Libyan Arab Jamahiriya, Oman, Qatar, Sudan, Emirates Republic Sudan Syrian Arab Country United Arab percentage. Weighted Tunisia a EMR = Eastern Mediterranean Region; GYTS = Global Youth Tobacco Surveys; NA = not available. Tobacco = Global Youth Region; GYTS EMR = Eastern Mediterranean Current smoking, 2 Current other current susceptible use, and never smokers tobacco Table to initiate smoking, 1999–2007 (concluded) EMR GYTS, by country and sex, Syrian Arab Republic and UAE can use the information as part

271 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Table 3 Gender ratio of smoking prevalence for adults and youth aged 13–15 years in selected countries of the Eastern Mediterranean Region Country Adult smoking prevalencea Most recent GYTS Males (%) Females (%) Male:female Boys (%) Girls (%) Boy:girl ratio ratio Egypt 45.4 12.1 3.8:1 5.9 1.4 4.2:1 Jordan 50.5 8.3 6.1:1 13.2 7. 1 1.9:1 Kuwait 34.4 1.9 18.1:1 17.7 4.5 3.9:1 Lebanon 42.3 30.6 1.4:1 11.8 5.6 2.1:1 Libyan Arab Jamahiriya NA NA NA 7.7 0.9 8.6:1 Gaza Strip NA NA NA 9.7 3.0 3.2:1 West Bank NA NA NA 27.6 8.7 3.2:1 Morocco 28.5 0.1 285.0:1 4.3 2.1 2.0:1 Oman 15.5 1.5 10.3:1 3.5 1.2 2.9:1 Qatar 37.0 0.5 74.0:1 13.4 2.3 5.8:1 Somalia–Somaliland NA NA NA 4.9 4.5 1.1:1 Sudan 23.5 1.5 15.7:1 10.2 2.1 4.9:1 Syrian Arab Republic 44.3 5.7 7.8:1 19.1 5.9 3.2:1 Tunisia 49.5 2.4 20.6:1 15.1 1.6 9.4:1 United Arab Emirates 17.3 1.3 13.3:1 12.1 3.6 3.4:1

aSource: [9]. GYTS = Global Youth Tobacco Surveys; NA = not available.

of their plans for monitoring the WHO feedback to evaluate and improve Na- implementation and evaluation of ef- Framework Convention on Tobacco tional Tobacco Control Action Plans fective tobacco control programmes Control (WHO FCTC) [10]; all of and to develop plans where none exists. in all WHO Member States [11]. The these countries have ratified the WHO GYTS can help measure many of the FCTC. Monitoring the WHO FCTC articles in the WHO FCTC. The WHO The WHO FCTC, adopted by the FCTC calls for countries to use consist- Monitoring tobacco control 56th World Health Assembly in May ent methods and procedures in their efforts 2003, is the world’s first public health surveillance efforts. The GYTS was de- National Tobacco Control Action treaty on tobacco control. It provides signed to provide consistent sampling Plans are developed by governments the driving force and blueprint for the procedures, core questionnaire items, to provide clear strategies for reducing global response to the pandemic of training in field procedures and analysis and controlling tobacco use. A compre- tobacco-induced death and disease. The of data across all survey sites. hensive tobacco control programme Convention embodies a coordinated, The WHO FCTC also requires generally includes: public counter-mar- effective and urgent action plan to curb countries to be able to monitor the keting campaigns to reduce the effects tobacco consumption, laying out cost- treaty’s application. The GYTS helps of tobacco advertising, community- effective tobacco control strategies on each country establish applied research based programmes to reduce tobacco population-wide public policies, such in public health and contributes to es- use, cessation-assistance programmes, as bans on direct and indirect tobacco tablishing continuous tobacco control school-based programmes, enforce- advertising, tobacco taxes and price surveillance and monitoring. The WHO ment of existing tobacco restrictions, increases, smoke-free environments in FCTC also contributes to strengthening monitoring and evaluation of the all public places and workplaces, and the leadership capacity of the ministries control programme and related policy large, clear, graphic health messages of health and other state health bodies efforts to support the programme, such on tobacco packaging. In addition, the responsible for tobacco control, not as increased excise taxes, chronic disease Convention encourages countries to only in terms of public health advocacy, programmes targeting tobacco-related address cross-border issues, such as il- but also in negotiations with other sec- health problems and environmental legal trade and duty-free sales [10]. tors with respect to tobacco control. tobacco smoke restrictions. GYTS data The WHO FCTC and GYTS The GYTS also enhances the role of the can provide countries with valuable share the same goal: the development, nongovernmental sector by supporting

272 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

civil society participation in surveillance, EMR countries and the Gaza Strip/ Samy Eissa Al-Nasser (Kuwait), monitoring, and policy and programme West Bank reliable baseline data that Georges Saade (Lebanon), Ahmed development. can be used to develop, implement and Buni (Libyan Arab Jamahiriya), Moain evaluate their comprehensive tobacco El Kariri (Gaza Strip/West Bank), control programmes. Abdelkhalek Moujarrade (Morocco), Conclusion Salah Al Muzahmi (Oman), Hamda Qotba (Qatar), Ali Sheikh Omaer Tobacco use represents the single Acknowledgement Kabil (Somalia), Ibrahim Ginawi (Su- greatest preventable cause of death dan), Bassam Abu Al Zahab (Syrian worldwide. Every year nearly 5 mil- The authors appreciate the contribu- Arab Republic), Radhouane Fakhfakh lion people die from tobacco-related tion of each of the following Country (Tunisia), Ayesha Al-Mutawa (United illnesses, and this number is expected GYTS research coordinators: Ehab Arab Emirates). The high quality and to more than double by 2030 [12]. Makram (Egypt), Mohamed Mehrez reliability of the GYTS data are a tribute The findings from this study offer 13 (Egypt), Iman Al Jaghbeer (Jordan), to their leadership and direction.

References

1. Jha P, Chaloupka FJ. Tobacco Control in Developing Countries. 7. Tobacco industry targeting of women and girls. Washington Oxford, Oxford University Press, 2000. DC, Campaign for Tobacco-Free Kids, 2007 (Fact sheet, May 2. Warren CW et al. Global Youth Tobacco Surveillance – 2000– 7). 2007. Centers for Disease Control and Prevention, Surveil- 8. Waldron I et al. Gender differences in tobacco use in Africa, lance Summaries, 2008. MMWR, 2008, 57(no. SS-1). Asia, the Pacific and Latin America.Social science and medicine, 3. Pierce JP et al. Validation of susceptibility as a predictor of 1988, 27:1269–75. which adolescents take up smoking in the United States. Health 9. Mackay J, Eriksen M, Shafey O. The tobacco atlas, 2nd ed. At- psychology, 1996, 15(5):355–61. lanta, American Cancer Society, 2006. 4. Shah BV, Barnwell BG, Bieler GS. Software for the statistical 10. WHO Framework Convention on Tobacco Control. Geneva, analysis of correlated data (SUDAAN): User’s Manual. Release World Health Organization, 2003. 7.5. 1997 (software documentation). Research Triangle Park, 11. Global Tobacco Surveillance System Collaborating Group. NC, Research Triangle Institute, 1997. Global Tobacco Surveillance system (GTSS): purpose, pro- 5. Pierce J et al. Tobacco industry promotion of cigarettes and duction, and potential. Journal of school health, 2005, 75(1):15– adolescent smoking. Journal of the American Medical Associa- 24. tion, 1998, 279(7):505–11. 12. Peto R, Lopez AD. Future worldwide health effects of current 6. Hochberg A. Critics fume over marketing of “Camel No. 9”. smoking patterns. In: Koop CE, Pearson CE, Schwarz MR, NPR, March 16, 2007 (http://www.npr.org/templates/story/ eds. Critical issues in global health. San Francisco, Jossey–Bass; story.php?storyId=8909745, accessed 21 March 2009). 2001.

273 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Assessing validity of the adapted Arabic Paediatric Asthma Quality of Life Questionnaire among Egyptian children with asthma R. Abdel Hai,1 E. Taher 1 and M. Abdel Fattah2

تقيـيم مصداقية استبيان جودة حياة مرىض الربو لدى األطفال َّاملعدل باللغة العربية يف أطفال مرصيني مصابني بالربو رحاب عبد احلي، إيامن طاهر، حممد عبد الفتاح

يرتكز اخلالصـة:تدبري الربو يف الوقت احلايل عىل إتاحة أفضل جلودة احلياة لدى األطفال. وقد ّقيم الباحثون يف هذه الدراسة استبيان جودة حياة مرىض الربو لدى األطفال َّاملعدل باللغة العربية يف أطفال مرصيـني، واشتمل التقييم 103من األطفال املصابني بالربو ممن تتـراوح أعامرهم بني 8 و16 ًعاما. وقد كانت املصداقية التميـيزية للمقاييس الوسطية أكثر بمقدار ُي ْع َت ُّد به ًإحصائيالدى املصابني بدرجة خفيفة من الربو مما لدى املصابني بدرجة متوسطة أو شديدة منه. وكان ارتباط املصداقية للمجاالت سلبية مع مقاييس الشدة الرسيرية )اإلكلينيكية(. كام َّقيم الباحثون املوثوقية والتوافق الداخيل باستخدام معامل ألفا كرونباخ )α = 0.84(.أما قابلية اإلنتاج املتكرر )للنتائج( واالستجابية فكانتا مرتفعتني لدى كل من املرىض املصابني بالربو املستقر وغري املستقر. واستنتج الباحثون أن استبيان جودة حياة مرىض الربو لدى األطفال َّاملعدل باللغة العربية أداة تتمتع باملصداقية واملوثوقية لتقيـيم جودة احلياة يف األطفال املرصيـني املصابني بالربو.

ABSTRACT The recent focus in asthma management is rendering children a better quality of life (QOL). Validity and reliability of an adapted Arabic translation of the Paediatric Asthma Quality of Life Questionnaire (PAQLQ-A) among Egyptians was assessed in a cohort of 103 asthmatic children aged 8–16 years. Discriminative validity of mean scores was significantly higher among mild asthmatics than those with moderate/severe asthma. Construct validity of domains was significantly negatively correlated with clinical severity score. Reliability and internal consistency were assessed using Cronbach alpha coefficient α( = 0.84). Reproducibility and responsiveness were high among both stable and unstable asthma patients. PAQLQ-A is valid and reliable for assessing QOL among Egyptian asthmatic children.

Évaluation de la validité de l’adaptation en arabe du questionnaire sur la qualité de vie des enfants asthmatiques pour les enfants égyptiens asthmatiques

RÉSUMÉ Le récent intérêt accordé à la gestion de l’asthme permet d’améliorer la qualité de vie des enfants. La validité et la fiabilité d’une traduction en arabe adaptée du questionnaire sur la qualité de vie des enfants asthmatiques (PAQLQ-A) pour les enfants égyptiens a été évaluée dans une cohorte de 103 enfants asthmatiques âgés de 8 à 16 ans. La validité discriminante des notes moyennes était nettement plus élevée chez les enfants légèrement asthmatiques que chez ceux souffrant d’asthme modéré ou sévère. Une nette corrélation négative a été observée entre la validité de construit des domaines et le score de sévérité clinique. La fiabilité et la cohérence interne ont été évaluées à l’aide du coefficient alpha de Cronbach α( = 0,84). La reproductibilité et la réactivité étaient fortes à la fois chez les patients asthmatiques stables et instables. Le PAQLQ-A est valable et fiable pour évaluer la qualité de vie des enfants égyptiens asthmatiques.

1Department of Public Health and Community Medicine; 2Department of Paediatrics, Faculty of Medicine, University of Cairo, Cairo, Egypt (Correspondence to R. Abdel Hai: [email protected]). Received: 24/11/07; accepted: 03/02/08

274

املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Introduction Methods Faculty of Medicine, University of Cai- ro. Data confidentiality was preserved Globally, bronchial asthma is a major Study setting and design throughout the study in accordance health problem especially among chil- This prospective cohort study was con- with the revised Helsinki declaration dren, due to the increasing prevalence ducted at the outpatient allergy clinic of of bioethics [17]. All children, as well and associated increase in its morbid- the New Children’s Hospital, University as their parents and/or guardians, were ity and mortality [1,2]. It has recently of Cairo, over a 9-month period from informed about the aims of the study. been recognized as the most common January to September 2006. Written informed consent was obtained cause of school absence, thus affecting from the caregivers of the participants Sample size and study children’s educational potential and who agreed to their child’s participation participants in the study. adversely affecting a child’s quality of Sample size calculation was conducted life [3]. The introduction of new and Data collection long-acting inhaled drugs has greatly using the Power and Sample Size software, version 2.1.31 [15], and were After inclusion, children were clini- improved the prognosis of asthma [4,5] cally evaluated and interviewed at 3 pre- and nowadays the focus of concern has based on the following inputs: power = 85%, significance level = 0.05, an equal defined dates: on entering the study, shifted towards improvements in chil- after 4 weeks and after another 4 weeks. dren’s quality of life (QOL) [6]. proportion of children with mild and moderate/severe asthma, an ability to The child’s clinical condition was as- QOL is essentially viewed as a sub- detect difference of mean scores be- sessed using a clinical severity score jective, multidimensional experience tween these 2 groups of children = 0.2 (CSS) that Juniper et al. had utilized involving summary evaluations of the and standard deviation = 0.5. A sample for the validation of the original English positive and negative attributes charac- size of 113 participants was found to PAQLQ [10]. Each item of the CSS terizing one’s life [7]. In children, QOL fulfil these inputs. was assigned 1 point, and according to used to be based on the conventional the total sum of scores (maximum = Over a 6-month period, we enrolled assessment of asthma severity, the pres- 6), participant children were clinically a total of 110 children and adolescents ence and intensity of symptoms, the classified as “mild” (score ≤ 2) or “mod- aged between 8 and 16 years, who were need for medication, pulmonary func- erate/severe” asthma (score > 2). Chil- registered and having regular follow-up tion testing and discussions with parents dren were assigned as “stable asthma” if at the clinic and who fulfilled the study or caregivers [8]. Conversely, evidence they remained in the same CSS group inclusion criteria. Of these, 7 were lost has shown that clinical parameters have throughout the 3 visits or “unstable to follow-up and 103 continued until asthma” if they changed groups, turning a weak correlation with the child’s emo- the end of the study (31 females and tional aspects and with his/her daily from mild into moderate/severe or vice 72 males), giving a participation rate of versa over the 3 visits. functions and expectations [9]. 93.6% and study power of 82.0%. The Paediatric Asthma Quality of The translated PAQLQ consists of The inclusion criteria were patients 23 questions divided over 3 domains: Life Questionnaire (PAQLQ) was one with intermittent or persistent asthma activity limitation (5 questions), symp- of the first questionnaires to be used in as defined by the Global Initiative for toms (10 questions) and emotional the assessment of QOL of asthmatic Asthma guidelines [16]. All patients function (8 questions). It uses a 7-point children. Juniper et al. developed the matching these definitions who visited Likert scale with higher scores indicat- questionnaire, which was validated and the clinic during the study period were ing better health-related quality of life published originally in English, in 1996 included. Exclusion criteria for children (HRQOL) perception. The maximum [10], and then translated into other were: age below 8 years or above 16 possible score for each item is 7 (good languages [11–14]. To the authors’ years; concomitant illnesses other than HRQOL) and the minimum score is knowledge, the Arabic version has not asthma that could affect QOL; suspi- 1 (poor HRQOL). The contribution previously been assessed for validity cion of having an alternative cause for of each item in the questionnaire to the and reliability in Egypt. The aim of the recurrent wheezing other than asthma; 3 domains of activity limitation, symp- present study was to assess the valid- illiterate; or refusal to participate. toms and emotional function was evalu- ity and reliability of an adapted Arabic ated by calculating the mean response to translation of PAQLQ, in order to utilize Ethical considerations all its items for each domain. In addition, this questionnaire in assessing the QOL The study design and methodology was the overall score was calculated as being in Egyptian children and adolescents approved by the scientific research com- the average of the means of the 3 do- with asthma. mittee of the Department of Paediatrics, mains. The children themselves selected

275 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

the first 3 distressing items from a list of measuring the variation in the question- and a predominance of males (69.9%). day-to-day practice items. naire’s mean domain scores at the 2nd Almost three-quarters (72.5%) were Before the start of the study, the and 3rd visits by the same observer); urban residents and the rest (27.5%) questionnaire was pre-tested in a pilot and internal consistency (measured by were rural residents. At initial recruit- phase to 20 patients (not included in correlation of items within the same do- ment 56% of children suffered from the final sample) with intermittent or main and with the overall questionnaire moderate to severe asthma, with no sta- persistent asthma, aged 8–16 years, score using Cronbach alpha coefficient, tistically significant differences detected mean age 10.9 (SD 2.3) years, 12 of which was set at an acceptable range of between males and females. Initially, whom were males. After this assessment, 0.70 to 0.84 [22]). CSS was evaluated with an overall mean 5 items in the original Arabic PAQLQ score of 2.7 (SD 0.9) among all patients. Statistical analysis were removed as they did not match Stratification by sex revealed a CSS of with the Egyptian culture (baseball, A pre-designed SPSS, version 11.01 file 2.7 (SD 0.9) in males and 2.8 (SD 0.9) roller-boarding, skateboarding, water- was used for data entry and analysis. in females with no significant statistical skating and sand-skating). The other The following tests were used to test ad- difference (Table 1). items were maintained according to the aptation and validation of the PAQLQ: Before applying the PAQLQ-A, translation. The new adapted question- chi-squared test, Student t-test, paired patients were asked to choose the first naire was termed the Paediatric Asthma t-test, analysis of variance, Spearman 3 most distressing items from a list of Quality of Life Questionnaire–adapted correlation and Cronbach alpha reli- day-to-day practice items. The most fre- (PAQLQ-A) and the same scoring ability coefficient, with 95% confidence quently chosen were: sleeping (28.2%), system (1–7) was maintained as that intervals (95% CI) and significant P- playing football (16.5%) and playing used in the original questionnaire. In all values of < 0.05. 3 visits, and after assessing the children with friends (11.7%). The initial assess- with the CSS, they were subjected to the ment of QOL showed the mean score interviewer-administered version of the Results of the symptoms domain was 3.7 (SD Arabic translation of the PAQLQ-A. 0.6), of the emotions domain was 3.9 The age of the study participants ranged (SD 0.6) and of the activities domain Assessing PAQLQ-A properties from 8 to 16 years, with a mean of 11.3 was 4.5 (SD 0.8), while the overall score The PAQLQ-A was assessed for its [standard deviation (SD) 1.7] years was 4.0 (SD 0.5) with no statistically validity and reliability [14,18–21]. The face validity of the PAQLQ-A was previously established during the process of Table 1 Baseline characteristics of asthma patients and clinical severity score and scores on the Paediatric Asthma Quality of Life Questionnaire–adapted developing the questionnaire through a (PAQLQ-A) domains at the 1st visit full literature review, expert opinion and Baseline characteristic Males Females Total P-value patient inputs [18]. The content validity No. (%) No. (%) No. (%) was established through the process of Sex 72 (69.9) 31 (30.1) 103 (100.0) – pre-testing and item reduction during Age group (years) 0.351a the pilot phase. Discriminative validity 19 (26.4) 11 (35.5) 30 (29.1) was assessed by testing the significant < 11 difference between mean scores for ≥ 11 53 (73.6) 20 (64.5) 73 (70.9) a patients with different disease severities Severity of asthma 0.814 (mild, and moderate/severe). To test Mild 32 (44.4) 13 (41.9) 45 (43.7) construct validity the different domains Moderate/severe 40 (55.6) 18 (58.1) 58 (56.3) of the PAQLQ-A were compared with Mean score Mean score Mean score (SD) (SD) (SD) the CSS to determine if it behaved as Clinical severity and predicted, i.e. the higher the CSS the PAQLQ-A domains 2.7 (0.9) 2.8 (0.9) 2.7 (0.9) 0.845b lower the PAQLQ-A score. Clinical severity, 1st visit The test reliability measures included Symptoms, 1st visit 3.7 (0.5) 3.8 (0.6) 3.7 (0.6) 0.812b reproducibility (evaluated in the stable Emotions, 1st visit 3.8 (0.6) 4.0 (0.7) 3.9 (0.6) 0.781b asthma group at the 2nd and 3rd visits Activities, 1st visit 4.5 (0.8) 4.4 (0.8) 4.5 (0.8) 0.607b by the same observer using Spearman Overall, 1st visit 4.0 (0.6) 4.1 (0.5) 4.0 (0.5) 0.298b

correlation testing); responsiveness aChi-squared test; bStudent t-test. (assessed among unstable patients by SD = standard deviation.

276 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

significant difference detected between Table 2 Comparison of clinical severity score and the Paediatric Asthma Quality males and females (Table 1). of Life Questionnaire–adapted (PAQLQ-A) domains among all patients during the follow-up period During follow-up, several patients Clinical severity and 1st visit 2nd visit 3rd visit P-valuea changed their asthma stability and PAQLQ-A domains Mean score Mean score Mean score were reclassified as stable or unstable. (SD) (SD) (SD) Thus, in the final assessment, 63 pa- Clinical severity 2.7 (0.9) 2.1 (0.8) 1.6 (0.6) 0.001 tients (61.2%) were classified as having Symptoms 3.7 (0.5) 4.3 (0.7) 4.9 (0.8) < 0.001 stable asthma, while 40 (38.8%) were Emotions 3.8 (0.6) 4.5 (0.6) 5.1 (0.8) 0.001 classified as unstable asthma, with no Activities 4.5 (0.8) 4.8 (0.8) 5.2 (0.8) < 0.001 statistical difference between males and Overall 4.0 (0.5) 4.6 (0.6) 5.1 (0.8) 0.001

females. Throughout follow-up, gradual aANOVA. improvement occurred so that the per- SD = standard deviation. centage of patients with mild asthma increased from 43.7% initially to 67.0% in the 2nd visit to reach 73.8% by the 2nd visit and –0.73, –0.66 and –0.73 at also assessed by Cronbach alpha for all 3rd visit. When assessing CSS, as well the 3rd visit respectively) (Table 4). patients, using correlations of different as all domains of the PAQLQ-A and When assessing reproducibility, items within its domain, yielding values the overall score, a gradual significant stable asthma patients were compared of 0.85 for symptoms, 0.83 for emo- improvement was detected over the 3 at 2 different times: the 2nd visit and tions and 0.72 for activities (P < 0.001) visits, with lower CSS and higher QOL 3rd visit with a 4-week interval between (Table 6). domain scores (Table 2). the 2 visits. A highly significant positive correlation was found in all domains Assessment of the psychometric Discussion properties of the PAQLQ-A was con- between the 2 visits: symptoms (r = ducted by evaluating its validity and 0.89), emotions (r = 0.79), activities (r Assessing only the clinical parameters reliability. Discriminant validity was as- = 0.88) and overall score (r = 0.87) (P of asthma provides just one aspect of its sessed during the 1st visit and a statisti- < 0.001). Assessment of responsiveness great impact on a child’s life. Growing cally significant difference was observed was done by comparing each domain interest in the impact of asthma on day- between patients with mild asthma and score at 2 different times (2nd and 3rd to-day function is leading investigators those with moderate to severe asthma. visits) among unstable asthma patients. to include asthma-specific HRQOL Those suffering from the mild type The results showed a statistically signifi- questionnaires in a broad range of showed higher mean QOL scores in cant improvement in all domains (P < clinical studies [7]. In an earlier study all domains, in comparison with their 0.001), and the same occurred with the conducted among Egyptian children counterparts with moderate to severe CSS (Table 5). with asthma, the burden of asthma on a asthma (Table 3). The reliability of the PAQLQ-A child’s QOL was estimated using total In terms of construct validity, the was assessed by the Cronbach alpha disabling days, and the study concluded different domains of the PAQLQ-A coefficient for all patients, as well as after that a child with asthma would suffer were compared with the CSS, and a stratification by stability, yielding val- approximately 3 weeks of disrupted significant negative correlation was ues of 0.84, 0.80 and 0.88 respectively sleep/year and 8 weeks of disrupted waking hours/year, in addition to observed with each domain. This was (Table 6). Internal consistency was especially evident for the symptoms and emotional function domains as well as Table 3 Comparison of Paediatric Asthma Quality of Life Questionnaire–adapted the overall score among stable patients (PAQLQ-A) domain scores at the 1st visit, according to disease severity (r = –0.65, –0.62 and–0.56 in the 2nd PAQLQ-A domains Mild asthma patients Moderate/severe P-value visit and –0.71, –0.79 and –0.69 at the asthma patients 3rd visit respectively). Among unstable Mean score (SD) Mean score (SD) patients a similar significant negative Symptoms, 1st visit 3.9 (0.6) 3.5 (0.5) < 0.001 correlation was also observed in all Emotions, 1st visit 4.2 (0.6) 3.6 (0.6) < 0.001 domains, that was evident for each of Activities, 1st visit 4.5 (0.9) 4.3 (0.7) 0.41 symptoms, emotions and overall score Overall, 1st visit 4.2 (0.5) 3.8 (0.5) < 0.001 (r = –0.55, –0.53, and–0.55 during the SD = standard deviation.

277 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Table 4 Correlation between clinical severity score and mean scores for each of strong significant correlation between the Paediatric Asthma Quality of Life Questionnaire–adapted (PAQLQ-A) domains the mean overall scores in these visits. at the 2nd and 3rd visits, according to stability of asthma The main purpose of a QOL scale is its Clinical severity and PAQLQ-A Stable asthma patients Unstable asthma domains patients ability to detect important changes in a patient’s QOL. Therefore, a “good” scale r P-value r P-value is one that is sensitive to change for the 2nd visit population under study [25]. Respon- Clinical severity × symptoms –0.65 < 0.001 –0.55 < 0.001 siveness was determined by studying Clinical severity × emotions –0.62 < 0.001 –0.53 < 0.001 unstable asthma patients. Comparison –0.23 < 0.001 –0.27 < 0.001 Clinical severity × activities of the mean domain scores in the 2nd –0.56 < 0.001 –0.55 < 0.001 Clinical severity × overall and 3rd visits separately showed statisti- 3rd visit cally significant differences, revealing Clinical severity × symptoms –0.71 < 0.001 –0.73 < 0.001 that in unstable asthma patients, when Clinical severity × emotions –0.78 < 0.001 –0.66 < 0.001 the clinical status of the patient changed Clinical severity × activities –0.31 < 0.001 –0.54 < 0.001 (improved or worsened), the question- Clinical severity × overall –0.69 < 0.001 –0.73 < 0.001 naire responded accordingly. Therefore r = correlation coefficient. the PAQLQ-A was sensitive to small but clinically significant changes in the health status and the child’s daily QOL. possible maximum hospitalization and significant negative correlation between These findings agree with other stud- emergency room visits of 2.5 months/ CSS and the means of the 3 question- ies that assessed these aspects of the year and 195 days/year respectively naire domains as well as the overall PAQLQ [14,24]. [23]: hence the importance of measur- score in both stable and unstable asth- The reliability and internal consist- ing QOL. ma groups at 2 different times. These ency of an instrument is accepted at a results were comparable to the results The PAQLQ is one of the most Cronbach alpha coefficient value above of other studies which reported that widely used instruments for measuring 0.7. Concerning the PAQLQ-A reliabil- the construct validity of their adapted HRQOL in children with asthma and ity, an alpha value above 0.8 was found questionnaire was confirmed in both has been validated in many countries when assessing all patients and when cross-sectional and longitudinal studies around the world [11–14,24]. When assessing stable and unstable asthma pa- by demonstrating a strong correlations testing the validity and reliability of the tients separately. Moreover, the internal between various PAQLQ-A domains adapted Arabic version of PAQLQ, consistency was established regarding with clinical asthma parameters (asth- the questionnaire displayed a signifi- each individual domain. These results ma diary, beta-agonist use and peak cant discriminant validity in both the are comparable to those of Reichen- expiratory flow rate) 14,24[ ]. symptoms and activities domains, with berg and Broberg [11], who applied the the exception of its discriminative abil- The reproducibility of the PAQLQ to Swedish children and found ity in the emotions domain, which was PAQLQ-A was assessed by applying the a Cronbach alpha coefficient of 0.92 weak. Moreover, the questionnaire had questionnaire to stable asthma patients for the overall mean score, 0.86 for the high construct validity with a strong at the 2nd and 3rd visits. There was a symptoms domain, 0.84 for the emotions domain and 0.79 for the activities domain. The results are also comparable Table 5 Comparison of clinical severity score and different domains of the Paediatric Asthma Quality of Life Questionnaire–adapted (PAQLQ-A) among to those of the Spanish study that found patients with unstable asthma at the 2nd and 3rd visits the alpha coefficient ranged between Clinical severity and 1st visit 2nd visit 3rd visit P-value 0.88 and 0.96 for stable and unstable PAQLQ-A domains Mean score Mean score Mean score asthma patients respectively [12]. (SD) (SD) (SD) Assessing HRQOL as a proxy meas- Clinical severity 2.2 (0.7) 1.6 (0.9) – < 0.001 ure of clinical outcome considers the Symptoms – 4.4 (0.7) 5.1 (0.9) < 0.001 impact of the condition and its treat- Emotions – 4.7 (0.7) 5.1 (0.9) < 0.001 ment on a child’s physical, emotional Activities – 4.5 (0.6) 5.1 (0.8) < 0.001 and social functioning. It provides a Overall – 4.3 (0.7) 5.0 (0.8) < 0.001 more subjective patient-led baseline SD = standard deviation. against which effects of interventions

278 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 6 Reliability coefficients and Cronbach alpha values of the Paediatric One drawback of the questionnaire Asthma Quality of Life Questionnaire–adapted (PAQLQ-A) among the patients arises when it is applied to children 8 with stable and unstable asthma and among the 3 PAQLQ-A domains or 9 years old, since children at this age Variable Single measure Average measure α-value interclass interclass sometimes are not able understand the difference between certain terms, such Coefficient (95% CI) Coefficient (95% CI) Type of asthma as “mildly”, “moderately” and “severely”. Stable 0.51* (0.38–0.63) 0.80* (0.71–0.87) 0.80 In this case in our study, these terms Unstable 0.64* (0.56–0.77) 0.88* (0.80–0.93) 0.88 were explained by the interviewer using words in Egyptian dialect in order All patients 0.57* (0.48–0.66) 0.84* (0.78–0.89) 0.84 to avoid information bias during data PAQLQ-A domains collection. Symptoms 0.34* (0.27–0.42) 0.85* (0.80–0.89) 0.85 Emotions 0.35* (0.28–0.44) 0.83* (0.77–0.87) 0.83 Activities 0.30* (0.22–0.40) 0.72* (0.63–0.80) 0.72 Conclusion *P < 0.001. CI = confidence interval. This study has confirmed the PAQLQ-A validity and reliability as can be evaluated. This can only be Thus it can be used to evaluate the out- a multidimensional asthma-specific achieved if the measurement tool is valid, come effects of treatment. and age-appropriate instrument for as- reliable, disease-specific and responsive Another positive aspects of the sessing the HRQOL of Egyptian asth- to change [26]. The PAQLQ-A in this PAQLQ-A is that it is relatively quick matic children and adolescents. The study has been validated among Egyp- and easy to use. It took from 10 to 20 PAQLQ-A did not modify the proper- tian asthmatic children from urban and minutes to be applied, depending on ties of the original questionnaire. The rural areas with an approximately equal the child’s age, intellectual capacity and adaptation of activities did not interfere distribution of disease severity. It was education level (it was noted that the with the application and properties of found to be able to discriminate between first interview usually was longer and the original PAQLQ. We recommend levels of disease severity and to inform on the last one shorter). This is consistent its introduction and routine application minimal, yet important, clinical changes with reports on the original and other when assessing QOL among Egyptian and how they affected the child’s QOL. adapted questionnaires [10–14]. children with asthma.

References

1. Lee TA et al. Agreement between caregiver reported health- generic core scales. Health and quality of life outcomes, 2005, care utilization and administrative data for children with 1(3):22. asthma. Journal of asthma, 2007, 44(3):189–94. 10. Juniper EF et al. Measuring quality of life in children with 2. Rance K, Trent SA. Profile of a primary care practice asthma asthma. Quality of life research, 1996, 5:35–46. program. Journal of pediatric health care, 2005, 19(1):25–32. 11. Reichenberg K, Broberg AG. Quality of life in childhood asth- 3. Ramos-Barbón D. Asma [Asthma]. Archivos de bronconeu- ma: use of the Paediatric Asthma Quality of Life Questionnaire mología, 2007, 43(Suppl. 2):3–14. in a Swedish sample of children 7 to 9 years old. Acta paediat- rica, 2000, 89:989–95. 4. Baena-Cagnani CE et al. New perspectives in the treatment of allergic rhinitis and asthma in children. Current opinion in al- 12. Badia X et al. Validacion de la version espanola del Pediatric lergy and clinical immunology, 2007, 7(2):201–6. Quality of Life Questionnaire en la valoracion de la calidad de vida del nino asmatico [Validity of the Spanish version 5. Blaiss MS, Hill B. Outcomes in pediatric asthma. Current allergy of the Pediatric Quality of Life Questionnaire for evaluating and asthma reports, 2005, 5(6):431–6. quality of life in asthmatic children]. Medicina clínica, 2001, 6. Baiardini I et al. Allergic diseases and their impact on qual- 116(15):565–72. ity of life. Annals of allergy, asthma and immunology, 2006, 13. Clarke E et al. Pediatric asthma quality of life questionnaire: 97(4):419–28. validation in children from Singapore. Asian Pacific journal of 7. Eiser C, Morse R. Quality of life measures in chronic diseases of allergy and immunology, 1999, 17(3):155–61. childhood. Health technology assessment, 2001, 5(4):1–7. 14. Raat H et al. Responsiveness, longitudinal- and cross-sectional 8. Van Gent R et al. Quality of life in children with undiagnosed construct validity of the Pediatric Asthma Quality of Life Ques- and diagnosed asthma. European journal of pediatrics, 2007, tionnaire (PAQLQ) in Dutch children with asthma. Quality of 166(8):843–8. life research, 2005, 14(1):265–72. 9. Upton P et al. Measurement properties of the UK-English ver- 15. PS: Power and Sample Size Calculation. Department of Bi- sion of the Pediatric Quality of Life Inventory 4.0 (PedsQL) ostatistics, Vanderbilt University [freeware] (http://biostat.

279 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

mc.vanderbilt.edu/twiki/bin/view/Main/PowerSampleSize, 21. French DJ, Christie MJ, Sowden AJ. The reproducibility of the accessed 31 August 2009). Childhood Asthma Questionnaires: measures of quality of life for children with asthma aged 4–16 years. Quality of life 16. GINA report. Global strategy for asthma management and pre- research, 1994, 3(3):215–24. vention. Revised 2006. Global Initiative for Asthma [online document] (http://www.ginasthma.org, accessed 12 August 22. Cronbach LJ. Coefficient alpha and the internal structure of 2009). tests. Psychometrika, 1951, 16(3):297–334. 17. World Medical Association Declaration of Helsinki. Ethical 23. Abdel Hai R. Bronchial asthma: a study among attendants of children’s hospital, Cairo University [MSc thesis]. Cairo, Egypt, principles for medical research involving human subjects. Revised Department of Public Health, Faculty of Medicine, Cairo Uni- by the 52nd WMA General Assembly, Edinburgh, Scotland, versity, 2000. October 2000 (http://www.wma.net/e/policy/b3.htm, accessed 12 August 2009). 24. Poachanukoon O et al. Pediatric Asthma Quality of Life Ques- tionnaire (PAQLQ): validation among asthmatic children in 18. Juniper EF. How important is quality of life in pediatric asthma? Thailand. Pediatric allergy and immunology, 2006, 17(3):207– Pediatric pulmonology. Supplement, 1997, 15:17–21. 12. 19. Oleson M. Content validity of the Quality of Life Index. Applied 25. Hyland ME. A brief guide to the selection of quality of life in- nursing research, 1990, 3(3):126–7. struments. Health and quality of life outcomes, 2003, 1:24. 20. Hays RD, Anderson RT, Revicki D. Assessing reliability and 26. Bowling A. Evaluating health services: multidisciplinary col- validity of measurement in clinical trials. In: Staquet MJ, Hays laboration. In: Research methods in health: investigating health RD, Fayers PM, eds. Quality of life assessment in clinical trials. and health services, 2nd ed. Buckingham, UK, Open University Oxford, Oxford University Press, 1998:169–82. Press, 2003.

Corrections Hepatitis B and C viral infection: prevalence, knowledge, attitude and practice among barbers and clients in Gharbia governorate, Egypt. S. Shalaby,1 I.A. Kabbash,1 G. El Saleet,1 N. Mansour,1 A. Omar1 and A. El Nawawy2. Eastern Mediterranean health journal, 2010, 16(1):10–17. The authors’ affiliations should read: 1Department of Public Health, Social and Preventive Medicine, University of Tanta, Tanta, Egypt (Corre- spondence to S. Shalaby: [email protected]). 2Department of Public Health and Community Medicine, Al-Azhar University, Cairo, Egypt.

Nosocomial infections in a neonatal intensive care unit in south-western Saudi Arabia. A.A. Mah- fouz, T.A. Al-Azraqi, F.I. Abbag, M.N. Al-Gamal, S.Seef and C.S. Bello. Eastern Mediterranean health journal, 2010, 16(1):40–4. In the Arabic abstract, the first sentence should read:

قاست دراسة أترابية وقوع عوامل االختطار لعدوى املستشفيات يف وحدة الرعاية املركزة للولدان يف مستشفى أهبا العام، باململكة العربية السعودية.

280 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Cigarette smoking, hypertension and diabetes mellitus as risk factors for erectile dysfunction in upper Egypt H. Zedan,1 A.A. Hareadei,1 A.A. Abd-Elsayed 2 and E.M. Abdel-Maguid 1

تدخني السجاير وارتفاع ضغط الدم والسكري كعوامل اختطار خللل وظيفة االنتصاب يف صعيد مرص حاتم زيدان، أمحد هريدي، عالء عبد السيد، إنصاف عبد املجيد

يؤثرخلل اخلالصـة:االنتصاب عىل جودة احلياة لدى ماليني الرجال يف مجيع أرجاء العامل. وهتدف هذه الدراسة إىل ُّالتعرف عىل العالقة بني عوامل اختطار منتقاة وبني خلل وظيفة االنتصاب لدى الرجال يف صعيد مرص. وقد أجرى الباحثون املسح عىل املرىض مستخدمني املنسب الدويل لوظائف االنتصاب ذي البنود اخلمسة، مع تقييم وجود ارتفاع ضغط الدم والسكري والتدخني. وشملت الدراسة 658 ًرجال يعاين من خلل وظيفة االنتصاب، ووجد الباحثون أن 17.3% منهم مصابون بارتفاع ضغط الدم وأن 21.4% منهم مصابون بالسكري وأن 40.1% مدخنون، كام شملت الدراسة 821 ًرجال من نفس عمر املرىض اعتربوا شواهد لعدم معاناهتم من خلل وظيفة االنتصاب، ووجد الباحثون أن 2.8% منهم مصابون بارتفاع ضغط الدم وأن 3.7% منهم مصابون بالسكري وأن 28.7% مدخنون. واتضح للباحثني من التحليل املتعدد املتغريات أن ً كالمن ارتفاع ضغط الدم )نسبة األرجحية 3.1 5.4 5.4 ( والسكري )نسبة األرجحية ( والتدخني )نسبة األرجحية ( من عوامل االختطار التي ُي ْع َت ُّد هبا يف خلل وظيفة االنتصاب.

ABSTRACT Erectile dysfunction impairs the quality of life of millions of men worldwide. This study aimed to determine the relationship between selected clinical risk factors and erectile dysfunction in men residing in upper Egypt. Patients were surveyed with the 5-item International Index of Erectile Function (IIEF-5) and assessed for the presence of hypertension, diabetes and smoking. Of 658 men with erectile dysfunction, 17.3% had hypertension, 21.4% had diabetes and 40.1% were smokers, whereas among 821 age-matched controls without erectile dysfunction, the corresponding figures were 2.8%, 3.7% and 28.7%. Multivariate analysis showed that hypertension (OR = 5.4), diabetes mellitus (OR = 5.4) and smoking (OR = 3.1) were significant risk factors for erectile dysfunction.

Tabagisme, hypertension et diabète comme facteurs de risque des troubles de l’érection en haute Égypte

RÉSUMÉ Les troubles de l’érection perturbent la qualité de vie de millions d’hommes dans le monde. Cette étude visait à définir la relation entre certains facteurs de risque cliniques et les troubles de l’érection chez des hommes vivant en haute Égypte. L’étude était basée sur l’Index International de la Fonction Érectile contenant 5 items (5-item International Index of Erectile Function ou IIEF-5) et les patients étaient contrôlés au niveau de l’hypertension, du diabète et du tabagisme. Sur 658 hommes présentant des troubles de l’érection, 17,3 % souffraient d’hypertension, 21,4 % étaient diabétiques et 40,1 % étaient des fumeurs, alors que parmi les 821 hommes du groupe témoin appariés selon l’âge ne souffrant pas de troubles de l’érection les chiffres correspondants étaient 2,8 %, 3,7 % et 28,7 %. L’analyse multivariée a montré que l’hypertension (risque relatif rapproché = 5,4), le diabète (risque relatif rapproché = 5,4) et le tabagisme (risque relatif rapproché = 3,1) étaient des facteurs de risque importants pour les troubles de l’érection.

1Department of Dermatology and Andrology; 2Department of Public Health and Biostatistics, Faculty of Medicine, University of Assiut, Assiut, Egypt (Correspondence to A.A. Abd-Elsayed: [email protected]). Received: 29/10/07; accepted: 13/02/08

281 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Introduction aimed to determine the relationship be- 21–25); mild (score 18–21); mild to tween cigarette smoking, hypertension moderate (score 12–17); moderate Erectile dysfunction (ED) is defined as and diabetes mellitus and ED in men (score 8–11); or severe (score 5–7). the inability to achieve or maintain an residing in Upper Egypt. The inclusion criteria for cases were: erection sufficient to permit satisfactory presented during the period of the sexual intercourse [1]. It is estimated Methods study; resident in Upper Egypt; > 20 that ED affects more than 150 million years old; diagnosed as having ED by men worldwide, and impairs psycho- Study design history and IIEF-5 score ≤ 21; free of logical well-being and personal relation- major physical or psychological disabili- ships and hence quality of life [2]. The This study was an analytical hospital- reported frequencies of ED vary sub- based case–control study comparing ties; and not receiving any psychiatric stantially between different countries cases of ED with a control group medication. The inclusion criteria for and populations, possibly due to differ- matched for age and residence (gover- controls were: matched with cases for ences in culture, diagnostic criteria, age norate and urban or rural). A pilot study age and residence (governorate and distribution and concomitant medical was carried out on 20 cases to identify urban or rural); diagnosed to be free conditions, as well as methodological potential problems before starting the from ED by history and IIEF-5 score differences 3[ –5]. The Massachusetts main study. > 21; free of major physical or psycho- Male Aging Study indicated that the Sample logical disabilities; and not receiving any prevalence of ED was 52% in men aged psychiatric medication. A total of 658 male patients aged over 40–70 years old and was strongly cor- 20 years complaining of ED were The presence or absence of the fol- related with age [5,6]. included in this study. They were all lowing risk factors—smoking, hyper- Common risk factors associated patients who attended the andrology tension and diabetes mellitus (type 1 or with ED include heart disease, hyper- clinic of University of Assiut Hospital, 2)—were determined in both patients tension, diabetes mellitus and cigarette Assiut, Egypt in the years 2003 to 2007. and controls. A man was considered a smoking [7–23]. Prolonged elevated This hospital is the largest tertiary care smoker if he had smoked > 1 cigarette/ blood pressure has detrimental effects centre in Upper Egypt, serving patients day for at least 1 year and a nonsmoker on the vascular system as a whole, in- from Beni Suef in the north to Aswan in if he had never smoked more than 1 cluding the penile blood supply [13,14]. the south. In addition, 821 control men It is now widely accepted that organic cigarette/day 22[ ]. Heavy smoking was were included in this study who were defined as tobacco consumption > 20 ED in a substantial majority of men normal patients visiting the hospital and cigarettes per day; moderate smoking occurs because of underlying vascular not complaining of ED. causes, especially atherosclerosis [15]. as 10–20 cigarettes per day; and mild All antihypertensive agents, regardless Data collection smoking as < 10 cigarettes per day. A of composition, have been implicated After providing informed, witnessed man was defined as having hyperten- in causing ED [16,17]. ED is also a com- verbal consent, all patients and controls sion if he had a systolic pressure > 160 mon complication of diabetes mellitus were surveyed with 2 instruments: the mmHg and/or a diastolic pressure > [19–21], and ED is 2–4 times more International Index of Erectile Func- 90 mmHg and/or was currently taking prevalent among diabetic patients than tion, 5-item version (IIEF-5) [24] and a drugs for hypertension [23]. A man was nondiabetic individuals [18] with the physician medical data collection form. defined as having diabetes mellitus if reported prevalence ranging between The IIEF-5 is internationally recognized he was currently taking drugs for this 2% [7] and 75% [19]. Several studies as a reliable, valid and sensitive evalua- condition and/or he had fasting blood have demonstrated that cigarette smok- tion instrument for assessing ED. It is a glucose ≥ 7.0 mmol/L. ing is associated with ED [8,10,11]. 5-point scale (score range 5–25) to rate The physician’s data collection form Smokers are 1.5 times more likely to a man’s confidence in achieving and have ED than nonsmokers [14] and keeping an erection; the frequency that consisted of history-taking, clinical ex- even 2 cigarettes per day may have a his erections are hard enough for pen- amination and investigation, e.g. level deleterious effect on erection 12[ ]. etration; his difficulty in maintaining an of smoking (mild, moderate or heavy) In Egypt, a greater understanding erection to completion of intercourse; and the presence, nature and treatment of the frequency and risk factors for ED and the frequency that attempted sexual of hypertension and diabetes, measure- is required for further development of intercourse is satisfactory. ED severity ment of blood pressure and estimation therapeutic strategies for ED. This study was defined as: none (IIEF-5 score of blood glucose level.

282 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 1 Age distribution of men with erectile dysfunction and control men 141 (21.4%) had diabetes mellitus and Age group Cases (n = 658) Controls (n = 821) P-value 264 (40.1%) were smokers. Among (years) No. % No. % the 821 control men, 23 (2.8%) had 20–< 30 89 13.5 102 12.4 0.6 hypertension, 30 (3.7%) had diabetes 30–< 40 110 16.7 133 16.2 0.4 mellitus and 236 (28.7%) were smok- 40–< 50 234 35.6 291 35.4 0.5 ers. The differences in the prevalence 50–< 60 154 23.4 186 22.7 0.4 values for all 3 risk factors were statisti- 60–< 70 58 8.8 83 10.1 0.2 cally significant between the groups (all 70–80 13 2.0 26 3.2 0.1 P < 0.001) (Table 2). Among the studied ED patients, severe ED (IIEF-5 score 5–7) was recorded among 12.2% of hypertensive Table 2 Prevalence of hypertension, diabetes and smoking among men with erectile dysfunction and control men patients, 8.5% of diabetic patients and Risk factor Cases (n = 658) Controls (n = 821) P-value 7.2% of smokers (Table 3). No. % No. % Multivariate analysis showed that Hypertension 114 17.3 23 2.8 < 0.001 the risk of having ED was more than 5 Diabetes mellitus 141 21.4 30 3.7 < 0.001 times higher in men with hypertension Smoking 264 40.1 236 28.7 < 0.001 [odds ratio (OR) = 5.4, 95% confidence interval (CI): 3.4–8.5, P < 0.001] and diabetes mellitus (OR = 5.4, 95% CI: Ethical considerations the different age classes between both 3.6–8.2, P < 0.001), and around 3 times groups, Yates corrected chi-squared higher in smokers (OR = 3.1, 95% CI: Approval was obtained from our institu- 2.4–3.8, P < 0.001) (Table 4). tion ethical committee. The aim of the test, Mann–Whitney test and logistic study was explained to all participants. regression for analysis of risk factors. It was emphasized to them that they Discussion would not be harmed in the study and that refusal to participate would Results A number of modifiable risk factors are not affect their access to services and associated with ED, including cigarette The study included 658 patients com- treatment in the hospital. Consent was smoking, hypertension and diabetes plaining of ED and 821 control men obtained from all the participants and mellitus [25]. ED is a common problem not complaining of ED. The duration of security and confidentiality of all the in- among ageing men and the prevalence formation obtained was observed. The complaint varied from 1 month to 120 of ED increases with each decade of pa- IIEF-5 data were collected by personal months. The mean age of the cases was tient age [5]. In the present study, about interview with participants. 46.3(SD 23.5) years and of the controls 70% of patients were over age 40 years, was 44.5 (SD 20.2) years (P = 0.11). indicating that age is strongly associated Data analysis About 70% of the ED patients were over with ED. Feldman et al. analysed 1290 Data entry and analysis were done using the age of 40 years (Table 1); 59.0% men in the Massachusetts Male Aging SPSS, version 15. Data analysis included were 40–60 years old and 10.8% were Study and concluded that “age was the descriptive analysis with means and 60–80 years old. variable most strongly associated with standard deviation (SD), z test for com- Of the 658 ED patients, 114 impotence” [5]. Of the patients in the paring the percentage of participants in (17.3%) suffered from hypertension, current study 59.0% were in the age

Table 3 Prevalence of hypertension, diabetes and smoking by International Index of Erectile Function, 5-item version (IIEF-5) scores among men with erectile dysfunction (ED) Risk factor Severe ED Moderate ED Mild to moderate ED Mild ED (score 5–7) (score 8–11) (score 12–16) (score 17–21) No. % No. % No. % No. % Hypertension 14 12.2 41 36.0 49 43.0 10 8.8 Diabetes mellitus 12 8.5 59 41.8 60 42.6 10 7. 1 Smoking 19 7.2 88 33.3 124 47.0 33 12.5

283 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

group 40–60 years old compared with Table 4 Logistic regression analysis of risk factors for erectile dysfunction 10.8% in the group 60–80 years old, Risk factor OR (95% CI) P-value which is not the same in more devel- Hypertension 5.4 (3.4–8.5) < 0.001 oped countries [5,26]. This reflects dif- Diabetes mellitus 5.4 (3.6–8.2) < 0.001 ferences in social and cultural aspects in Smoking 3.1 (2.4–3.8) < 0.001

addition to different life expectancy rates OR = odds ratio; CI = confidence interval. in Upper Egypt compared to developed countries. It is known that men above the age of 60 years in Upper Egypt rarely the total number of cases compared patients had severe ED. The probability seek medical advice for ED probably with 2.8% of the controls. According of developing ED in men with diabetes because they consider this problem as a to IIEF-5 scores, 12.2% of hypertensive was 5 times higher than in nondiabetic normal consequence of ageing. patients had severe ED. A high preva- men. Previous studies demonstrated The association between smoking lence of ED (43.2%) was previously that ED was 2 to 4 times more prevalent and ED has been reported in numerous demonstrated in the Egyptian male among diabetic patients than nondia- studies that reported an increased inci- population with hypertension [31]. Ac- betic individuals [18]. dence of ED among smokers [5,9,27]. cording to Ponholzer et al. patients with Longitudinal data from the Mas- hypertension developed ED 2 times sachusetts Male Aging Study showed more than nonhypertensive patients Conclusion that the incidence of ED was twice as [24]. However, our study showed that high among smokers compared with the probability of developing ED in hy- Our study estimated the prevalence of nonsmokers over a 9-year period [28]. pertensive patients was 5 times higher smoking, hypertension and diabetes There is evidence of a beneficial effect of than in nonhypertensive patients. This mellitus among patients with ED in Up- smoking cessation on ED, especially at difference could be attributed to differ- per Egypt compared with controls. We younger ages [29]. In our study, smok- ences in study design, age distribution, found that the probability of developing ing was reported by 40.1% of patients duration of hypertension and the anti- ED was very high in the presence of complaining of ED compared with hypertensive drugs used by the study smoking, hypertension and diabetes 28.7% of controls. The probability of patients. mellitus. The results differ from those developing ED in smokers was 3 times ED is a common complication of in other populations and countries higher than nonsmokers, which is simi- diabetes, with a prevalence ranging be- possibly due to differences in culture, lar to the results of Shiri et al. [30] and tween 27% and 75% [19]. In our study, race, health services and methodology. higher than that reported by Kupelian 21.4% of the total number of ED pa- Larger-scale studies on a national basis et al. [25]. tients had diabetes mellitus compared are needed to survey a greater number Hypertension was detected in 114 with 3.7% of the controls. According of populations for all risk factors associ- patients who represented 17.3% of to the IIEF-5 scores, 8.5% of diabetic ated with ED.

References

1. NIH Consensus Conference. Impotence. NIH Consensus De- 7. Benet AE, Melman A. The epidemiology of erectile dysfunc- velopment Panel on Impotence. Journal of the American Medi- tion. Urologic clinics of North America, 1995, 22:699–709. cal Association, 1993, 270:83–90. 8. Martin-Morales A et al. Prevalence and independent risk fac- 2. Lue TF. Erectile dysfunction. New England journal of medicine, tors for erectile dysfunction in Spain: results of the Epidemiolo- 2000, 342:1802–13. gia de la Disfuncion Erectil Masculina Study. Journal of urology, 2001, 166:569–75. 3. Papatsoris AG, Triantafyllidis A, Gekas A. Prevalence of erectile 9. Bacon CG e et al. A prospective study of risk factors for erectile dysfunction in the European Union. Asian journal of andrology, dysfunction. Journal of urology, 2006, 176:217–21. 2003, 5:255. 10. Seyam RM et al. Prevalence of erectile dysfunction and its cor- 4. Nicolosi A et al. Epidemiology of erectile dysfunction in four relates in Egypt: a community based study. International journal countries: cross national study of the prevalence and corre- of impotence research, 2003, 15:237–45. lates of erectile dysfunction. Urology, 2003, 61:201–6. 11. McVary KT, Carrier S, Wessells H. Smoking and erectile dys- 5. Feldman HA et al. Impotence and its medical and psychosocial function: evidence based analysis. Journal of urology, 2001, correlates: Results of the Massachusetts Male Aging Study. 166:1624–32. Journal of urology, 1994, 151:54–61. 12. Gilbert DG, Hagen RL, D’Agostino JA. The effects of cigarette 6. Wessells H et al. Erectile dysfunction. Journal of urology, 2007, smoking on human sexual potency. Addictive behaviours, 1986, 177:1675–81. 11:431–4.

284 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

13. Ellis JL, Campos-Outcalt D. Cardiovascular disease risk factors 23. Fonseca V, Jawa A. Endothelial and erectile dysfunction, dia- in native Americans: a literature review. American journal of betes mellitus, and the metabolic syndrome: common path- preventive medicine, 1994, 10:295–307. ways and treatments? American journal of cardiology, 2005, 14. Hypertension control. A report of WHO Expert Commit- 96(Suppl.):13M–8M. tee. World Health Organization technical report series, 1996, 24. Rosen RC et al. Development and evaluation of an abridged, 862:1–83. 5-item version of the International Index of Erectile Function 15. Virag R, Bouilly P, Frydman D. Is impotence an arterial disor- (IIEF-5) as a diagnostic tool for erectile dysfunction. Interna- der? A study of arterial risk factors in 440 impotent men. Lancet, tional journal of impotence research, 1999, 11:319–26. 1985, 1:181–4. 25. Saigal CS et al. Predictors and prevalence of erectile dysfunc- 16. Goldstein I, Krane RJ. Drug induced sexual dysfunction. World tion in a racially diverse population. Archives of internal medi- journal of urology, 1983, 1:239–47. cine, 2006, 166:207–12. 17. Slag MF et al. Impotence in medical clinic outpatients. Journal 26. Ponholzer A et al. Prevalence and risk factors for erectile dys- of the American Medical Association, 1983, 249:1736–40. function in 2869 men using a validated questionnaire. Euro- 18. Rosen RC et al. The multinational Men’s Attitudes to Life Events pean urology, 2005, 47:80–6. and Sexuality (MALES) study. I. Prevalence of erectile dysfunc- 27. Kupelian V, Link CL, McKinlay JB. Association between smok- tion and related health concerns in the general population. ing, passive smoking, and erectile dysfunction: results from Current medical research and opinion, 2004, 20:607–17. the Boston Area Community Health (BACH) Survey. European 19. Fedele D et al. Erectile dysfunction in type I and type II diabetics urology, 2007, 52:416–22. in Italy. International journal of epidemiology, 2000, 29:524–31. 28. Feldman HA et al. Erectile dysfunction and coronary risk fac- 20. Sasaki H et al. Prevalence and risk factors for erectile dysfunc- tors: prospective results from the Massachusetts Male Aging tion in Japanese diabetics. Diabetes research and clinical prac- Study. Preventive medicine, 2000, 30:328–38. tice, 2005, 70:81–9. 29. Polsky JY et al. Smoking and other lifestyle factors in relation to 21. Romeo JH et al. Sexual function in men with diabetes type 2: erectile dysfunction. BJU international, 2005, 96:1355–9. association with glycemic control. Journal of urology, 2000, 163:788–91. 30. Shiri R et al. Smoking causes erectile dysfunction through vas- 22. Austoni E. Smoking as a risk factor for erectile dysfunction: data cular disease. Urology, 2006, 68:1318–22. from the andrology prevention weeks 2001–2002. A study 31. Mittawae B et al. Incidence of erectile dysfunction in 800 of the Italian Society of Andrology. European urology, 2005, hypertensive patients: a multicenter Egyptian national study. 48:810–8. Urology, 2006, 67:575–8.

Global Adult Tobacco Survey in Egypt

The Ministry of Health (MoH) of Egypt, the Central Agency for Public Mobilization and Statistics (CAPMAS) and the World Health Organization (WHO) released key findings from the Global Adult Tobacco Survey in Egypt on Thursday, 28 January 2010. The Global Adult Tobacco Survey, which was conducted in 2009, is a standardized global survey for systematically monitoring adult tobacco use (smoked and smokeless) and tracking key tobacco control indicators. The MoH, CAPMAS and the WHO Representative’s Office in Egypt are the three national partners who collaborated in this survey.

The survey was conducted in all of Egypt’s governorates, covering both males and females, aged 15 years and older, in a total sample size of 23 760. One of the key findings revealed by the survey is that 38% of Egyptian males use some form of tobacco product. Of this percentage, nearly 32% smoke cigarettes, about 6% smoke shisha and almost 5% chew tobacco.

Tobacco use constitutes a serious health burden on Egyptian society, and also on the health system, due to the cost of providing health care for smokers.

Further information about the Tobacco Free Initiative can be found at: http://www.emro.who.int/tfi/tfi.asp

285 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Behavioural and clinical factors associated with depression among individuals with diabetes Z. Yekta,1 R. Pourali 1 and R. Yavarian 2

العوامل السلوكية والرسيرية )اإلكلينيكية( املصاحبة لالكتئاب لدى املصابني بالسكري زهرا يكتا، رضا بور عيل، رويا ياوري ارتبط اخلالصـة:االكتئاب بمعدالت أعىل للوفيات وللمراضة لدى مرىض السكري، ورغم ذلك مل حتظ هذه القضية بدراسات كافية يف مجهورية إيران اإلسالمية. وقد أجرى الباحثون هذه الدراسة العرضية لوصف معدل انتشار االكتئاب لدى املراجعني لعيادة معاجلة السكري يف أوريميا، وللتعرف عىل العوامل الرسيرية )اإلكلينيكية( والسلوكية واالجتامعية والديموغرافية املصاحبة. وقد شملت الدراسة 295 ًمريضا، كان لدى 128 )43.4%( منهم أحراز االكتئاب تساوي أو تزيد عن وفق 15مرسد بيك لالكتئاب. وبلغ احلرز الوسطي جلميع املرىض 15.4 )واالنحراف املعياري (. 9.5وقد كان مرىض السكري املصابون باكتئاب أكرب ً سناوأقل ًتعليامبشكل واضح ممن ال يعانون من االكتئاب، كام كان لدهيم فرتة أطول من اإلصابة بالسكري، ومعاناة أكثر من مضاعفاته. وبالتحليل للتحوف اللوجستي وجد الباحثون أن التقدم يف العمر كان هو املتغري الوحيد املصاحب لالكتئاب ًتصاحبا ُي ْع َت ُّد به ًإحصائيا.

ABSTRACT Depression has been linked to greater mortality and morbidity in diabetic patients, but this issue has not been adequately studied in the Islamic Republic of Iran. This cross-sectional study described the prevalence of depression in patients attending a diabetes clinic in Urmia and determined the associated sociodemographic, behavioural and clinical factors. Of 295 patients, 128 (43.4%) had depression scores (≥ 15) on the Beck Depression Inventory. The mean score for all patients was 15.4 (SD 9. 5). Those with depression were significantly older and less educated than those without depression, had a longer duration of diabetes and were more likely to suffer complications. On logistic regression analysis, older age was the only variable significantly associated with depression.

Facteurs comportementaux et cliniques associés à la dépression chez les individus diabétiques

RÉSUMÉ Une relation a été établie entre une mortalité et une morbidité plus fortes chez les patients diabétiques, mais le sujet n’a pas été correctement étudié en République islamique d’Iran. Cette étude transversale décrivait la prévalence de la dépression chez les patients suivis dans une clinique du diabète à Urmia et définissait les facteurs socio-démographiques comportementaux et cliniques qui y sont associés. Sur 295 patients, 128 (43,4 %) avaient un score de dépression (≥ 15) sur l’Inventaire de la dépression de Beck. Le score moyen pour l’ensemble des patients était 15,4 (écart type 9,5). Les patients souffrant de dépression étaient nettement plus âgés et moins éduqués que ceux qui n’en souffraient pas, étaient diabétiques depuis plus longtemps et présentaient davantage de risques de complications. À l’analyse de régression logistique, l’âge était la seule variable significative associée à la dépression.

1Department of Community Medicine, Faculty of Medicine; 2Department of Psychology, Urmia University of Medical Sciences, Urmia, Islamic Republic of Iran (Correspondence to Z. Yekta: [email protected]). Received: 04/11/07; accepted: 17/02/08

286 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Introduction cause of disease burden worldwide after Data collection ischaemic heart disease [12]. A questionnaire was used to collect data Depression is recognized as an impor- In the Islamic Republic of Iran re- about clinical status: age at onset and tant co-morbidity in a number of chron- search on depression in diabetic patients duration of diabetes; complications of ic medical conditions such as diabetes is lacking, and depression frequently re- diabetes; treatment intensity; current [1]. Reports indicate that patients with mains unrecognized and undertreated. smoking (daily and occasional smok- diabetes are 1.5–2 times more likely The aim of this study was to describe the ers); and body mass index (BMI). to have depression than those without prevalence of depression in individuals Treatment intensity was classified into: [1]. A recent meta-analysis of 39 studies with diabetes in Urmia, Islamic Repub- insulin therapy (alone or combination) of patients with diabetes estimated the lic of Iran, and determine the associated or other therapy (oral agents, diet). BMI prevalence of major depression (based sociodemographic, behavioural and was divided into 2 categories: normal on psychiatric interviews) as 11% and clinical factors. < 25 kg/m2 or overweight ≥ 25 kg/m2. elevated depression symptoms (based Demographic data were also collected on depression rating scales) as 31% about: age, sex, marital status, educa- [2]. Depression contributes to poor Methods tional level and income level. Marital metabolic control, decreased quality of status was classified as unmarried (sin- life and increased medical morbidity This was a cross-sectional study con- gle/divorced/widowed) or married. and mortality in patients with diabetes ducted from February to August 2006. Participants were classified into low or [2–5], as well as significantly higher Sample middle/high income groups based on medical costs [6]. More encouragingly, total household income divided by the however, there is evidence that therapy A convenience sample was drawn from number of household members. This to treat depressive conditions is effective attendees at the Taleghani diabetes categorization was based on patients’ and improves the mood, functioning clinic in Urmia, Islamic Republic of Iran. reports. Educational level was classi- and quality of life in diabetic patients Based on a power analysis using a mod- fied as low (illiterate/primary/middle [7,8]. erate effect size and probability level of school) or high (high school/college/ Some sociodemographic, behav- 0.05, significance level of 0.05 and 0.80 university). ioural and clinical factors seem to be power, a sample size of 300 participants All patients completed a Farsi ver- associated with depression in diabetic was judged to be adequate. sion of the Beck Depression Inventory patients although the results are con- Physicians referred patients for (BDI) [13]. The BDI is a standard self- flicting. Research has shown a high risk participation in the study based on the reported questionnaire containing 21 for female sex, younger age, lower edu- following criteria: currently taking any items that measure the presence and cation and lower income [2]. Another diabetic medication, fasting blood glu- severity of cognitive and somatic symp- study indicated that age and chronic cose > 126 mg/mL (confirmed with a toms of depression on a scale from conditions may be significant independ- second out-of-range test); and a hos- 0–63, with higher scores indicating ent predictors of depressive symptoms, pital discharge diagnosis of diabetes. greater depression. Each item evalu- but that depression was unrelated to Patients with a current diagnosis of type ates a category according to a scale of sex, ethnic group, duration or type of 2 diabetes and age > 30 years were as- 4 possible responses of increasing se- diabetes [9]. signed consecutively to the study. Pa- verity. It has been validated in patients Despite the many reports of an tients diagnosed with type 1 diabetes with diabetes and screens effectively association between depression and (onset < 30 years of age and insulin for major depression in this population diabetes, the validity of these findings as the first treatment prescribed) were [3,14,15]. Standard cut-offs were used in different cultures and communities excluded [2]. in this study: 0–15 (no depression); remains to be shown [10]. Diabetes cur- The aim of the study was explained 16–30 (mild), 31–46 (moderate) and rently affects approximately 3%–5% of to all participants, who signed the con- ≥ 47 (severe). the population of the Islamic Republic sent form. Permission to conduct this All information was collected by in- of Iran and this figure is expected to research was granted by the ethics com- terview. Two trained medical students rise considerably by the year 2025 [11]. mittee at the Urmia Medical Sciences interviewed the patients and filled out Furthermore, the Global Burden of University. Only 5 patients refused to the forms. They also helped patients to Disease Survey estimated that by 2020 participate and therefore 295 patients fill out the BDI if they had any prob- major depression will be second leading were enrolled in this study. lems.

287 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Analysis combination). Half of patients (50.9%) < 0.05) (Table 1). Patients who were ≥ All statistical analyses were done us- reported having at least 1 complication 50 years had a significantly higher risk ing SPSS, version 12.0. The prevalence of diabetes, e.g. cardiovascular disease, of having depression symptoms than was determined by simple percentages. nephropathy, retinopathy and diabetic those < 50 years (OR 1.64, 95% CI: We also examined group differences foot; 10.1% experienced 2 or more com- 1.21–2.42) (P < 0.01). in sociodemographic, behavioural and plications. There were higher rates of More women (94/205, 45.9%) clinical variables between patients with cardiovascular disease and nephropathy than men (34/90, 37.8%) had depres- and without depression by using the compared with other complications. sion scores on the BDI but this was not chi-squared test for categorical variables Patients with 1 or more complication statistically significant. However, mean and independent t-test for continuous had a significantly higher mean duration BDI scores for women were significant- data. The effect of variables on depres- of diabetes than those without any com- ly higher than for men [16.6 (SD 10.4) sion was also estimated in men and plications [9.4 (SD 6.8) years versus versus 12.8 (SD 8.1)] (P < 0.02). women separately. The P-value was 7.3 (SD 6.0) years] (P < 0.005) and a Depressed patients were less edu- significant at < 0.05. significantly higher mean age [56.0 (SD cated than nondepressed patients (only Logistic regression models were 13) years versus 47.5 (SD 12) years] (P 81.3% versus 68.3% had low educational constructed to model the odds of hav- < 0.001). level) (P < 0.05) and poorer (64.1% versus 52.4% had low income level) ing depression versus no depression. All Prevalence of depression variables were examined for the associa- (P < 0.05) (Table 1). There was also tion with depression including sociode- Of the 295 patients, 128 (43.4%) met significant difference regarding marital mographic factors (age, sex, education, the diagnostic criteria for depression status; a higher proportion of depressed patients were married (97.4%) than marital status and income), behavioural (BDI score ≥ 15); 36.3% were classi- nondepressed patients (92.9%) (P < risk factors (BMI and smoking) and fied as minor (BDI score 16–30), 5.8% 0.05). diabetes clinical factors (duration of dia- as moderate (BDI score 31–46) and betes, treatment intensity and number 1.3% as severe depression (BDI score ≥ Clinical factors associated with de- of complications). Odds ratios (OR) 47). The mean score on the BDI for the pression were duration of diabetes and and their 95% confidence intervals (CI) whole sample of diabetic patients was having complications of diabetes. The were estimated for all variables in the 15.4 (SD 9.5). mean duration of diabetes was 9.2 (SD models. 6.8) years in depressed and 7.8 (SD Factors associated with 5.1) years in nondepressed patients (P depression < 0.05). Other clinical and behavioural Results Compared with the group without factors (treatment intensity, smoking, depression, those with depression were overweight) had the same distribution Background characteristics significantly older [mean age 55.6 (SD in depressed and nondepressed diabetic The mean age of this sample of diabetic 12) years versus 50.4 (SD 13) years] (P patients (Table 1). patients was 52.4 [standard deviation (SD) 12.0] years. Of the 295 patients Table 1 Demographic, behavioural and clinical features of diabetic patients with 205 (69.5%) were women. The mean and without depression scores on the Beck Depression Inventory (BDI) age of males and females was not sig- Variable No depressiona Depressionb P-valuec nificantly different [51.6 (SD 13) years (n = 167) (n = 128) versus 53.5 (SD 12) years]. Three-quar- Sex (male/female) (no.) 56/111 34/94 0.19 ters (74.8%) of participants had a low Age [mean (SD) years] 50.4 (13) 54.6 (12) 0.04 education level, below high school. The High educational level (%) 30.6 18.7 0.03 sample was predominantly low income Married (%) 92.9 97.4 0.04 with 55.9% of participants reporting an Low income level (%) 52.4 64.1 0.03 income < US$ 150 per month. Overweight (%) 79.9 76.5 0.09 The mean BMI was 28.4 (SD 4.9) Current smoker (%) 9.0 11.8 0.48 2 kg/m , and 78.2% of participants were Duration of diabetes [mean (SD) years] 7.8 (5.1) 9.2 (6.8) 0.03 2 overweight (BMI ≥ 25 kg/m ). Treat- Complications of diabetes (1+) (%) 46.2 55.6 0.04 ment intensity showed that 83.0% were Treatment intensity (insulin therapy) (%) 19.4 14.6 0.34

managed with oral agents/diet only aBDI score < 15; bBDI score ≥ 15; cχ2 test or t-test. and 17.0% were on insulin (alone or SD = standard deviation.

288 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 2 Demographic, behavioural and clinical features of diabetic men and women with depression (Beck Depression Inventory score ≥ 15) Variable Men Women OR (95% CI) P-value (n = 34) (n = 94) Age [mean (SD) years] 55.7 (16.9) 54.2 (12.4) – 0.88 Family history of diabetes (%) 47.1 38.3 1.43 (0.64–3.10) 0.19 High educational levela (%) 38.2 11.7 4.20 (1.10–5.60) < 0.0001 Married (%) 91.2 98.8 1.50 (1.30–3.10) 0.04 Low income level (%) 51.5 64.9 0.57 (0.20–1.20) 0.06 Overweightb (%) 72.7 87.1 2.35 (1.95–6.70) < 0.0001 Current smoker (%) 30.0 4.3 9.40 (2.10–11.30) < 0.0001 Duration of diabetes [mean (SD) years] 9.8 (6.8) 8.9 (6.7) – 0.84 Complications of diabetes (%) 44.1 44.7 0.97 (0.44–2.10) 0.88 Treatment intensity (insulin therapy) (%) 14.5 14.8 0.90 (0.70–1.30) 0.73

aHigh school and more; bBMI = body mass index ≥ 25 kg/m2. SD = standard deviation; OR = odds ratio; CI = confidence interval.

Female patients with depression female subjects. Low income was the reported in other studies in the Islamic were more likely to be overweight (OR only variable that influenced having Republic of Iran [15–17]. 2.35, 95% CI: 1.95–6.70) and to be depression in diabetic men; 32.1% non- Epidemiological studies of psychi- married (OR 1.50, 95% CI: 1.30–3.10) depressed versus depressed 51.5% men atric disorders in the general popula- compared with men. Male patients had a low income). tion in the Islamic Republic of Iran were more to have a high educational Using logistic regression analysis, are limited and the prevalence varies level (OR 4.20, 95% CI: 1.10–5.60) entering all confounder variable in the across different studies. A survey in west than women. Other variables, e.g. family model, older age was the only variable Azerbaijan reported the prevalence of history of diabetes, complications, low that remained significantly related to psychiatric disorders as 29.0% (40.2% income, duration of diabetes, treatment depression (OR = 2.02, 95% CI: 1.52– in women and 17.7% in men), anxi- intensity and age, had no relationship 3.80, exp(B) = 1.023, P < 0.01, constant: ety 14.0% and mood disorders 8% [18]. with the sex of depressed patients (Ta- –1.49). Another epidemiological assessment of ble 2). mental disorders among people aged 15 Although depressed men and wom- years or older in urban Natanz showed en had a similar prevalence of compli- Discussion that the most prevalent disorders were cations, there were sex differences in dysthymia (5.8%), generalized anxiety the relationship between complications In view of the increasing prevalence of (5.3%) and depression (3.3%) [19]. In and depression, and the association was diabetes in developing countries and Hamedan, 8.1% of studied individuals statistically significant for women but its association with depression, the suffered from at least one psychiatric not for men. Comparing women with detection and control of depression disorder, of which 2.6% had major de- depression with nondepressed women, symptoms has some benefits in the pression [20]. Anxiety and mood disor- older age (> 50 years) (58.2% versus management of patients with diabetes. ders were the most prevalent psychiatric 50.5%), having complications (55.3% versus 40.5%), being unmarried (92.8% The overall prevalence of depres- disorders in east Azarbaijan province versus 98.9%) and duration of diabe- sion in our setting, based on BDI scores (8.5% and 2.9%) [21]. Trends in per- tes [mean 7.3 (SD 5.2) versus 8.9 (SD ≥ 15, was 43.4%. Severe depression was ceptions of depression may be partly 6.1) years] had a significant effect on detected in only 1.4% of patients. The attributable to cultural attitudes toward depression (P < 0.05). Twice as many overall frequency of elevated depression depression [7,10]. women patients with depression had symptoms was quite high compared In our study, diabetic patients with 2 or more complications of diabetes with studies in other countries, where depression scores on the BDI had fewer (12.0%) compared with women with investigators found elevated depression years of education, lower income and no depression (6.5%). However, we scores on the BDI in 26% of diabetic were older than those who did not. In- could find no relation between other patients [7,14]. However, a similar fre- deed financial stress and social support behavioural and clinical factors among quency of depression as our study was may contribute the most to depression

289 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

among diabetic patients. These findings stress; therefore, increased attention One limitation of the study was agree with studies which found that the should be given to these cases [25]. the use of self-reported questionnaires frequency of depression was associated A gender effect on the frequency of instead of psychiatric interviews. Many with social and financial factors [2,22]. depression has been documented for studies, however, have used the same Age was the only variable related to several conditions [14]. In our study methods. Another limitation in our depression in our multiple regression the prevalence of elevated depression study was to evaluate education and in- model. The greater risk of depression symptoms was the same in women and come based on patients’ self-reporting. in older versus younger patients may men but the mean BDI score was signifi- Despite these limitations the results are be because older patients have more cantly higher in women. Although some valuable in that they confirm the impor- problems in coping physically and psy- studies found that women with diabetes tance of age as an associated variable in depression in diabetic patients [27]. chologically with a greater number of experience more psychological distress diabetes complications. than men [16,17,25], a meta-analysis In conclusion, the high prevalence of An association between disease revealed that the increased odds of depression in diabetes patients in Urmia is an important finding, as depression duration and depression was revealed depression associated with diabetes ac- has a negative impact on quality of life. in our study. In contrast, Fisher et al. cording to sex were about the same [7]. Depression has additional importance found no relation between length of BMI had no effect on depression in diabetes because of its association diabetes and depression [23]. Other score in our diabetic patients. This is in with poor compliance with diabetes studies showed that disease duration contrast to Robinson et al. and Gavard treatment, poor glycaemic control and was significantly correlated with depres- et al. who reported that currently de- an increased risk of micro- and macro sion scores [15–17,22,24]. Meanwhile, pressed diabetic individuals had a signif- vascular disease complications [2]. treatment type had only a weak influ- icantly higher BMI than nondepressed People with diabetes and depression ence on depression severity in some diabetics [26]. However, the frequency have greater use of emergency care, out- 2 studies [17,25,26] and the intensity of of BMI ≥ 25 kg/m was high in our patient primary care, medical and psy- therapy (insulin or other) had no effect study sample and significantly more chiatric specialty care, medical inpatient on depression in our study. Patients depressed women had high BMI than care and higher prescription costs [28]. with complications of diabetes also had did depressed men. The combination of This will become an important consid- higher depression scores in our study. depression and obesity may worsen the eration as the prevalence of diabetes Patients with complications typically course of diabetes because both factors is estimated to rise in Asia with a cor- require long-term medical care and are are associated with an increased risk of responding increase in the economic consequently suffer more psychological adverse cardiac outcomes. burden of the disease [10,29,30].

References

1. Brown LC et al. History of depression increases risk of type 2 8. De Groot M et al. Depression treatment and satisfaction in diabetes in younger adults. Diabetes care, 2005, 28(5):1063–7. a multicultural sample of type 1 and type 2 diabetic patients. 2. Katon W et al. Behavioral and clinical factors associated with Diabetes care, 2006, 29(3):549–53. depression among individuals with diabetes. Diabetes care, 9. Eaton WW et al. Depression and risk for onset of type II diabe- 2004, 27(4):914–20. tes. A prospective population-based study. Diabetes care, 1996, 19(10):1097–102. 3. Lustman PJ, Clouse RE. Depression in diabetic patients: the relationship between mood and glycemic control. Journal of 10. Zahid N et al. Depression and diabetes in a rural commu- diabetes and its complications, 2005, 19(2):113–22. nity in Pakistan. Diabetes research and clinical practice, 2007, 79(1):124–7. 4. Katon WJ et al. The association of comorbid depression with mortality in patients with type 2 diabetes. Diabetes care, 2005, 11. King H, Aubert RE, Herman WH. Global burden of diabetes, 1995–2025: prevalence, numerical estimates, and projections. 28(11):2668–72. Diabetes care, 1998, 21(9):1414–31. 5. Bruce DG et al. A prospective study of depression and mortal- 12. Murray CJL, Lopez AD, eds. The Global Burden of Disease. ity in patients with type 2 diabetes: the Fremantle Diabetes A comprehensive assessment of mortality and disability from Study. Diabetologia, 2005, 48(12):2532–9. diseases, injuries and risk factors in 1990 and projected to 2020. 6. Katon W et al. Cost-effectiveness and net benefit of enhanced Cambridge, Massachusetts, Harvard School of Public Health treatment of depression for older adults with diabetes and on behalf of the World Health Organization and the World depression. Diabetes care, 2006, 29(2):265–70. Bank, 1996. 7. Anderson RJ et al. The prevalence of comorbid depression 13. Kaviani H, Musavi A, Mohit A. [Psychological interviews and in adults with diabetes: a meta-analysis. Diabetes care, 2001, scales.] Tehran, Islamic Republic of Iran, SANA Publishers, 2001 24(6):1069–78. [in Farsi].

290 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

14. Tellez-Zenteno JF, Cardiel MH. Risk factors associated with 22. Musselman DL et al. Relationship of depression to diabetes depression in patients with type 2 diabetes mellitus. Archives of types 1 and 2: epidemiology, biology, and treatment. Biological medical research, 2002, 33(1):53–60. psychiatry, 2003, 54(3):317–29. 15. Taziki A et al. [The association between depression and dia- 23. Fisher L et al. Contributors to depression in Latino and Euro- betes.] Journal of Gorgan University of Medical Sciences, 2001, pean–American patients with type 2 diabetes. Diabetes care, 8:59–64 [in Farsi]. 2001, 24(10):1751–7. 16. Sepehrmanesh Z et al. [The prevalence of depression and its 24. Starostina EG. Depression and anxiety in type 2 diabetes in related factors among diabetic patients.] Journal of Kashan relation to disease characteristics and psychosocial variables: University of Medical Sciences, 2002, 7:69–75 [in Farsi]. a pilot assessment. Journal of psychosomatic research, 2006, 17. Larijani B et al. [The association between depression and dia- 60(6):663. betes among diabetic patients in Shariati hospital.] Iranian journal of diabetes and lipid disorders, 2003, 3(1):77–82 [in Farsi]. 25. Bai YL et al. Correlates of depression in type 2 diabetic elderly patients: a correlational study. International journal of nursing 18. Mohammadi MR et al. [An epidemiological study of psychiatric disorders in west Azerbaijan province.] Urmia medical journal, studies, 2008, 45(4):571–9. 2004, 14(4):242–52 [in Farsi]. 26. Sevincok L et al. Depression in a sample of Turkish type 2 dia- 19. Omidi ATA, Sazvar SA, Akkasheh G. [Epidemiology of mental betes patients. European psychiatry, 2001, 16(4):229–31. disorders in urbanized areas of Natanz.] Iranian journal of psy- 27. Gilmer TP et al. Predictors of health care costs in adults with chiatry and clinical psychology, 2001, 8(4):32–8 [in Farsi]. diabetes. Diabetes care, 2005, 28(1):59–64. 20. Mohammadi MR et al. [An epidemiological study of psychiatric 28. Zhao W et al. Association between diabetes and depression: disorders in Hamadan province, 2001.] Scientific Journal of Ha- sex and age differences. Public health, 2006, 120(8):696–704. madan University of Medical Sciences and Health Services, 2004, 11(3):28–36 [in Farsi]. 29. Simon GE et al. Diabetes complications and depression as predictors of health service costs. General hospital psychiatry, 21. Mohammadi MR et al. [An epidemiological study of psychiat- 2005, 27(5):344–51. ric disorders in east Azerbaijan province.] Medical Journal of Tabriz University of Medical Sciences and Health Services, 2005, 30. Shah ZC, Huffman FG. Depression among Hispanic women 64:67–73 [in Farsi]. with type 2 diabetes. Ethnicity and disease, 2005, 15(4):685–90.

Mental health systems in selected low- and middle-income countries

Mental health systems in selected low- and middle-income countries summarizes descriptive data on mental health systems of selected low- and middle-income countries (LAMICs) using the World Health Organization Assessment Instrument for Mental health Systems (WHO-AIMS).

The comprehensive and detailed information gathered through WHO-AIMS and summarized in this report provides a better understanding of mental health systems in these countries. Results indicate that mental health resources and activities are scarce, inequitably distributed and inefficiently used; community-based mental health services are underdeveloped; mental health systems are often not well connected to other relevant sectors, such as the primary health care system; and that insufficient attention is given to human rights.

This report highlights the urgent need for additional resources, and the importance of ensuring better use of the limited resources available. The information derived from this WHO-AIMS study is being used to develop plans for strengthening community care and scaling up services for people with mental disorders.

Further information about this and other WHO publications can be found at: http://www.who.int/publications/en/

291 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Profile of diabetic ketoacidosis at a teaching hospital in Benghazi, Libyan Arab Jamahiriya R.R. Elmehdawi 1 and H.M. Elmagerhei 2

َمرت َسم احلامض الكيتوين السكري يف مستشفى تعليمي يف بنغازي، اجلامهريية العربية الليبية رفيق رمضان املهدوي، حنان حممد املقرحي

اخلالصـة: تصف هذه الدراسة َمرت َسممئة حالة من احلامض الكيتوين السكري يف مستشفى تعليمي يف بنغازي، يف اجلامهريية العربية الليبية. واتضح للباحثني أن احلامض الكيتوين السكري أكثر ًشيوعا لدى الشابات املصابات بالنمط األول من السكري، وأنه ينجم يف غالب األحيان عن أسباب يمكن ِّتوقيهامثل اختالل املعاجلة باإلنسولني أو اإلصابة بالعدوى. كام حيدث احلامض الكيتوين السكري ً أيضالدى السكريـني من النمط الثاين، وبمعدل مرتفع للوفيات ًنظرا ألهنم أكرب ًعمرا ولدهيم حاالت مرضية مرافقة. وأكثر املظاهر الرسيرية )اإلكلينيكية( ًشيوعا هي تعدد َالب ْيالت والتعب واألمل البطني والقيء، أما ُالسبات )الغيبوبة( ُفهم أندر ً.وكان حدوثااحلامض السكري الكيتوين لدى عدد كبري من احلاالت هو التظاهرات األوىل للسكري من النمط األول؛ ومن هنا ينبغي أن يؤخذ تشخيص احلامض السكري الكيتوين لدى مجيع املرىض الذين يشكون من بطن حاد أو من نقص مستوى الوعي. وتعود أسباب ارتفاع معدل الوفيات يف هذه الدراسة )10%( إىل عوامل متعددة.

ABSTRACT This study described the profile of 100 cases of diabetic ketoacidosis (DKA) at a teaching hospital in Benghazi, Libyan Arab Jamahiriya. DKA was more frequent in young women with type 1 diabetes and mostly due to preventable causes, e.g. disrupted insulin treatment and/or infection. DKA also occurred in type 2 diabetics, with a higher mortality rate, as they were older patients with co-morbidity. Polyurea, fatigue, abdominal pain and vomiting were the most common clinical features, while coma was rarer. A high number of cases were first presentations of type 1 diabetes; hence this diagnosis should be considered in all patients with acute abdomen or decreased level of consciousness. The reasons for high mortality rate in this study (10%) were multifactorial.

Profil de l’acidocétose diabétique dans un hôpital universitaire de Benghazi (Jamahiriya arabe libyenne)

RÉSUMÉ Cette étude décrit le profil de 100 cas d’acidocétose diabétique dans un hôpital universitaire de Benghazi (Jamahiriya arabe libyenne). L’acidocétose diabétique était plus fréquente chez les jeunes femmes souffrant de diabète de type 1, et était essentiellement due à des causes telles qu’une interruption de l’insulinothérapie et/ou une infection. Elle a également été observée chez des personnes souffrant de diabète de type 2, avec un taux de mortalité plus élevé, les patients étant plus âgés et présentant une comorbidité. Les principales manifestations cliniques étaient une polyurie, un état de fatigue, des douleurs abdominales et des vomissements, le coma étant plus rare. Un grand nombre de cas étaient des premières présentations de diabète de type 1. C’est pourquoi ce diagnostic devrait être envisagé chez tous les patients présentant un abdomen aigu ou une altération de la conscience. Les causes du taux de mortalité élevé dans cette étude (10 %) étaient multifactorielles.

1Department of Internal Medicine, Faculty of Medicine, Al-Arab Medical University, Benghazi, Libyan Arab Jamahiriya (Correspondence to R.R. Elmehdawi: [email protected]). 2Seventh of October Hospital, Benghazi, Libyan Arab Jamahiriya. Received: 04/11/07; accepted: 11/02/08

292 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Introduction Methods The data were analysed using SPSS, version 11.0. Differences between Type 1 diabetes mellitus (DM) is char- A descriptive cross-sectional retrospec- groups were tested statistically using the acterized by the development of a state tive analysis was made of the records chi-squared test and independent sam- of complete insulin deficiency which of adult patients admitted to the ples t-test. Differences were considered makes patients prone to ketoacidosis. Seventh of October teaching hospital, statistically significant at P-value < 0.05. Diabetic ketoacidosis (DKA) is a fre- Benghazi, and diagnosed with DKA quent cause of morbidity and mortality between January 2003 and December Results in type 1 diabetes patients. However, 2006. During this 4-year period 200 suspected DKA cases were admitted it also occurs occasionally in patients We studied 100 episodes of DKA in 81 with type 2 DM during severe stress, for to the medical ward and intensive care patients; 8 patients were responsible for example sepsis and myocardial infarc- unit of the hospital. Out of these, we 19 episodes (3 patients had 3 episodes tion [1–5]. included only 100 episodes in this study and 5 patients had 2 episodes). These About 12.8% of the diabetic adults in because for the other 100 the records 100 admissions accounted for 769 Benghazi have type 1 DM [6]. The inci- were missing important information. medical and intensive care bed-days dence of type 1 DM in the Libyan popu- The following variables were re- (2.4% of the total medical and intensive lation up to 34 years old is estimated to corded: age, sex, nationality, duration care bed-days). According to the ADA be approximately 9 per 100 000 per year of diabetes, precipitating factors for criteria, 29 of the episodes were mild, [7]. DKA is the presenting feature of DKA, duration of symptoms before 37 were moderate and 34 were severe. about 30% of Libyan children with type seeking medical advice, clinical features, Type 1 DM was the diagnosis in 90 of 1 DM [8]. However, the annual inci- laboratory findings at presentation with the episodes, while the other 10 had dence of DKA in Libyan diabetics is not DKA (plasma glucose, serum blood type 2. There were 19 episodes in newly known. Even so, the disorder seems to urea nitrogen, serum creatinine, serum diagnosed cases and 81 episodes in pre- contribute appreciably to the number of sodium and potassium, urine acetone, viously known cases of DM. hospital admissions as it constitutes 9% arterial blood pH, plasma bicarbonate Patient and disease and plasma osmolarity), associated of all diabetes-related admissions [9] characteristics and 3.2% of all acute admissions to the complications, duration of hospitaliza- medical intensive care unit in Benghazi tion and outcome. Patients were classi- A higher number of episodes were in females (71) than males (29) (female [10]. A report from Benghazi revealed fied as having type 1 or type 2 diabetes to male ratio 2.3:1) (Table 1). The that 1.6% of the adult diabetic patients based on their treatment history. age distribution is shown in Figure 1. had been admitted to hospital in the The criteria used to diagnose DKA The mean age of all patients was 29.2 previous year because of DKA 6[ ]. The in the patients included in the study years (range 15–68 years); males were overall mortality rate from DKA in the were: typical clinical features such as slightly older than females [mean age Libyan Arab Jamahiriya ranges from hyperventilation, vomiting and dehy- 32.0 (SD 14.9) years versus 28.4 (SD 2% to 9.4% of all DKA admissions ac- dration; hyperglycaemia ≥ 250 mg/dL; 9.8) years], but this difference was not cording to different reports 1,2,11[ ]. arterial blood pH ≤ 7.3 and/or serum statistically significant. Although many DKA-associated deaths bicarbonate ≤ 18 mmol/L; and pres- Patients with type 2 DM were signifi- are due to concomitant illnesses (e.g. ence of ketonuria (2+ on dipstick). cantly older than those with type 1 DM myocardial infarction, septicaemia), The severity of DKA was defined (mean age 52.6 versus 26.5 years) (P < others are preventable and are due to according to the American Diabetes 0.001) (Table 1). Previously diagnosed delay in presentation, diagnosis and/or Association (ADA) criteria [12,13] as patients were also older on average than errors in management. mild (arterial blood pH 7.25–7.30), the new cases of DM (mean age 30.1 The aim of this study was to deter- moderate (pH 7.00–7.24) or severe versus 25.1 years), but the difference mine the pattern of DKA at a teach- (pH < 7.00). Effective osmolarity was was not significant (Table 1). ing hospital in Benghazi, Libyan Arab calculated using the ADA formula: [(2 Jamahiriya, to assess the clinical and × sodium) + plasma glucose/18], with Clinical features at admission biochemical characteristics of the pa- normal values being 290 mmol/kg Disruption of insulin treatment (com- tients and the precipitating factors for water. Blood urea nitrogen was not in- plete stoppage or reduction of insulin DKA, and to correlate these parameters cluded in this calculation because urea dose or frequency) was the commonest with the outcome. has less osmotic activity [14]. precipitating factor for the episode of

293 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

DKA in 38 cases. Infection was the next most common reason – – – – 0.021 0.15 0.57 in 30 cases: urinary tract infection (13 cases), upper respiratory < 0.001 < 0.001 P -value tract infection (8), lower respiratory tract infection (3), abscess (2) and gastroenteritis, tuberculosis, diabetic foot, peritonitis (1 case each). In 19 cases DKA was the first manifestation – – _ n/a 6/4 60.0 57 (23)

122 (33) of diabetes. Other causes were rarer: myocardial infarction 103 (25) 42.5 (17.0) Outcome Died ( n = 10) (2 cases), emotional stress (2) and trauma (1). In 3 cases no obvious precipitating factor could be identified and data were – – _ missing for 5 episodes. 5.5 n/a 66/25 98 (15) 115 (28) 70 (20) ( n = 90) 27.7 (9.4) 27.7 Survived The commonest presenting symptoms in this study were general weakness and fatigue (100 cases), polyurea and poly- – – – dipsia (100), abdominal pain nausea and/or vomiting (64). 0.36 0.04 0.19 0.62 P -value < 0.001 < 0.001 < 0.001 Other symptoms were drowsiness (15 cases), dyspnoea (12), coma (5), headache (3) and haematemesis (2). Table 1 shows – – the haemodynamic features at admission. 4/6 50.0 40.0 Type 2 Type ( n = 10) 5.3 (2.7) 5.3 5.0 (4.5) 5.0 8.4 (5.0) 8.4 108 (20) 52.6 (8.1) The mean duration of symptoms before presentation to the hospital was 2.7 (SD 2.2) days (range 1–10 days) (Table

Type of diabetes Type 1), with 48% of the cases presenting within the same day of the – – 6.6 6.6 onset of symptoms and 62% within 48 hours. The duration of 66/24 Type 1 Type 99 (18) 7.4 (7.2) 7.4 ( n = 90) 7.6 (4.2) 7.6 3.0 (2.5) 3.0 26.5 (8.6) symptoms was significantly longer in patients with type 2 DM as compared with type 1 DM (P < 0.05) (Table 1). Also the – – – – – mean duration of symptoms was longer in new cases of DM as 0.76 0.004 0.089 0.60 0.93

P -value compared with the previously diagnosed cases (Table 1). The mean duration of symptoms was significantly longer in males than females [4.0 (SD 3.7) days versus 2.8 (SD 2.3) days] (P – – – – 10.5

11/8 < 0.05). First 101 (18) ( n = 19) 5.8 (5.1) 5.8 8.2 (4.3) 25.1 (8.8) 25.1

presentation Biochemical abnormalities at admission Table 2 summarizes the biochemical parameters of DKA History of diabetes cases at presentation. Plasma glucose levels were higher, – – – – 9.8

59/22 but not significantly so, in cases presenting for the first time ( n = 81) 2.4 (1.3) 2.4 7.5 (4.9) 7.5 100 (19) 30.1 (11.8) 30.1 Previously Previously diagnosed than in those with a previous diagnosis (Table 2). Serum sodium was significantly higher in the new cases of DM than the previously diagnosed cases (P < 0.05) (Table 2). Sodium n/a n/a n/a 1–10

1–25 levels were also significantly higher in males than in females 15–68 0–120 Range 0.2–27 64–170 60–200 [mean 137.9 (SD 13) versus 136.9 (SD 8) mmol/L] (P < 0.05). Serum potassium at admission was significantly lower in patients who subsequently died than in the survivors (P < 11 10 0.05) (Table 2). Blood urea nitrogen was higher for those with Total cases ( n = 100) Total 71/29 Value 69 (21) 101 (19) 115 (28) 7.7 (4.7) 7.7 7.3 (6.4) 7.3 2.7 (2.2) 29.2 (11.6) 29.2 type 2 DM than type 1 and in the new cases compared with previously diagnosed cases (Table 2) but these differences were not significant. Serum creatinine was raised in 91% of cases at admission. Complications and outcome The mean hospital stay was 7.7 days, range 1–25 days (Table 1). About 66% of the cases had some sort of complication. Some of these were due to the episode of DKA itself, e.g. coma (5 cases), hypokalaemia (serum potassium < 3.5 mmol/L) (13 cases). However, some complications developed during Systolic BP [mean (SD) mmHg] Systolic BP [mean Pulse rate [mean (SD) beats/min] [mean rate Pulse Duration of (SD) years] diabetes [mean Duration of (SD) days] symptoms [mean Female:male ratio (no.) Female:male (SD) days] Hospital stay [mean Diastolic BP [mean (SD) mmHg] Diastolic BP [mean Mortality rate (%) rate Mortality Co-morbidity (%) Co-morbidity Variable Age [mean (SD) years] Age [mean SD = standard deviation;SD = standard n/a = not applicable. BP = blood pressure; Background characteristics 1 Background of and clinical features cases of diabeticTable at presentation ketoacidosis management due to improper therapeutic interventions,

294 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Figure 1 Age distribution of episodes of diabetic ketoacidosis (n = 100)

e.g. pneumothorax (1 case) and hy- DKA cases; none of the patients with Patient and disease pokalaemia from treatment (3 cases). mild DKA died. characteristics The most common complications en- An important comorbidity was While most of the episodes of DKA in countered during DKA were prerenal associated with 6 of the 10 deaths: this study were due to type 1 DM, 10% azotaemia (urea > 40 mmol/L) (28 myocardial infarction (2 cases), acute were due to type 2 DM; this compares cases) and electrolyte disturbances, abdomen, pulmonary tuberculosis, with 15% reported by Roaeid from the e.g. hypokalaemia (16), hyperkalae- acute renal failure or diabetic foot (1 same hospital between 1997 and 1999 mia (13) and hypernatraemia (serum case each). The following parameters [1] and 11.6% reported by El-Sharief in sodium > 145 mmol/L) (10). More were associated with a significantly in- another Libyan study from Tripoli [2]. serious complications were rare: e.g. up- creased mortality: age > 40 years, type Although DKA is usually considered as per gastrointestinal bleeding (2 cases), 2 DM, depressed level of conscious- a unique marker for type 1 DM these acute gastric dilatation (1), acute renal ness at presentation, comorbidity, pulse reports clearly demonstrate that it can failure (1 case), adult respiratory dis- rate ≥ 115 beats/min, systolic blood occur among type 2 diabetic patients. tress syndrome (1) and disseminated pressure ≤ 105 mmHg, diastolic blood This fact is increasingly recognized intravascular coagulation (1). pressure ≤ 65 mmHg, plasma glucose throughout the world [3–5]. There were 10 deaths among the ≥ 525 mg/dL, serum sodium ≥ 144 Females were 70% of all DKA DKA cases (10% of episodes, 12.3% mmol/L, blood urea nitrogen ≥ 50 mg/ admissions. A female predominance of patients) (4 males and 6 females), dL, serum creatinine ≥ 4 mg/dL, arterial was previously reported in Benghazi and the mortality rate among males blood pH ≤ 7 and plasma osmolality ≥ by Roaeid and Kablan [1], whereas El- was slightly higher than females (13.8% 325 mOsm/kg water (Table 3). Sharief reported a nearly equal ratio of versus 8.4%). The mean age of females sexes in Tripoli [2]. This discrepancy who died was significantly lower than in the sex ratio between different co- the males [37 (SD 17) years versus Discussion horts was also observed in studies from 51 (SD 16) years] (P < 0.001). The Saudi Arabia [15] and Jordan [16]. We mortality rate was significantly higher Despite the small sample size of this think that the cause of these variations is among patients with type 2 DM than study, and the fact that it was a retro- multifactorial, including environmental, those with type 1 (40.0% versus 6.6%) spective analysis conducted at a teach- genetic and social differences between (P < 0.001) (Table 1). Regarding the ing hospital, all of which are important the cohorts. However, the small study severity of DKA, there were 8 deaths limitations, this study gives a basic samples and errors in sampling could (23.5%) among the severe cases and profile of DKA in this part of the world be important contributing factors. The 2 deaths (5.4%) among the moderate where studies about DKA are few. female predominance was maintained

295 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

in the group of known diabetics, among the new cases and in those with type 1 DM. However, 0.20 0.25 0.008 0.043 0.013 0.97 0.47 0.005 P -value this predominance was reversed in patients with type 2 DM, where males constituted 60% of the cases. 5.3 (2.6) 5.3 4.18 (1.1) 4.18 527 (88) 67.5 (40) 67.5 311.6 (36) 311.6 7.20 (5.6) 7.20 142.0 (18) The mean age of all patients (29.2 years) is 6.80 (0.17) Mean (SD) Mean Died ( n = 10) similar to that previously reported by Roaeid Outcome and Kablan (28.9 years) [1]. Males were slightly older than females, although this was not statistically significant. However, patients with type 3.3 (2.0) 3.3 41.4 (29) 41.4 135.0 (8) 135.0 494 (116) 300.7 (17) 4.50 (0.8) 6.60 (4.0) 7.05 (0.21) 7.05 Mean (SD) Mean 2 DM were significantly older than those with

Survived ( n = 90) type 1 (52.6 and 26.5 years respectively) pre- sumably because type 2 DM tends to develop in adults older than 40 years while type 1 DM 0.90 0.78 0.12 0.40 0.29 0.99 0.45 0.65 P -value usually occurs at a much younger age. Clinical features at admission Poor drug compliance was the major precipitat- 4.8 (2.1) 523 (131) 300 (23) 60.4 (31) 60.4

Mean (SD) Mean ing factor of DKA in our patients (38%); which 4.49 (1.0) 4.49 6.68 (4.1) 139.6 (14) 139.6 7.00 (0.23) 7.00 Type 2 ( n = 10) Type is similar to El-Sharief’s report from Tripoli

Type of diabetes Type (36.7%) [2], but lower than 54.4% from Saudi Arabia [15]. However, it worth mentioning that the rate of DKA due to treatment disruption 3.5 (2.5) 3.5 45.2 (41) 45.2 494 (111) 302 (20) 136.9 (9) in the current study (from 2003–06) is much 6.68 (4.3) 4.50 (0.8) Mean (SD) Mean 7.03 (0.21) 7.03

Type 1 ( n = 90) Type lower than in the previous study in Benghazi from the years 1997–99 (64%) [1]; this might indicate an improvement in patients’ compli- 0.006 0.89 0.08 0.16 0.11 0.007 0.017 0.11 P -value ance and education. Infections were the second most common precipitating factor (30%), similar to other reports from Benghazi (29.7%) [1], Tripoli (21.7%) [2] and Saudi Arabia ( n = 19) 4.3 (2.4) 553 (124) 142.4 (14) 142.4 4.54 (1.0) 70.0 (43) 5.28 (4.4) 5.28 (28%) [15]. In developed countries infection 309.7 (25) 309.7 Mean (SD) Mean 6.96 (0.25) is the commonest precipitating factor of DKA, First presentation while insulin disruption ranks second, perhaps

History of diabetes due to better patient education 17[ ]. Cases of DKA caused by drug omission and infection are potentially preventable with good patient 3.4 (2.0) 3.4 300 (19) 135.8 (7) 135.8 41.3 (24) 41.3 7.20 (4.1) 7.20 484 (108) 4.50 (0.9) 7.05 (0.20) 7.05 Previously Previously Mean (SD) Mean education and proper outpatient management.

diagnosed ( n = 81) About 3% of our patients had no obvious precipitating factor, which is in concordance with 1–8 1–17.4 2%–10% reported from other parts of the world 10–152 Range 118–176 6.7–7.3 2.9–6.8 250–836 270–385 [18]. The mean duration of symptoms before presentation to the hospital was 2.7 days, with

3.1 (1.8) 3.1 48% of the patients presenting within the same Total cases ( n = 100) Total 43.4 (31) 43.4 497 (114) 137.0 (10) 137.0 6.68 (4.2) 301.9 (20) 301.9 4.50 (0.9) 7.03 (0.21) 7.03 Mean (SD) Mean day of the onset of symptoms, and this probably contributed to the low rate of serious clinical features at presentation. However, the duration was longer in those who were not known diabetics (5.8 versus 2.7 days), because patients may not realize what is happening at the first Parameter Serum potassium Serum sodium Serum creatinine Plasma osmolality nitrogen Blood urea Plasma glucose Plasma bicarbonate (mmol/L) (mmol/L) (mOsm/kg water) (mg/dL) (mmol/L) (mg/dL) (mg/dL) Arterial blood pH Arterial SD = standard deviation.SD = standard Laboratory parameters of parameters 2 Laboratory cases of diabetic at presentationTable ketoacidosis episode of DKA. The duration of symptoms

296 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 3 Risk factors for mortality among 100 episodes of diabetic ketoacidosis at in males than in females, suggesting that presentation males postpone medical consultation Variable Mortality (%) P-value longer than females. Age (years) Fatigue, polyurea, abdominal pain ≥ 40 27.7 < 0.001 and vomiting were the most common < 40 6.0 presenting symptoms in our study, Sex which is similar to the report from Male 13.8 0.46 Tripoli [2]. Only 5% of our patients had Female 8.5 coma at presentation compared with Type of diabetes 25% reported from a study in Nairobi Type 2 40.0 < 0.001 [19], perhaps due to ethnic differences Type 1 6.6 in diabetes severity or better education Co-morbidity among our patients and easier avail- Yes 50.0 < 0.001 ability of immediate health care services No 4.5 in our country which could contribute Level of consciousness to earlier presentation and hence less Unconscious 35.0 < 0.001 severe clinical features. Fully conscious 3.8 Biochemical abnormalities Pulse (beats/min) The mean plasma glucose level was ≥ 115 22.7 0.024 higher in the new cases of DM and < 115 7. 4 in those with type 2 DM, probably Blood pressure (mmHg) because patients belonging to these Systolic ≤ 105 21.6 0.005 groups presented 2 to 3 days later than Systolic > 105 3.5 the known cases of type 1 DM, allow- Diastolic ≤ 65 22.5 0.008 ing a higher degree of hyperglycaemia Diastolic > 65 4.8 to develop. The same was true for the Plasma glucose (mg/dL) higher serum sodium and blood urea in ≥ 525 12.0 0.04 these groups, which indicates a more se- < 525 8.6 vere dehydration due to longer duration Arterial blood pH of glucose-mediated osmotic diuresis ≤ 7 16.3 < 0.001 and water loss in excess of electrolytes. > 7 3.9 The great majority of patients (75%) Blood urea nitrogen (mg/dL) were eukalaemic at presentation, 13% ≥ 50 24.0 0.012 were hypokalaemic, while another 3% < 50 5.5 developed hypokalaemia as a result Serum creatinine (mg/dL) of potassium under-replacement dur- ≥ 4 24.1 0.039 ing management. Hyperkalaemia was < 4 4.9 nearly as common as hypokalaemia at admission. The raised serum creatinine Serum sodium (mmol/L) seen in most of the DKA patients was ≥ 144 28.5 0.02 likely a false elevation, as acetoacetate < 144 7.6 artificially raises measured creatinine in Plasma osmolality (mOsm/kg water) the standard colorimetric assay [20,21]. ≥ 325 44.0 0.018 Metabolism of the acetoacetate follow- < 325 7. 1 ing the administration of insulin will lower the measured serum creatinine concentration towards its true value. was significantly longer for those with slowly due to secondary insulin defi- type 2 DM than type 1 DM because ciency which builds up gradually as a re- Outcome these patients are not primarily ketosis- sult of glucose toxicity. The duration of The mean hospital stay was 7.7 days, prone but ketoacidosis may develop symptoms was also significantly longer with some cases discharged on their

297 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

request as early as the first day of admis- previously reported in Benghazi (3.8%) Conclusion sion and a few others who stayed at [1], which might indicate a deficiency hospital more than 2 weeks for treating in diagnostic and therapeutic facilities DKA was more common in Libyan the precipitating factors and compli- at hospitals or a decline in doctors’ and females than in males. It was frequently cations or for social reasons. This is co-workers’ skills and experience in the seen in young women with type 1 DM, similar to the average length of stay in mostly caused by potentially prevent- management of DKA. The mortality other reports from the Libyan Arab able causes such as disruption of insulin rate was also higher than figures from Jamahiriya [1,2]. However, it is longer treatment and/or infection. DKA also than the average length of stay for DKA Saudi Arabia [15] and Jordan [16] occurred in type 2 DM patients and in the United States (3.6 days) [22] (2.9% and 4.8% respectively); however, when this happened it resulted in higher and also longer than in southern Jordan it was better than from Nairobi, where mortality as it occurred in older persons where the average length of stay was the mortality rate was nearly 30% [19]. with coexisting morbidity. 3.4 days [16]. Our longer hospitaliza- This high variation in the mortality rates Polyurea, fatigue, abdominal pain and tion will result in higher costs of DKA may reflect variability in access to medi- vomiting were the most common clini- management. cal services and standards of inpatient cal features in our patients, while coma Fortunately, despite the high rate management as well as patient educa- was relatively uncommon. DKA as a first of DKA-related complications (68%) tion. presentation of type 1 DM accounted for most of the complications were not a considerable number of cases; hence serious and were easy to treat, e.g. mild DKA patients with type 2 DM had this diagnosis should be considered in electrolyte disturbances and prerenal a significantly higher mortality rate all patients with acute abdomen or de- azotaemia. However, serious complica- (40.0%) than patients with type 1 DM creased level of consciousness, even if tions were encountered in some cases, (6.6%), probably because type 2 DM they are not known diabetics. e.g. severe electrolyte disturbances. patients were significantly older and The high mortality rate in this study Unfortunately, some of the complica- more likely to have comorbidity. Over- (10% of episodes) was due to multi- tions were iatrogenic, e.g. such as pneu- all, patients who died were significantly factorial reasons, partly patient-related mothorax due to insertion of a central older, with more comorbidity, a lower factors such as old age, and comorbidity, venous line and hypokalaemia due to and partly disease-related factors such level of consciousness at presentation, potassium under-correction. as DKA severity. However some may lower blood pressure, faster pulse rate, Despite the low rate of serious com- have been due to therapy-related factors plications, the mortality rate of our DKA higher plasma glucose, higher serum such as errors in management. Further cases was high (10%). While this is simi- sodium, higher serum creatinine, higher studies are needed to clarify the factors lar to the report from Tripoli (9.4%) [2], plasma osmolality and lower blood behind this high mortality rate and steps it is much higher than the rate that was pH. to reduce the rate should be taken.

References

1. Roaeid R, Kablan A. Diabetic ketoacidosis in Benghazi charac- 7. Kadiki OA, Reddy MR, Marzouk AA. Incidence of insulin-de- teristics and outcome in 211 patients. Garyounis medical journal, pendent diabetes (IDDM) and non-insulin-dependent diabe- 2004, 21(1):11–4. tes (NIDDM) (0–34 years at onset) in Benghazi, Libya. Diabetes 2. El-Sharief HJ. Diabetic ketoacidosis: Tripoli Medical Center research and clinical practice, 1996, 32(3):165–73. experience. Jamahiriya medical journal, 2006, 5 (1):51–4. 8. Kadiki OA, Gerryo SE, Khan MM. Childhood diabetes mellitus 3. Seyoum B, Berhanu P. Profile of diabetic ketoacidosis in a in Benghazi (Libya). Tropical pediatrics, 1987, 33(3):136–9. predominantly African American urban patient population. 9. Roaeid RB. Hospital admissions of diabetic patients in Beng- Ethnicity and disease, 2007, 17(2):234–7. hazi. Diabetes international, 2002, 12 (1):24–5. 4. Newton CA, Raskin P. Diabetic ketoacidosis in type 1 and type 10. Roaeid RB et al. Admission patterns and outcome in an adult 2 diabetes mellitus: clinical and biochemical differences. Ar- medical intensive care unit in Benghazi. Garyounis medical chives of internal medicine, 2004. 27, 164 (17):1925–31. journal, 2005, 22(1):61–6. 5. Umpierrez GE, Smiley D, Kitabchi AE. Narrative review: ketosis- 11. Lakhdar AA, Elharboush S. Characteristics and outcome of prone type 2 diabetes mellitus. Archives of internal medicine, ketoacidosis in Libyan diabetic patients. Diabetes international, 2006, 144(5):350–7. 1999, 16(6):171–3. 6. Roaeid RB, Kablan AA. Profile of diabetes health care at Beng- 12. Kitabchi AE et al. Hyperglycemic crises in adult patients with hazi Diabetes Centre, Libyan Arab Jamahiriya. Eastern Mediter- diabetes: a consensus statement from the American Diabetes ranean health journal, 2007, 13(1):168–76. Association. Diabetes care, 2006, 29 (12):2739–48.

298 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

13. Kitabchi AE et al. Hyperglycemic crises in diabetes. Diabetes totic syndrome. American journal of medical science, 1996, care, 2004, 27(Suppl. 1):S94–102. 311(5):225–33. 14. Siperstein MD. Diabetic ketoacidosis and hyperosmolar coma. 19. Mbugua PK et al. Diabetic ketoacidosis: clinical presenta- Endocrinology and metabolism clinics of North America, 1992, tion and precipitating factors at Kenyatta National Hos- 21(2):415–32. pital, Nairobi. East African medical journal, 2005, 82(12 15. Qari FA. Precipitating factors for diabetic ketoacidosis. Saudi Suppl.):S191–6. medical journal, 2002, 23(2):173–6. 20. Molitch ME et al. Spurious serum creatinine elevations in keto 16. Tahboub I, Shalan JB. Diabetic ketoacidosis in southern Jordan: acidosis. Annals of internal medicine, 1980, 93(2):280–1. five-year experience. Eastern Mediterranean health journal, 2000, 6(5–6):1035–8. 21. Kemperman FA et al. The influence of ketoacids on plasma 17. Wallace TM, Matthews DR. Recent advances in the monitoring creatinine assays in diabetic ketoacidosis. Journal of internal and management of diabetic ketoacidosis. Quarterly journal of medicine, 2000, 248(6):511–7. medicine, 2004, 97(12):773–80. 22. Diabetes data and trends. Centers for Disease Control and Pre- 18. Umpierrez GE, Khajavi M, Kitabchi AE. Review: diabetic vention [website] (http://www.cdc.gov/Diabetes/statistics/ ketoacidosis and hyperglycemic hyperosmolar nonke- dkafirst/fig2.htm, accessed 5 September 2009).

Mortality and burden of disease attributable to selected major risks

Mortality and burden of disease attributable to selected major risks uses a comprehensive framework for studying health risks that was developed for the World Health Report 2002, which presented estimates for the year 2000. The report provides an update for the year 2004 for 24 global risk factors. It uses updated information from WHO programmes and scientific studies for both exposure data and the causal associations of risk exposure to disease and injury outcomes. The burden of disease attributable to risk factors is measured in terms of lost years of healthy life using the metric of the disability-adjusted life year (DALY). The DALY combines years of life lost due to premature death with years of healthy life lost due to illness and disability.

Health risks are in transition: populations are ageing owing to successes against infectious diseases; at the same time, patterns of physical activity and food, alcohol and tobacco consumption are changing. Low- and middle-income countries now face a double burden of increasing chronic, noncommunicable conditions, as well as the communicable diseases that traditionally affect the poor.

Further information about this and other WHO publications can be found at: http://www.who.int/publications/en/

299 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Efficacy of metoclopramide and dexamethasone for postoperative nausea and vomiting: a double-blind clinical trial M. Entezariasl,1 M. Khoshbaten,2 K. Isazadehfar 1 and G. Akhavanakbari 1

نجاعة )كفاءة( امليتوكلوبراميد وديكساميثازون يف معاجلة الغثيان والقيء تلو العمليات اجلراحية: دراسة رسيرية )إكلينيكية( مزدوجة التعمية مسعود انتظاري أصل، منوجهر خوش باطن، خاطره عيسى زاده فر، قدرت اخوان أكربي

اخلالصـة:يشيع حدوث الغثيان والقيء يف مضاعفات التخدير. ِّم ويقيالباحثون يف هذه الدراسة الرسيرية )اإلكلينيكية( املزدوجة التعمية، معدل حدوث الغثيان والقيء، بعد إجراء جراحة لقدح الساد )الكتاراكت( مع التخدير الوريدي لدى مئة مريض عينهم الباحثون ًعشوائيا إلعطائهم دواء عقل placebo )حملول ملحي( عند التحضري للتخدير، أو إلعطائهم 10مييل غرام من امليتوكلوبراميد أو 8 مييل غرام من الديكساميثازون أو الدوائني ً. معاوقد بلغ معدل حدوث الغثيان يف غرفة اإلفاقة لدى من أعطي الدواء العقل 44%، ولدى من أعطي امليتوكلوبراميد 20%، ولدى من أعطي الديكساميثازون 16% ولدى من أعطي الدوائني ًمعا 8%. أما معدل حدوث القيء فكان لدى من أعطي الدواء العقل 20%، ولدى من أعطي امليتوكلوبراميد 4%، ولدى من أعطي الديكساميثازون 4% ولدى من أعطي الدوائني ًمعا 0%. وهكذا اتضح أن إعطاء توليفة امليتوكلوبراميد والديكساميثازون ُي ِنقص بمقدار ُي ْع َت ُّد به ًإحصائيا من حدوث الغثيان والقيء يف غرفة اإلفاقة وبعد 24 ساعة من ذلك، لذا يويص باستعامل هذه التوليفة لدى املجموعة املعرضة خلطر مرتفع َّوالسياميف اجلراحات للمرىض اخلارجيني.

ABSTRACT Postoperative nausea and vomiting are common complications of anaesthesia. This double-blind clinical trial assessed the incidence of nausea and vomiting after cataract surgery with intravenous anaesthesia in 100 patients randomly assigned to preinduction placebo (saline), metoclopramide (10 mg), dexamethasone (8 mg) or the 2 drugs combined. The incidence of nausea in the recovery room was 44% with placebo, 20% with metoclopramide, 16% with dexamethasone and 8% with the combination. The incidence of vomiting was 20%, 4%, 4% and 0% respectively in the 4 groups. Metoclopramide plus dexamethasone combination significantly decreased nausea and vomiting both in the recovery room and 24 hours afterwards and is recommended for high-risk groups, especially in outpatient surgeries.

Efficacité du métoclopramide et de la dexaméthasone sur les nausées et vomissements postopératoires : une étude clinique en double aveugle

RÉSUMÉ Les nausées et vomissements postopératoires sont des complications courantes de l’anesthésie. Cette étude clinique en double aveugle évaluait l’incidence des nausées et des vomissements après une chirurgie de la cataracte avec anesthésie par voie intraveineuse chez 100 patients auxquels un placebo (eau saline), du métoclopramide (10 mg), de la dexaméthasone (8 mg) ou les deux médicaments associés avaient été administrés de manière randomisée avant l’opération. L’incidence des nausées en salle de réveil était de 44 % avec le placebo, 20 % avec le métoclopramide, 16 % avec la dexaméthasone et 8 % avec l’association des deux médicaments. L’incidence des vomissements était de 20 %, 4 %, 4 % et 0 %, respectivement, dans les quatre groupes. L’association du métoclopramide et de la dexaméthasone a considérablement réduit les nausées et vomissements, non seulement en salle de réveil mais également dans les 24 heures qui suivent, et est recommandée pour les groupes à haut risque, notamment pour les patients opérés en ambulatoire.

1Department of Anaesthesiology, Ardebil University of Medical Sciences, Ardebil, Islamic Republic of Iran. 2Drug Applied Research Centre, Tabriz, Islamic Republic of Iran (Correspondence to M. Khoshbaten: [email protected]). Received: 13/01/08; accepted: 28/02/08

300 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Introduction The sample size was calculated based on Pain intensity was assessed using a 10 probability of reducing PONV by 40%, cm visual analogue scale (0 = no pain to Postoperative nausea and vomiting with α = 0.05 and β = 20%. The baseline 10 = most severe pain). Since pain after (PONV) is one of the most common exclusion criteria were: suffering from cataract surgery is relatively slight, pa- complications of anaesthesia [1]. De- diabetes mellitus or other underlying tients did not receive further analgesic spite significant improvements in thera- disorders, use of antiemetics during the treatment after discharge. Any patient peutics and anaesthesia techniques, the past 24 hours and a positive history of experiencing moderate to severe pain rate of this complication still remains motion sickness or PONV. (score ≥ 5) would receive 50 mg pethi- high. PONV afflicts around one-third Patients were enrolled 1 hour before dine and be ineligible to finish the study; of patients undergoing operation with going to the operating room and only if however no cases were reported. general anaesthesia [2], and can lead to they agreed to participate in the study af- The data were analysed using SPSS, wound dehiscence, bleeding, aspiration ter it had been fully explained them and version 13.0. A series of 1-way analyses of gastric contents, water and electrolyte they signed an informed consent form. of variance were conducted to exam- abnormalities, prolonged hospitaliza- This was prepared by the ethical com- ine differences between the 4 groups tion, rehospitalization and patients’ low mittee of Ardebil University of Medical with respect to parametric variables. If Sciences and allowed participants to satisfaction with postoperative care [3]. a significant difference was found, the withdraw at any point in the study no It is so unpleasant that most patients Bonferroni t-test was used to detect explanation and without jeopardizing would prefer to experience postopera- intergroup differences. Categorical vari- the quality of care they received. tive pain rather than nausea and vomit- ables were analysed by using a series of ing [4]. The patients were randomly as- 4 × 2 chi-squared tests to determine the signed by block randomization to one Occurrence of PONV is influenced differences between the 3 treatment of 4 groups of 25 patients: placebo (2 by various factors such as duration of groups versus placebo. anaesthesia, type of surgery, use of nar- mL normal saline); metoclopramide cotics, patient’s age and sex, and history (10 mg); dexamethasone (8 mg); or of PONV, motion sickness or smoking metoclopramide plus dexamethasone Results [4]. Cataract surgery, however, has not (10 mg/8 mg). All the treatments were been reported to increase postoperative given intravenously 1 minute before A total of 100 patients were included nausea and vomiting. Evidence shows induction of anaesthesia by injection in the final analysis. There were no sig- that PONV may be prevented by using of fentanyl (2 μg/kg), propofol (2 mg/ nificant differences between the back- prophylactic injection of certain drugs kg) and atracurium (0.5 mg/kg). They ground characteristics of patients in the [5]. Dexamethasone and metoclopra- were given by the anaesthetist who was 4 groups in terms of age, sex, duration unaware of the drug allocation. Follow- mide are commonly used separately of surgery or history of smoking (Table ing tracheal intubation, anaesthesia was for this purpose, and most studies have 1). demonstrated their efficacy in decreas- continued with an infusion of propofol (50 μg/kg/min) while the patient The incidence of nausea and vomiting PONV [6–8]. In the current study, ing in the recovery room and 24 hours the effect of preoperative injection of breathed a combination of O2 (50%) after the operation is shown in Table 2. dexamethasone, metoclopramide and and N2O (50%). Local anaesthetics Patients in the metoclopramide, dexam- their combination on PONV was in- were not applied during surgery. ethasone and combination groups had vestigated. The duration of surgery was almost the same (i.e. range of 35–40 minutes) a lower incidence of nausea (20%, 16% for all patients. Following surgery, pa- and 8% respectively) in the recovery Methods tients were transferred to the recovery room when compared with the placebo room, where the incidence of PONV group (44%). The incidence of vomiting The study was a double-blind clinical and the frequency of those needing in the recovery room was 4%, 4% and trial. The study group was 100 patients treatment for PONV were evaluated by 0% respectively compared with 20% referred to Alavi Hospital in Ardebil nurses, who were blind to the patient’s in the placebo group. Although both (northwest of Islamic Republic of Iran) group as the drug administered was metoclopramide and dexamethasone for cataract surgery under general anaes- not recorded on the patient’s anaes- separately were effective in reducing thesia who met the American Society of thesia chart. The occurrence of PONV the symptoms of PONV, the difference Anesthesiologists’ physical status classi- was also recorded 24 hours following was only statistically significant for the fication class 1 (normal healthy patient) surgery. Nausea and vomiting were re- combination. Similar results were found or 2 (patient with mild systemic disease). corded by nurses as present or absent. after 24 hours (Table 2).

301 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Table 1 Age, sex, duration of surgery and history of smoking in the 4 patient groups Variable Placebo (n = 25) Metoclopramide Dexamethasone Metoclopramide + P-valuea (n = 25) (n = 25) dexamethasone (n = 25) Age [mean (SD) years]b 67.1 (9.3) 61.5 (10.1) 65.9 (11.4) 71 (5.9) 0.80 Sex [no. (%)] Female 13( 52) 10 (40) 11 (44) 14 (56) 0.70 Male 12 (48) 15 (60) 14 (56) 11 (44) Duration of surgery 36.2 (2.3) 38.5 (1.8) 35.4 (1.1) 37.2 (2.4) 0.65 [mean (SD) min] History of smoking 3 (12) 5 (20) 3 (12) 3 (12) 0.09 [no. (%)]

aChi-squared test and ANOVA; bRange 50 to 80 years. SD = standard deviation.

There was no significant difference of surgery, causing delays in patient dis- underlines the beneficial effect of com- in the incidence of nausea and vomit- charge from hospital, especially in out- bined use of these 2 drugs. ing between males and females in the patient surgeries. Therefore, therapeutic Some previous placebo-controlled 4 study groups (data not shown, P = strategies preventing this complication studies have also demonstrated the 0.76). are of utmost importance [9]. Replacing preventive effect of preoperative use of The intensity of postoperative pain nitrous oxide with intravenous propofol dexamethasone (especially in combina- was relatively minor, with patients in the has reduced, but not eliminated, the tion with metoclopramide) on PONV. 4 groups reporting similar low/median incidence of PONV [10,11]. In the In 204 patients undergoing lumbar disc pain scores: placebo 1.7; metoclopra- current study, the preventive effect of surgery, Wallenborn et al. showed that mide 2.3; dexamethasone 2.4; combina- metoclopramide, dexamethasone and while the incidence of nausea and vom- tion 1.9. Furthermore, when compared their combination in decreasing PONV iting was 35.8% in the placebo group, the with the placebo group, no significant in elderly people undergoing cataract preoperative use of combined metoclo- difference was observed in any of the surgery was investigated. The results pramide and dexamethasone decreased 3 treatment groups (i.e. metoclopra- of this study showed that although pa- the incidence of postoperative nausea mide, dexamethasone or combination) tients given metoclopramide or dexam- to 10% and vomiting to 3% [12]. In an- in terms of the frequency of need for ethasone alone had a lower incidence of other study by Feo et al. on laparoscopic postoperative analgesics. PONV, the effect was not statistically surgery patients, the incidence of nausea significant whereas their combined use and vomiting was 46% in the placebo had a significant effect. The incidence of group and 14% in the dexamethasone Discussion nausea with metoclopramide plus dex- group [13]. Laiq et al., in a study on amethasone was 8% versus 44% with 100 patients undergoing gynaeco- PONV is still among the most com- placebo and the incidence of vomiting logical surgery, found that nausea and mon and troublesome complications was 0% versus 20% with placebo. This vomiting decreased from 30% and 24%

Table 2 Incidence of nausea and vomiting in the recovery room, and after 24 hours in the 4 patient groups Group Placebo (n = Metoclopramide (n = 25) Dexamethasone (n = 25) Metoclopramide + 25) dexamethasone (n = 25) % % P-valuea % P-valuea % P-valuea In recovery room Nausea 44 20 0.72 16 0.65 8 0.01 Vomiting 20 4 0.55 4 0.55 0 0.02 After 24 hours Nausea 52 20 0.68 20 0.68 8 0.008 Vomiting 20 4 0.55 4 0.55 0 0.02

aCompared with the placebo group by the chi-squared test.

302 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

respectively with placebo to 20% and 6% PONV incidence was not significantly in patients undergoing cataract surgery. with dexamethasone [14]. According different between the dexamethasone Third, it was conducted only in older to Huang et al., 5 mg dexamethasone and metoclopramide groups [17]. The patients. However, our study protocol significantly decreased the incidence of difference between our results and was a strong point. PONV from 70% to 28% [15], while theirs was probably due to the time of In conclusion, although prophylac- Wang et al. showed dexamethasone de- drug injection, since its effect on PONV tic injection of 10 mg metoclopramide creased nausea and vomiting from 63% might be diminished due to the short or 8 mg dexamethasone separately to 23% in patients having laparoscopic half-life of the drug. The limited number can decrease the incidence of PONV, cholecystectomy [16]. of patients in Chekman’s study was their the combined use of these drugs has a In contrast, Chekman et al., compar- main limitation. more marked and significant effect. In ing the effect of dexamethasone and Our study has some limitations view of the low cost of these drugs, their metoclopramide (injected 10 minutes that might have influenced the find- combined prophylactic use is recom- before anaesthesia) in 45 patients ings. First, our sample size was small. mended in the groups at risk for PONV, divided into 3 groups, showed that Second, the study was conducted only especially in outpatient surgeries.

References

1. Golembiewski J, Gernin E, Chopra T. Prevention and treatment lational agents. European journal of anaesthesiology, 1998, of postoperative nausea and vomiting. American journal of 15(4):433–45. health-system pharmacy, 2005, 62(12):1247–60. 11. Yavosgo R et al. Propofol-nitrous oxide versus sevoflurane– 2. Apfel CC et al. A factorial trial of six interventions for the pre- nitrous oxide for strabismus surgery in children. Paediatric vention of postoperative nausea and vomiting. New England anaesthesia, 1999, 9:495–9. journal of medicine, 2004, 350(24):2441–51. 12. Wallenborn J et al. Metoclopramid und Dexamethason zur 3. Ku CM, Ong BC. Postoperative nausea and vomiting: a re- Prophylaxe von postoperativer Ubelkeit und Erbrechen nach view of current literature. Singapore medical journal, 2003, balancierter Anasthesie [Metoclopramide and dexametha- 44(7):366–74. sone in prevention of postoperative nausea and vomiting after 4. Apfel CC, Roewer N. Postoperative nausea and vomiting. Anes- inhalation anesthesia]. Anästhesiologie, Intensivmedizin, Not- thesia, 2004, 53(4):377–89. fallmedizin, Schmerztherapie, 2003, 38(11):695–704. 5. Tatic M et al. Postoperativna muka i povracanje [Postop- 13. Feo CV et al. Randomized clinical trial of the effect of post- erative nausea and vomiting]. Medicinski pregled, 2003, operative dexamethasone on nausea and vomiting after 56(9–10):431–5. laparoscopic cholecystectomy. British journal of surgery, 2006, 6. Langer R. Postoperative nausea and vomiting. Educational 93(3):295–9. synopses in anaesthesiology and critical care medicine, 1996, 1(3) 14. Laiq N et al. Dexamethasone as antiemetic during gynaecologi- (http://anestit.unipa.it/gta/nausea.html, accessed 29 August cal laparoscopic surgery. Journal of the College of Physicians and 2009). Surgeons—Pakistan, 2005, 15(12):778–81. 7. Habib AS, Gan TJ. Evidence-based management of postopera- 15. Huang JC et al. Low dose dexamethasone effectively pre- tive nausea and vomiting: a review. Canadian journal of anaes- vents postoperative nausea and vomiting after ambulatory thesia, 2004, 51(4):283–5. laparoscopic surgery. Canadian journal of anaesthesia, 2001, 8. Henzi I, Walder B, Tramèr MR. Dexamethasone for the pre- 48(10):973–7. vention of postoperative nausea and vomiting: a quantitative 16. Wang JJ et al. Dexamethasone reduces nausea and vomiting systematic review. Anesthesia and analgesia, 2000, 90:186–91. after laparoscopic cholecystectomy. British journal of anaesthe- 9. Garrett K et al. Managing nausea and vomiting. Critical care sia, 1999, 83(5):772–5. nurse, 2003, 23:31–50. 17. Ekmen N et al. Comparison of the effects of dexamethasone 10. Sneyd JR et al. A meta-analysis of nausea and vomiting fol- and metoclopramide on postoperative nausea and vomiting. lowing maintenance of anaesthesia with propofol or inha- Erciyes medical journal, 2003, 25(3):137–43.

303 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Chlamydia trachomatis and cervical intraepithelial neoplasia in married women in a Middle Eastern community M. Valadan,1 F. Yarandi,1 Z. Eftekhar,1 S. Darvish,1 M.S. Fathollahi 2 and A. Mirsalehian 3

الكالميديا التـراخومية واألورام داخل البطانية يف عنق الرحم لدى املتزوجات يف إحدى املجتمعات الرشق أوسطية مهرناز ولدان، فريبا يارندي، زهرا افتخار، سودابه درويش، حممود فتح اهلي، أكرب مريصاحليان

اخلالصـة:إن الغرض َّاملتوخى من هذه الدراسة هو ُّالتعرفعىل مدى الرتابط بني العدوى الفرجية بالكالميديا وبني األورام دخل البطانية يف عنق الرحم. وقد مجع الباحثون املعطيات ضمن دراسة للحاالت والشواهد شملت 60 مريضة بورم داخل بطاين يف عنق الرحم شخصت إصابتهن باخلزعة، إىل جانب 85إمرأة كان تنظري عنق الرحم واخلزعة منه ً سويالدهين. وقد الحظ الباحثون أن األضداد املصلية للكالميديا التـراخومية تتصاحب مع ازدياد خطر اإلصابة بالورم داخل البطاين يف عنق الرحم )وكان معدل األرجحية 3 .7، وكانت فتـرة الثقة 95%، إذ تراوحت النتائج بني 35.2 1.5 و (. وكان هناك ترابط ُي ْع َت ُّد به ًإحصائيابني وجود أضداد مشتملة للكالميديا التـراخومية وبني الورم داخل البطاين يف عنق الرحم )فكان معدل األرجحية 5.5، وكانت فتـرة الثقة 95%، إذ تراوحت النتائج بني 2.4 و12.4(. وتشري هذه النتائج إىل ترابط قوي بني األورام داخل البطانية يف عنق الرحم وبني التهاب عنق الرحم بالكالميديات.

ABSTRACT The objective of this study was to determine the association between vaginal Chlamydia infection and cervical intraepithelial neoplasia (CIN). Data were collected in a case–control study for 60 patients with CIN in biopsy and 85 control subjects with normal colposcopy and biopsy. Serum antibodies to C. trachomatis were associated with an increased risk for CIN [odds ratio (OR) = 7.3; 95% confidence interval (CI) 1.5–35.2)]. There was also a significant association between presence of inclusion bodies for C. trachomatis and CIN (OR = 5.5; 95% CI 2.4–12.4). These results indicate a strong association between CIN and chlamydial cervicitis.

Les infections à Chlamydia trachomatis et la néoplasie cervicale intraépithéliale chez les femmes mariées d’une communauté du Moyen-Orient

RÉSUMÉ L’objectif de cette étude était de déterminer la relation entre les infections vaginales à Chlamydia et la néoplasie cervicale intraépithéliale (CIN). Des données ont été collectées lors d’une étude cas-témoin portant sur 60 patientes dont la biopsie confirmait une CIN et un groupe témoin de 85 personnes dont la colposcopie et la biopsie étaient normales. Les anticorps sériques anti-C. trachomatis étaient associés à une augmentation du risque de CIN (odds ratio = 7,3 ; intervalle de confiance (IC) 95 % 1,5 – 35,2)]. Une association significative a également été observée entre la présence de corps d’inclusion de C. trachomatis et la CIN (odds ratio= 5,5 ; IC 95 % 2,4 – 12,4). Ces résultats indiquent une association importante entre la CIN et une cervicite à Chlamydia.

1Department of Obstetrics and Gynaecology, Mirza Koochak Khan Hospital; 2Department of Biostatistics; 3Department of Pathology, Tehran University of Medical Sciences, Tehran, Islamic Republic of Iran (Correspondence to M. Valadan:[email protected]). Received: 30/10/07; accepted: 11/02/08

304 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Introduction patients had been referred from other adjusted for possible confounders, with cities, 20 patients were not available. 95% confidence interval, were calcu- Infection with Chlamydia trachomatis, From the 180 remaining, we selected 60 lated. For the statistical analysis SPSS, a highly prevalent sexually transmitted patients with CIN in biopsy and 85 con- version 13.0 for Windows, was used. All agent worldwide, is mostly asympto- trol subjects with normal colposcopy P-values were 2-tailed, with statistical matic (70%–80%) and often remains and biopsy after statistically matching significance defined at P ≤ 0.05. undetected. Besides causing cervicitis for age and social status. We used the and urethritis, infection may result in se- Bethesda system for classification of rious secondary complications such as CIN. Results pelvic inflammatory disease and pelvic Each participant was tested for pain, tubal infertility and ectopic preg- C. trachomatis using 2 methods: 1) an All participants were married and the nancy [1]. In addition, C. trachomatis has immunuofluorescence technique to age range was 20–65 years: 28 in the pa- been suggested to be a cofactor in the measure specific antichlamydial IgG- tient group and 33 in the control group development of cervical cancer [2,3]. antibody in blood samples (titres of were over 40 years old. However, in some studies no associa- ≥ 1/64 were considered positive for The overall prevalence of positive tion was found between C. trachomatis C. trachomatis); and 2) Giemsa staining serum antibodies for C. trachomatis in- and cervical neoplasia [4]. for detecting inclusion bodies charac- fection for cases and controls together The objective of this study was to teristic of C. trachomatis. was 26.2% (Table 1); 45.0% among determine the association between Each participant completed a de- women with CIN and 12.9% among vaginal chlamydial infection and cervi- tailed questionnaire administered by controls, which was a highly significant cal intraepithelial neoplasia (CIN). a trained interviewer covering demo- difference P( < 0.001, unadjusted OR graphic data, general health, obstetrical = 5.5). In addition, the prevalence of and gynaecological history and other inclusion bodies for C. trachomatis in Methods risk factors. All participants gave writ- the CIN group was higher than in the ten informed consent to write their control group (15.0% vs 2.4%, P = 0.005, For this retrospective case–control participation. unadjusted OR = 7.3). Participants study, participants were selected from negative for anti-Chlamydia antibody the total of 609 women visiting the gy- Statistical methods also had a negative evaluation for C. tra- naecology clinic in our hospital between Categorical variables were summarized chomatis inclusion bodies. January 2002 and May 2003 and who by absolute frequencies and percentages Of the CIN group and controls, were referred to the colposcopy clinic at and were compared using chi-squared 76.7% and 51.8% respectively had their Mirza Koochak Khan Hospital, Tehran. or Fisher exact test, as appropriate. In first sexual contact at age ≤ 20 years, Colposcopy showed suspicious/abnor- order to evaluate the association of a statistically significant difference (P mal cervix in 200 women and biopsy C. trachomatis with CIN, both crude = 0.002). In addition, 30.0% of the was taken for them. Since many of the and Mantel–Hansel odds ratio (OR), women in the CIN group and 10.6%

Table 1 Association between CIN with the presence of IgG antibodies to Chlamydia trachomatis or inclusion bodies of C. trachomatis Variable Women with CIN Controls P-value Crude OR (95% CI) (n = 60) (n = 85) No. % No. % Positive C. trachomatis IgG 27 45.0 11 12.9 < 0.001 5.5 (2.4–12.4) Presence of inclusion body 9 15.0 2 2.4 0.005 7.3 (1.5–35.2) First sexual contact < 20 years 46 76.7 44 51.8 0.002 3.1 (1.4–6.3) First pregnancy ≤ 16 years 18 30.0 9 10.6 0.003 3.6 (1.4–8.7) Multiparity 36 60.0 35 41.2 0.02 2.1 (1.1–4.2) History of abortion 31 51.7 25 29.4 0.007 2.6 (1.2–5.1) Used OCP for ≥ 1 year 22 36.7 22 25.9 0.1 1.6 (0.8–3.3) History of genital infection 42 70.0 42 49.4 0.01 2.4 (1.1–4.7)

CIN = cervical intraepithelial neoplasia; OR = odds ratio; CI = confidence interval. OCP = oral contraceptive pills.

305 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

in the control group had their first preg- Table 2 Odds ratio (OR) for Chlamydia trachomatis infection in cases with CIN nancy at age ≤ 16 years (P = 0.003), after adjustment for other risk factors again a statistically significant difference. Risk factor adjusted for: Adjusted OR (95% CI) There was a history of genital tract infec- First sexual contact < 20 years 5.4 (2.3–12.7) tion in 70.0% of cases compared with First pregnancy ≤ 16 years 5.0 (2.1–11.5) 49.4% of controls (P = 0.01). Use of Multiparity 5.1 (2.2–10.9) oral contraceptive pills for ≥ 1 year was History of abortion 4.5 (1.9–10.5) not significantly different between the 2 History of genital infection 5.0 (2.2–11.1)

groups (P = 0.1). CIN = cervical intraepithelial neoplasia; CI =confidence interval. None of the women in this study had a history of cigarette smoking, therefore the association of smoking with CIN controls [5]. Moreover, Koskela et al. • We had limited statistical power relat- was not evaluated. Furthermore, we found that the presence of C. tracho- ed to the number of cases of CINII/ could not assess the relationship be- matis antibodies was associated with an CINIII, leading us to combine them. tween CIN and number of sexual part- increased risk of cervical squamous cell • Lastly, we did not evaluate human ners, because sexual relations outside carcinoma [6]. Wallin et al. reported a papilloma virus (HPV). It is possible marriage are punishable by law in the similar association in a Swedish study that concurrent infection with HPV Islamic Republic of Iran, and this is why [7]. However, there are a number of may exacerbate the effect of Chlamy- our specimen was taken from married studies with opposing results. In a study dia infection on cervical cells. women only. on 128 women with clinical signs of Discrepancies between our results Of the 60 women with CIN, 54 were cervicitis, genital chlamydial infection and those from studies conducted else- diagnosed with CINI, 3 had CINII and did not directly influence the develop- where may be related to differences in 3 had CINIII. The prevalence of serum ment of CIN [4]. the characteristics of the studied popu- antibodies positive for C. trachomatis Our study does provide information lations (such as age, patterns of sexual was 22 (41%) for CINI, 3 (100%) for on this association in relation to this part behaviour or openness in reporting CINII and 2 (66%) for CINIII. of the world. To our knowledge, this is sexual behaviour). The association ofC. trachomatis and one of a few reports on the association Our results show that chlamydial CIN remained significant after adjust- between cervical abnormalities and C. cervicitis is a strong risk factor for CIN. ment for first sexual contact < 20 years, trachomatis infection from the Middle The link between bacterial infections first pregnancy ≤ 16 years, multiparity, East. However, when interpreting the and carcinogens is not clear, but ge- history of abortion and genital infection results, many points need to be borne netic damage and neoplastic changes (Table 2). in mind: can be induced in vitro by co-culturing cells. Release of nitric oxide occurs in • Ours is an exceptional population, in C. trachomatis infections. Recent evi- Discussion which the prevalence of both cervi- dence has also shown that C. tracho- cal abnormalities and C. trachomatis matis inhibits host cell apoptosis; these infection are low. We found a significant association mechanisms could initiate or promote between CIN and the presence of C. • Our sample included a greater cervical carcinogens [8]. This study was trachomatis IgG (crude OR = 5.5) and number of older women compared a preliminary attempt to establish the the presence of inclusion bodies for with the majority of studies published prevalence of cervical abnormalities and C. trachomatis (crude OR = 7.3), which in Western countries. chlamydial infection, and the associa- remained significant after adjusting for • This study only addressed the associ- tion between the 2 in a low-prevalence, other risk factors. Schachter et al. also ation between cervical abnormalities Middle Eastern population. Further reported a significant excess of anti- and C. trachomatis infection, and not research is needed to investigate the bodies against C. trachomatis in women between frank cancers and C. tracho- apparent association between CIN and with cervical neoplasia compared to matis infection. C. trachomatis infection.

References

1. Paavonen J, Eggert-Kruse W. Chlamydia trachomatis: impact 2. Fischer N. Chlamydia trachomatis infection in cervical intra on human reproduction. Human reproduction update, 1999, epithelial neoplasia and invasive carcinoma. European journal 5:433–47. of gynaecological oncology, 2002, 23(3):247–50.

306 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

3. Hakama M et al. Serum antibodies and subsequent cervical 6. Koskela P et al. Chlamydia trachomatis infection as a risk fac- neoplasms: a prospective study with 12 years of follow-up. tor for invasive cervical cancer. International journal of cancer, American journal of epidemiology, 1993, 137:166–70. 2000, 85:35–9. 7. Wallin KL et al. A population-based prospective study of 4. Borisov I, Mainkhard K. The relationship between genital Chlamydia trachomatis infection and cervical carcinoma. Inter- chlamydial infection and the presence of cervical intraepithe- national journal of cancer, 2002, 101(4):371–4. lial neoplasia. Akush ginecol (Sofia), 1995, 34(3):39–40. 8. Anttila T et al. Serotypes of Chlamydia trachomatis and risk for 5. Schachter J et al. Chlamydia trachomatis and cervical neoplasia. development of cervical squamous cell carcinoma. Journal of Journal of the American Medical Association, 1982, 248:2134–8. the American Medical Association, 2001, 285:47–51.

Towards a strategy for cancer control in the Eastern Mediterranean Region

Towards a strategy for cancer control in the Eastern Mediterranean Region was developed in response to the increasing burden of cancer and the need for coordinated action in this regard. This publication reflects a shared commitment to reducing the incidence of cancer and improving the quality of life of those who develop cancer. By promoting an integrated approach to the provision of cancer control activities and services, it is hoped the publication will encourage and assist government and nongovernmental service providers to work more closely together in addressing cancer control:

Towards a strategy for cancer control in the Eastern Mediterranean Region is targeted at government and nongovernmental agencies whose work impacts on the delivery of cancer services and activities, as well as the wide range of individuals involved in the management and delivery of activities and services related to cancer and people affected by cancer. The full text of this publication is freely available at: http://www.emro.who.int/dsaf/dsa1002.pdf

307 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Pregnancy outcome in women with antiphospholipid syndrome on low-dose aspirin and heparin: a retrospective study T. Naru,1 R.S. Khan 1 and R. Ali 1

نتائج احلمل لدى املصابات بمتالزمة مضادات الفوسفولبيد الاليت يعاجلن بجرعة منخفضة من األسربين واهليبارين: دراسة استعادية طاهرة ناور، روزيال صديقة خان، رمحت عيل

هذه اخلالصـة:دراسة استعادية لسجالت املستشفى لتحليل نتائج احلمل بعد معاجلتني خمتلفتني للحوامل املصابات بإسقاط حديث التشخيص مع متالزمة مضادات الفوسفولبيد يف احلمل املشار إليه. وقد شملت الدراسة 64 امرأة، تلقت منهن35 األسربين واهليبارين. وقد اعترب من املضاعفات التي تصيب األمهات واألجنّة كلمن ارتفاع ضغط الدم املحرض باحلمل، واخلداج، ونقص النمو داخل الرحم، وموت الوليد. ومل يكن هناك اختالفات يُعْتَدُّ هبا إحصائياً يف املضاعفات عند األمهات أو أثناء احلمل لدى املجموعات. وباملقابل هناك اختالفات يُعْتَدُّ هبا إحصائياً يف املستويات الوسطية للمضادات من الربوتينات املناعية جي للكارديوليبني. وقد كان إعطاء األسربين لوحده أو بالتشارك مع اهليبارين ذي نجاعة )كفاءة( متساوية لدى املصابات بمتالزمة مضادات الفسفولبيد وإسقاطات متكررة.

ABSTRACT This retrospective review of hospital records analysed pregnancy outcome with 2 different treatments for women with recurrent miscarriage diagnosed with antiphospholipid syndrome in the index pregnancy. Of 64 women, 29 had received aspirin and 35 aspirin plus heparin. Pregnancy-induced hypertension, prematurity, intrauterine growth restriction and neonatal death were considered as maternal and fetal complications. There were no significant differences in antenatal and maternal complications between the groups. However, there were significant differences in mean anticardiolipin IgG antibody levels. Aspirin alone or in combination with heparin was equally efficacious in women with antiphospholipid syndrome and recurrent miscarriage.

Issue de la grossesse de femmes présentant un syndrome des anticorps antiphospholipides traité par l’aspirine à faible dose et par l’héparine : une étude rétrospective

RÉSUMÉ Cette étude rétrospective des dossiers hospitaliers a analysé l’issue de la grossesse chez des femmes ayant fait des fausses couches à répétition, chez lesquelles un syndrome des anticorps antiphospholipides a été diagnostiqué et qui ont reçu deux traitements différents. Sur 64 femmes, 29 d’entre elles avaient reçu de l’aspirine, tandis que 35 étaient traitées avec de l’aspirine et de l’héparine. L’hypertension induite par la grossesse, la prématurité, les retards de croissance intra-utérine et les décès néonatals ont été considérés comme des complications maternelles et fœtales. Aucune différence significative en termes de complications prénatales et maternelles n’a été constatée entre les deux groupes. Cependant, des différences importantes ont été observées dans les taux moyens d’anticorps anticardiolipine IgG. L’aspirine seule possède la même efficacité que l’aspirine associée à l’héparine chez les femmes présentant un syndrome des anticorps antiphospholipides et ayant fait des fausses couches à répétition.

1Department of Obstetrics and Gynaecology, The Aga Khan University, Karachi, Pakistan (Correspondence to R.S. Khan: [email protected]). Received: 07/08/07; accepted: 09/12/07

308 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Introduction APS and recurrent pregnancy loss who had been commenced from the start were treated with aspirin alone or aspi- of the pregnancy either after a positive Antiphospholipid antibody syndrome rin in combination with heparin during pregnancy test or presence of cardiac (APS) is a disorder characterized by the index pregnancy. activity on ultrasound scan until the the presence of medium to high lev- 34th week of gestation, while in group els of lupus anticoagulant antibodies B heparin had been started in the 2nd (LAC) and anticardiolipin antibodies Methods trimester and continued until the 36th week of gestation. (aCL)—the so-called antiphospholi- Sample and setting pid antibodies (aPL). APS was first Most of the women were assessed described in the early 1980s in patients This study was conducted at the Aga before pregnancy when they came to with systemic lupus erythematosus Khan University Hospital in Karachi, the clinic for counselling regarding more [1,2] and is termed primary APS when Pakistan, a tertiary care unit with an than 2 pregnancy losses (intrauterine it occurs in otherwise healthy people average of 3500 deliveries per year. A and/or neonatal deaths). Women were [3]. APS is associated with arterial and dedicated outpatient service caters to monitored fortnightly during the 1st tri- venous thrombosis and both early preg- high-risk obstetric care, early pregnancy mester for hypertension and thrombo- nancy loss and fetal death in advanced complications and investigations in- cytopenia. Fetal monitoring was started cluding Doppler ultrasound facilities in pregnancy [1–4]. While placental from 24–26 weeks with monthly growth the fetomaternal unit. The nursing staff thrombosis and infarction are common scans until delivery. Two-weekly scans provides regular antenatal sessions as findings in aPL-related intrauterine fetal were performed if severe early onset of well as patient education and training deaths [4], thrombosis is not a universal IUGR was suspected along with umbili- for self-administered injections. finding and alternative mechanisms of cal artery Doppler studies. The timing pregnancy loss have been sought [5,6]. The criteria for entry to the study of delivery was assessed individually on were women who suffered 2 or more Prospective clinical studies have the basis of current and past obstetric 1st or 2nd trimester fetal deaths or at confirmed that aPL are risk factors for history and the onset of complications, least 1 intrauterine death or neonatal pregnancy loss, both in patients with e.g. pre-eclampsia, IUGR or abnormal death and 1 of the following laboratory systemic lupus erythematosus [5] and Doppler results. Women whose preg- findings before or during the index preg- in healthy nulliparous women [7]. In nancy reached 37 completed weeks nancy: aCL > 5 GPL [IgG phospholipid women with a history of 3 or more con- gestation were either induced or deliv- units] or > 5 MPL [IgM phospholipid secutive pregnancy losses persistently ered by elective caesarean section. In units]. positive tests for aPL were found in up to women who received heparin during 15%, and in these cases, subsequent fetal Data collection pregnancy, the treatment was used until the end of the hospital stay and low- loss rates of up to 60%–90% were noted The study was a retrospective analysis without specific treatment [8,9]. aPL dose aspirin was continued until 6–8 of women who voluntarily chose differ- weeks postpartum. are also associated with a high incidence ent treatments. From January 1994 to of pre-eclampsia, intrauterine growth December 2003 the data were retrieved Analysis restriction (IUGR), fetal distress and from a data indexing and coding system premature delivery [10,11]. The data were analysed using SPSS, based on the International classification version 10. Independent sample t-tests A number of studies have evaluated of diseases, 9th edition. Of 84 women were used to compare 1st, 2nd and 3rd the efficacy of treatment with low-dose identified, 20 were excluded due to trimester pregnancy losses, neonatal aspirin, prednisolone, unfractionated incomplete documentation and the deaths and status of aPL. Chi-squared low-molecular weight heparin and most remaining 64 women were divided into and Fisher exact tests were used for recently intravenous gammaglobulin, 2 groups: group A (n = 29) had received the association between maternal age, either alone or in various combinations. low-dose aspirin (75 mg once daily gestational age, birth weight, status of However, the findings have not been orally) while group B (n = 35) had re- aPL and pregnancy outcome. A P-value consistent [12,13]. Low-dose aspirin in ceived aspirin (75 mg once daily orally) < 0.05 was considered significant. combination with heparin was demon- plus heparin (5000 IU subcutaneously strated in 2 randomized controlled trials twice daily) (group B) during the in- to lead to a significant improvement in dex pregnancy. The group of patient Results the live birth rate [14,15]. on aspirin treatment alone were those This study aimed to determine the who had refused heparin, as injections A total of 64 women with APS and pregnancy outcome in women with were not acceptable to them. Aspirin recurrent miscarriage were included in

309 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Table 1 Age and pregnanacy characteristics of women treated with aspirin or aspirin plus heparin (n = 64) Characteristic Group A Group B P-value Aspirin (n = 29) Aspirin + heparin (n = 35) Mean (SD) Mean (SD) Maternal age (years) 28.8 (5.3) 29.6 (5.2) 0.575 1st and 2nd trimester pregnancy loss (No.) 2.28 (1.13) 2.43 (2.10) 0.726 3rd trimester pregnancy loss (No.) 0.21 (0.41) 0.34 (0.39) 0.300 Neonatal death (No.) 0.28 (0.53) 0.20 (0.47) 0.547

SD = standard deviation.

the study: 29 women in group A used the mean gestational ages at delivery combinations, to improve the poor live only low-dose aspirin and 35 women in in groups A and B were 35.9 (SD 4.1) birth rates among these women, with group B received low-dose aspirin plus months and 35.6 (SD 3.6) months re- live births reported from 30% to 100% heparin. Tables 1 and 2 show the demo- spectively. The mean birth weights in of pregnancies [1]. However, treatment graphic details and antiphospholipid groups A and B were 2.47 (SD 0.85) of pregnant aPL-positive women to antibody status of the women. There kg and 2.36 (SD 0.77) kg respectively. improve pregnancy outcome remains were no significant differences between There were 6 neonatal deaths (20.7%) completely empirical. The paucity of the groups with respect to maternal age, in group A and 3 (8.3%) in group B. One data from large well-designed trials of number of previous 1st, 2nd and 3rd baby in group A died due to congenital different management options in com- trimester pregnancy losses and neonatal heart disease (Table 3). None of the parable groups of pregnant women with deaths. women developed a thromboembolic this complex disease contributes to the There was a significant difference in or cerebrovascular event. difficulty in formulating recommenda- the mean concentrations of aCL IgG tions for managing these pregnancies. titres between group A and group B The approach is guided by observa- [8.88 (SD 7.17) GPL versus 15.88 (SD Discussion tional studies, personal knowledge and 13.77) GPL] (P = 0.016) (Table 2). the opinions of others experienced in The mean concentrations of aCL IgM APS is widely recognized as a risk factor the management of such women. titres were not significantly different for numerous obstetric complications, Our study investigated women with between the groups [10.73 (SD 7.19) including recurrent miscarriage, IUGR, APS in the index pregnancy treated MPL versus 12.30 (SD 7.61) MPL] (P pre-eclampsia, fetal death and preterm with either aspirin alone or aspirin and = 0.403). labour. Since its original description, heparin together. Analysis of the data There were no significant differ- APS has emerged as the most impor- did not reveal any statistically significant ences between groups A and B in the tant treatable cause of recurrent mis- difference between the treatment groups rates of pre-eclampsia [6/29 (20.7%) carriage [2]. The prevalence of aPL in in terms of live birth rate or the antenatal versus 10/35 (28.6%)], preterm births recurrent miscarriage is approximately complication rates. Preterm deliveries [9/29 (31.0%) versus 12/35 (34.3%)] 15.5% [9,16]. The risk of subsequent were experienced by 31.0% of women in and IUGR [5/29 (17.2%) versus 8/35 pregnancy loss in women with aPL and group A compared with 34.3% in group (22.9%)] (Table 3). previous pregnancy loss is unknown but B, while 17.2% and 22.9% of women suf- The mean gestational age and the probably exceeds 60% [10]. fered IUGR in groups A and B respec- neonatal birth weight were not signifi- A variety of treatment regimens tively. Similarly, severe pre-eclampsia cantly different between the groups; have been used, both single agents and was observed in 20.7% and 28.6% of

Table 2 Antiphospholipid antibody levels of women treated with aspirin or aspirin plus heparin (n = 64) Antibody levels Group A Group B P-value Aspirin (n = 29) Aspirin + heparin (n = 35) Mean (SD) Mean (SD) Anticardiolipin IgG (GPL) 8.88 (7.17) 15.88 (13.77) 0.016 Cardiolipin IgM (MPL) 10.73 (7.19) 12.30 (7.61) 0.403

SD = standard deviation; GPL = IgG phospholipid units; MPL = IgM phospholipid units.

310 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 3 Maternal and fetal outcomes of index pregnancies of women treated with aspirin or aspirin plus heparin Maternal and fetal outcome Group A Group B P-value Aspirin (n = 29) Aspirin + heparin (n = 35) Pre-eclampsia (No.) 6 10 0.568 Preterm deliverya (No.) 9 12 0.498 Small-for-gestational age baby (No.) 5 8 0.757 Gestational age [Mean (SD) weeks] 35.9 (4.2) 35.6 (3.6) 0.720 Birth weight [Mean (SD) g] 2.5 (0.9) 2.4 (0.8) 0.581 Neonatal death (No.) Yes 6b 3 0.152 No 23 32

aAt 26–37 weeks. b1 baby had congenital heart disease. SD = standard deviation.

the women respectively. These results Kutteh excluded women with LAC 14[ ], in the IgG values between the aspirin are comparable to international data whereas in the study of Rai et al., 80% of group and the aspirin plus heparin [11,12,16,17]. the patients had LAC in the absence of group (P = 0.016). The maternal and Stone et al. analysed the outcome aCL [15]. Meanwhile, Farquharson et al. fetal complications in the 2 groups were of pregnancies after introduction of a found the birth rate to be similar in both similar. We cannot exclude the pos- standard protocol in 33 patients with groups (72% with aspirin alone compared sibility of a better outcome in the aspirin APS and achieved a live birth rate of with 78% when heparin was added to the and heparin group due to the addition 91% in women with primary APS even regimen) [20]. The uncertainty as to of heparin. The persistent presence of in those with significant past pregnancy whether heparin is always needed comes IgG and IgM aCL needs to be clini- morbidity with or without thrombosis. from this uncontrolled prospective study cally monitored during pregnancy for with 91% live births with low-dose aspirin The outcome in the index pregnancy optimal outcome. This may have some alone in aPL-positive pregnant women, was influenced the most by past preg- implications for further study and coun- who had a minimum of 2 spontaneous nancy outcome [18]. Branch and Kha- selling of these women. mashta pointed out that treatment of abortions and a history of only 6% live As complications still occur and APS must be individualized and related births [18]. Differences in the inclusion to previous pregnancy complications and exclusion criteria used in different the exact pathogenetic mechanism of [19]. studies are a major cause of such discord- these events in APS remains unclear, further randomized trials are required A prospective study of aPL-positive ant findings [21]. pregnant women with at least 3 sponta- Our study suggests that the women to determine the optimum regime for neous consecutive miscarriages recorded with aPL and recurrent pregnancy loss the management of APS pregnancies. 44% of live births in women assigned to can be treated effectively with low-dose Among combination therapies, aspirin low-dose aspirin alone and 80% for those aspirin alone. As this study did not in- together with heparin is currently one treated with aspirin plus subcutaneous clude untreated controls, we cannot of the best treatments offered, with the heparin twice daily [14]. A comparable exclude the possibility that aspects of fewest adverse drug effects. The pos- trial also found aspirin alone inferior to obstetric care other than the treatment sibility of adverse effects with heparin aspirin plus heparin (42% versus 71% live per se influenced pregnancy outcome. therapy for a longer duration, and the births) [15]. In view of the heterogene- An important finding in this study cost implications for low resource areas ous definition of aPL in the literature, was the statistical significant difference needs consideration.

References

1. Petri M. Pathogenesis and treatment of the antiphospholi- syndrome: report of an international workshop. Arthritis and pid antibody syndrome. Medical clinics of North America, 1997, rheumatism, 1999, 42:1309–11. 81:151–77. 3. Asherson RA et al. The “primary” antiphospholipid syndrome: 2. Wilson WA et al. International consensus statement on pre- major clinical and serological features. Medicine (Baltimore), liminary classification criteria for definite antiphospholipid 1989, 68:366–74.

311 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

4. Out HJ et al. Histopathological findings in placentae from 13. Empson M et al. Recurrent pregnancy loss with antiphospholi- patients with intra-uterine fetal death and antiphospholipid pid antibody: a systemic review of therapeutic trials. Obstetrics antibodies. European journal of obstetrics and gynecology, 1991, and gynecology, 2002, 99:135–44. 41:179–86. 14. Kutteh WH. Antiphospholipid antibody associated recurrent 5. Di Simone N et al. Antiphospholipid antibodies regulate the pregnancy loss: treatment with heparin and low dose aspirin is expression of trophoblast cell adhesion molecules. Fertility and superior to low dose aspirin alone. American journal of obstet- sterility, 2002, 77:805–11. rics and gynaecology, 1996, 174:1584–9. 6. Di Simone N et al. In vitro effect of antiphospholipid antibody- 15. Rai R et al. Randomised controlled trial of aspirin plus heparin containing sera on basal and gonadotrophin releasing hor- in pregnant women with recurrent miscarriage associated with mone dependent human chronic gonadotrophin released by antiphospholipid antibodies. British medical journal, 1997, cultured trophoblast cells. Placenta, 1995, 16:75–83. 314:253–7. 7. Love PE, Santoro SA. Antiphospholipid antibodies: anti- 16. Backos M et al. Pregnancy complications in women with recur- cardiolipin and the lupus anticoagulant in systemic lupus rent miscarriage associated with aPL treated with low dose as- erythematosus (SLE) and in non-SLE disorders. Prevalence pirin and heparin. British journal of obstetrics and gynaecology, and clinical significance. Annals of internal medicine, 1990, 1999, 106:102–7. 112:682–98. 17. Vinatier D et al. Antiphospholipid syndrome and recurrent 8. Lynch A et al. Antiphospholipid antibodies in predicting miscarriages. European journal of obstetrics, gynecology, and adverse pregnancy outcome. A prospective study. Annals of reproductive biology, 2001, 96:37–50. internal medicine, 1994, 120:470–5. 18. Stone S et al. Primary antiphospholipid syndrome in preg- 9. Rai RS et al. High prospective fetal loss rate in untreated pregnancy: an analysis of outcome in a cohort of 33 women treated nancies of women with recurrent miscarriage and antiphos- with a rigorous protocol. Journal of thrombosis and haemostasis, pholipid antibodies. Human reproduction, 1995, 10:3301–4. 2005, 3(2):240–2. 10. Branch DW et al. Outcome of treated pregnancies in women 19. Branch DW, Khamashta MA. Antiphospholipid syndrome: with antiphospholipid syndrome: an update of the Utah expe- obstetric diagnosis, management and controversies. Obstetrics rience. Obstetrics and gynecology, 1992, 80:614–20. and gynecology, 2003, 101(6):1333–44. 11. Lima F et al. A study of sixty pregnancies in patients with the 20. Farquharson RG, Quenby S, Greaves M. Antiphospholipid antiphospholipid syndrome. Clinical and experimental rheuma- syndrome in pregnancy: a randomized, controlled trial of treat- tology, 1996, 14:131–6. ment. Obstetrics and gynecology, 2002, 100:408–13. 12. Stone S, Poston L. Antiphospholipid antibody syndrome in 21. Balasch J et al. Low-dose aspirin for prevention of pregnancy pregnancy: onset to outcome. Fetal and maternal medicine losses in women with primary antiphospholipid syndrome. review, 2004, 15:273–97. Human reproduction, 1993, 8:2234–9.

Mental health aspects of women’s reproductive health. A global review of the literature

This book has reviewed the research undertaken on a broad range of reproductive health issues and their mental health determinants/consequences over the last 15 years from both high- and low-income countries. Evidence from peer- reviewed journals has been used wherever possible but has been augmented with results of a specific survey initiated to gather state of the art information on reproductive and mental health issues from a variety of researchers and interested parties. Valuable data from consultant reports, national programme evaluations and postgraduate research work was also compiled, analyzed and synthesized.

Further information about this and other WHO publications can be found at: http://www.who.int/publications/en/

312 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Accuracy of ultrasound, clinical and maternal estimates of birth weight in term women T. Ashrafganjooei,1 T. Naderi,2 B. Eshrati 3 and N. Babapoor 2

دقة التقديرات باملوجات فوق الصوتية والتقديرات الرسيرية )اإلكلينيكية(، وتقديرات األمهات لوزن ولداهنن يف متام احلمل طاهرة أرشفكنجويي، طيبه نادري، بابك عرشيت، نوشني بابابور إن اخلالصـة:التقدير الدقيق لوزن الوليد وقت والدته أثناء احلمل مفيد يف تدبري املخاض والوالدة. ويقارن الباحثون يف هذه الدراسة بني دقة التقديرات باألمواج فوق الصوتية والتقديرات الرسيرية )اإلكلينيكية( وتقديرات 264من األمهات لوزن ولداهنن يف متام محلهن بجنني واحد، وقد أدخلن املستشفى ًمتهيدا إلجراء العملية القيرصية. وبلغت حساسية التنبؤ بوزن الوليد يف متام احلمل باستخدام املوجات فوق الصوتية 12.6% ونوعيته 92.1% فيام بلغت حساسيته باستخدام الفحص الرسيري باجلس 11.8% ونوعيته 99.6%، وبلغت حساسية تقديرات األمهات 6.3% ونوعيتها %98.0. إندقة تقديرات األطباء لوزن الوليد يف متام احلمل متاثل دقة تقديره باستخدام املوجات فوق الصوتية يف األسبوع الذي يسبق الوالدة. وتقديرات احلوامل لوزن الوليد يف متام احلمل أكثر دقة من كل من التقديرات الرسيرية )اإلكلينيكية( والتقديرات باملوجات فوق الصوتية.

ABSTRACT Accurate prenatal estimation of birth weight is useful in the management of labour and delivery. This study compared the accuracy of ultrasound, clinical and maternal estimates of fetal weight in 246 parous women with singleton, term pregnancies admitted for scheduled caesarean section. The sensitivity and specificity of predicting birth weight by ultrasound measures were 12.6% and 92.1%, by clinical palpation were 11.8% and 99.6% and by maternal estimate were 6.3% and 98.0% respectively. Clinicians’ estimates of birth weight in term pregnancy were as accurate as routine ultrasound estimation in the week before delivery. Parous women’s estimates of birth weight were more accurate than either clinical or ultrasound estimation.

Précision de l’estimation échographique, clinique et maternelle du poids du bébé à la naissance chez des femmes enceintes à terme

RÉSUMÉ L’estimation anténatale du poids à la naissance est utile au niveau de la gestion du travail et de l’accouchement. Cette étude a comparé la précision de l’estimation échographique, clinique et maternelle du poids du fœtus chez 246 femmes pares, enceintes d’un seul enfant, et ayant mené leur grossesse à terme, admises pour une césarienne programmée. La sensibilité et la spécificité de la prédiction du poids à la naissance au moyen de l’échographie étaient de 12,6 % et 92,1 % respectivement, alors qu’elles étaient de 11,8 % et 99,6 % pour la palpation clinique et de 6,3 % et 98,0 % pour l’estimation maternelle. Pour les grossesses à terme, les estimations du poids à la naissance par les cliniciens étaient aussi précises que les estimations échographiques habituellement réalisées la semaine avant l’accouchement. Les estimations des femmes pares étaient plus précises que celles des cliniciens et de l’échographie.

1Department of Gynaecologic Oncology; 2Department of Obstetrics and Gynaecology, Kerman University of Medical Sciences, Kerman, Islamic Republic of Iran (Correspondence to T. Ashrafganjooei: [email protected]). 3Faculty of Paramedical Sciences, Arak University of Medical Sciences, Arak, Islamic Republic of Iran. Received: 17/08/07; accepted: 04/03/08

313 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Introduction to evaluate the benefits of ultrasound weight, patient demographic data and EFW by comparing the accuracy of actual birth weight were recorded on Accurate prenatal estimation of fetal maternal, clinical and ultrasound EFW data sheets that were kept separate from weight (EFW) in late pregnancy and in term women. In addition, we sought the patient’s chart. labour is extremely useful in the man- to evaluate potential variables that may Neonatal birth weight was consid- agement of labour and delivery, permit- affect the accuracy of the EFW during ered as the gold standard. Correlations ting obstetricians to make decisions labour. were made of the actual birth weight about instrumental vaginal delivery, trial with the ultrasound, clinical and ma- of labour after caesarean delivery and ternal EFW independently. We also elective caesarean section for patients Methods calculated the sensitivity, specificity and suspected of having a macrosomic fetus positive and negative predictive values [1–5]. An accurate diagnosis of mac- Between July 2002 and December for the ultrasound, clinical and maternal rosomia for patients with gestational 2004, this prospective study evaluated EFW compared with actual birth weight. diabetes can reduce perinatal morbidity 3 different methods of EFW—clinical, Normal birth weight was considered as as it may assist the physician and staff maternal and ultrasound measure- 2500–4000 g. We used ROC curve in deciding the appropriate route of ments—on 246 parous women admit- with the chi-squared test to compare delivery to prepare for shoulder dysto- ted for scheduled caesarean section the accuracy of different methods of cia or to prevent a traumatic injury [6]. within 1 week of delivery in the hospital estimation of fetal weight. A P-value Correct EFW values are also important of Kerman University of Medical Sci- of < 0.05 was considered statistically when intrauterine growth is restricted ences, Kerman, Islamic Republic of significant. Data were analysed using and in preterm labour [7,8]. Iran. The sample size was calculated to Medcalc, version 7.4.4.1 and Stata, ver- EFW can be done by mothers (if estimate a sensitivity and specificity of sion 8 software. they are parous), by clinicians using 80% with a precision of 5%. The inclu- Leopold manoeuvres or by ultrasound. sion criteria were: singleton pregnancy In the 1970s, the use of ultrasound to and live-born infant without congenital Results estimate fetal weight gained popular- malformations or hydrops fetalis. ity because of the perceived ability to The mothers were instructed about A total of 246 mothers participated in standardize and reproduce measure- the purpose of the study and gave in- the study. The mean maternal age was ments [3,9], although the technique formed consent for participation. Ap- 27.6 [standard deviation (SD) 5.4] can be challenging, depending on the proval for the study was obtained from years and mean parity was 1.2 (SD 1.2). mother’s physique, uterine anomalies the research ethical committee of the The mean actual birth weight was 3339 or amniotic fluid index 10[ ]. Clinical University. (SD 443) g, while the mean estimated EFW has been shown to accurately pre- Ultrasound EFW was obtained for fetal weights by ultrasound, clinical dict birth weight. For example, Baum all women by the same physician with a assessment and maternal report were et al. showed no significant difference 3.5 MHz transducer (Hitachi EUB-500, 3305 (SD 335) g, 3321 (SD 449) g and between clinical and sonographic es- Tokyo, Japan) using standard Hadlock 3158 (SD 463) g respectively. timates of fetal weight; 64.0% versus reference tables that used biparietal For the clinical EFW, there was 62.5% of the estimates respectively were diameter, abdominal circumference no statistically significant difference in within 10% of the actual birth weight and femur length for calculating fetal the mean estimates comparing the 3 [1]. Maternal EFW is comparable to weight. The physician performing the physicians with different years of ex- both clinical or ultrasound predic- ultrasound was unaware of the 2 other perience: obstetrician with 14 years’ tions in both term and postdate babies estimations. Clinical EFW was obtained experience [3316 (SD 355) g, range [11,12]. Some researchers concluded on the day of operation by palpation 2500–4200], obstetrician with 13 years’ that clinical EFW has higher accuracy using Leopold manoeuvres. A total of 3 experience [3319 (SD 343) g, range than ultrasound EFW [2,13], but other physicians (2 obstetricians and 1 senior 2000–4200g] or senior resident with studies showed that ultrasound EFW resident) examined every woman and 3 years’ experience[3327 (SD 378) g, is more accurate [14] and Chauhan et we took an average of the 3 estimates. 2500–4800 g]. al. showed that the accuracies of both All were unaware of the other clinical The sensitivity values of predicting methods are the same [15]. and ultrasound EFWs. Maternal EFW birth weights for ultrasound, clinical and Due to difficulties in accessing ultra- was obtained by asking mothers, who maternal EFW were 17.6%, 11.8% and sound equipment in rural areas of the Is- were all parous, to estimate the weight 6.3%, with specificity of 93.5%, 99.6% lamic Republic of Iran, this study aimed of their baby. The 3 estimates of fetal and 98.0%, respectively (Table 1).

314 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 1 Specificity, sensitivity, and positive predictive values for estimation of fetal weight using ultrasound, clinical and maternal estimations Method of estimation Normal weight childrena Abnormal weight children Sensitivity Specificity Positive Correctly Incorrectly Correctly Incorrectly predictive estimated estimated estimated estimated value No. No. No. No. % % % Ultrasound 211 18 3 14 17.6 93.5 14.3 Clinical 228 1 2 15 11.8 99.6 67.0 Maternal 200 4 1 15 6.3 98.0 20.0

aNormal birth weight was considered as 2500–4000 g.

Figure 1 shows the results of ROC Discussion We found that clinicians’ estimates curve analysis. Using the chi-squared of birth weight in term pregnancy were test it was evident that the accuracy Birth weight is a key variable affecting as accurate as routine ultrasound esti- of the 3 tests were not significantly dif- fetal and neonatal morbidity, particu- mation in the week before delivery. Fur- ferent from each other (P = 0.35). Ac- larly in preterm and small-for-dates thermore, parous women’s estimates of cording to the maximum point of each babies. In addition, it is of value in the birth weight were more accurate than curve of the 3 measures, a new cut-off management of breech presentations, either clinical or ultrasound estimation. point was used for each method of EFW diabetes mellitus, trial of labour, mac- There have been differing results and the sensitivity and specificity were rosomic fetuses and multiple births about the accuracy of the various estimated (Table 2). [16]. methods of estimating fetal weight.

Figure 1 Receiver operating characteristic (ROC) curves for estimation of fetal weight

315 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Table 2 Sensitivity and specificity for estimation of fetal weight using ultrasound, [11,12], whereas others showed that clinical and maternal estimates based on the maximal point of receiver operating clinical and ultrasound EFW were more characteristic (ROC) curves accurate than maternal EFW [20]. In Method of estimation Cut-off point Maximum Maximum ROC area for weight (g) specificity sensitivity our study we found, as previously re- % % ported by O’Reilly-Green and Divon Ultrasound 3625 62.5 81.7 0.76 [27], that mother’s EFW was the best method of fetal weight screening. Clinical 3500 75.0 76.1 0.84 Maternal 3400 62.5 69.5 0.75 The major strength of our study, in contrast to similar studies, is that the number of patients was sufficient to Clinical estimation of fetal weight using ultrasound EFW at term, particularly in ensure statistical validity of the finding abdominal palpation has been shown macrosomic fetuses [25,26]. of no difference between methods of to be accurate to within 500 g in 85% of The advantage of using ultrasound EFW. cases, with more accuracy in the aver- for EFW has been questioned. Baum age, term fetus than in the preterm and et al. concluded that ultrasound offered macrosomic fetus [17–20]. Diase and no advantage over clinical estimates of Conclusion Monga showed that in diabetic women, fetal weight at term [1]. Mother’s esti- Our results are supported by previous neither parity nor maternal weight af- mates should be viewed as equally valid fected the accuracy of any of the birth as clinical estimates, especially in the studies that indicate that ultrasound weight estimates [6]. Humphries et light of the need for realistic, achievable EFW offers no advantage over clini- al. showed that the accuracies of birth standards. They also reported that senior cian’s EFW when performed during late weight estimation, both clinical and ul- resident clinical estimates were superior pregnancy or labour. An EFW should trasound, were still relatively low [16]. to junior resident estimates. In contrast be recorded in the assessment of all Some studies showed the ultrasound to this research, we found that the accu- patients who are at term and again when EFW was the best method for EFW, es- racy of physicians’ estimates was similar they are in labour, with full awareness of pecially in preterm fetuses [2,12,21,22], regardless of their length of experience, the limitations of the methods for mak- but other studies, such as ours, did not which agrees with Ben-Aroya et al. [7]. ing such estimates. This study and oth- conclude any difference between these Two studies reported no difference in ers show that parous mother’s estimates methods [1,13,14,23,24]. Other stud- the accuracy of EFWs by physician’s of fetal weight are accurate and should ies have reported limited accuracy of palpation versus maternal estimation be given due consideration.

References

1. Baum JD, Gussman D, Wirth III JC. Clinical and patient estima- 7. Ben-Aroya Z et al. Effect of ob/gyn residents’ fatigue and train- tion of fetal weight vs. ultrasound estimation. Obstetrical and ing level on the accuracy of fetal weight estimation. Fetal diag- gynecological survey, 2002, 57(9):558–9. nosis and therapy, 2002, 17:177–81. 2. Chauhan SP et al. Limitations of clinical and sonographic 8. Ott WJ. Sonographic diagnosis of fetal growth restriction. Clini- cal obstetrics and gynecology, 2006, 49(2):295–07. estimates of birth weight: experience with 1,034 parturients. Obstetrics and gynecology, 1998, 91:72–7. 9. Ratanasiri T et al. Comparison of the accuracy of ultrasonic fetal weight estimation by using the various equations. Journal of the 3. Kurmanavicius J et al. Ultrasonographic fetal weight estima- Medical Association of Thailand, 2002, 85:962–7. tion: accuracy of formulas and accuracy of examiners by birth 10. Alsulyman OM, Ouzounian JG, Kjos SL. The accuracy of intra- weight from 500 to 5000 g. Journal of perinatal medicine, 2004, partum ultrasonographic fetal weight estimation in diabetic 32:155–61. pregnancies. American journal of obstetrics and gynecology, 4. Ben-Haroush A et al. Accuracy of sonographically estimated 1997, 177:503–6. fetal weight in 840 women with different pregnancy complica- 11. Chauhan SP et al. Clinical estimate of birth-weight in labour: tions prior to induction of labor. Ultrasound in obstetrics and factors influencing its accuracy. Australian and New Zealand gynecology, 2004, 23(2):172–6. journal of obstetrics and gynaecology, 1993, 33(4):371–3. 5. McIntire DD et al. Birth weight in relation to morbidity and 12. Chauhan SP et al. Intrapartum clinical, sonographic and parous patients’ estimates of newborn birth weight. Obstetrics and mortality among newborn infants. New England journal of medi- gynecology, 1992, 79:956–8. cine, 1999, 340(16):1234–8. 13. Chauhan SP et al. Parous patients’ estimate of birth weight in 6. Diase K, Monga M. Maternal estimates of neonatal birth postterm pregnancy. Journal of perinatology, 1995, 15:192–4. weight in diabetic patients. Southern medical journal, 2002, 95(1):92–4.

316 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

14. Chauhan SP et al. Intrapartum prediction of birth weight: clini- 21. Sherman DJ et al. A comparison of clinical and ultrasonic cal versus sonographic estimation based on femur length alone. estimation of fetal weight. Obstetrics and gynecology, 1998, Journal of obstetrics and gynaecology, 1993, 81(51):695–7. 91:212–7. 15. Baum JD, Gussman D, Wirth JC 3rd. Clinical and patient es- 22. Zayed F, Abu-Heija A. A comparison between ultrasound and timation of fetal weight vs. ultrasound estimation. Journal of clinical methods for predicting fetal weight. Journal of obstetrics reproductive medicine, 2002, 47(3):194–8. and gynaecology, 1999, 19(2):159–61. 16. Chauhan SP et al. Intrapartum detection of a macrosomic fetus: 23. Sacks DA, Chen W. Estimating fetal weight in the management clinical versus and sonographic models. Australian & New Zea- of macrosomia. Obstetrical and gynecological survey, 2000, land journal of obstetrics and gynaecology, 1995, 35:266–70. 55(4):229–39. 17. Humphries J et al. Sonographic estimate of birth weight: rela- 24. Nahum GG, Stanislaw H. Ultrasonographic prediction of term tive accuracy of sonographers versus maternal–fetal medicine birth weight: How accurate is it? American journal of obstetrics specialists. Journal of maternal–fetal and neonatal medicine, and gynecology, 2003, 188:566–74. 2002, 11:108–12. 25. Ramon S et al. Clinical versus ultrasound estimation of fetal 18. Chauhan S et al. Estimate of birth weight among post-term weight. Australian and New Zealand journal of obstetrics and pregnancy: clinical versus sonographic. Journal of maternal- gynaecology, 1992, 32(3):196–9. fetal and neonatal medicine, 1994, 3:208–11. 26. Sherman DJ et al. A comparison of clinical and ultrasound 19. Hirata G et al. Ultrasonographic estimation of fetal weight in estimation of fetal weight. Obstetrics and gynecology, 1998, the clinically macrosomic fetus. American journal of obstetrics 91(2):12–7. and gynecology, 1990, 162:238. 27. O’Reilly-Green C, Divon M. Sonographic and clinical methods 20. Patterson R. Estimation of fetal weight during labor. Obstetrics in the diagnosis of macrosomia. Clinical obstetrics and gynecol- and gynecology, 1985, 65:330–2. ogy, 2000, 43(2):309–20.

Women and health: today’s evidence tomorrow’s agenda

This WHO report provides the latest and most comprehensive evidence available to date on women’s specific needs and health challenges over their entire life-course.

Women’s health has long been a concern for WHO but today it has become an urgent priority. This report explains why. Using current data, it takes stock of what we know now about the health of women throughout their lives and across the different regions of the world. Highlighting key issues - some of which are familiar, others that merit far greater attention - the report identifies opportunities for making more rapid progress. It points to areas in which better information - plus policy dialogue at national, regional and international levels - could lead to more effective approaches.

Further information about this and other WHO publications can be found at: http://www.who.int/publications/en/

317 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Road traffic fatalities in Qatar, Jordan and the UAE: estimates using regression analysis and the relationship with economic growth A. Bener,1 S.J. Hussain,2 M.A. Al-Malki,3 M.M. Shotar,4 M.F. Al-Said 5 and K.S. Jadaan 6

الوفيات النامجة عن حوادث املرور يف قطر واألردن واإلمارات العربية املتحدة: تقديرات باستخدام حتليل ّفالتحو والعالقة ّ بالنمواالقتصادي عبد الباري بنر، سيد جعفر حسني، حممد أ. املالكي، منهل شوطر، مغرية السعيد، خري سعيد َجدعان اخلالصـة: تستخدم معادلة سميد smeed عىل نطاق واسع كنموذج للتنبوء بالوفيات النامجة عن حوادث املرور، وقد وجد أن استخدامها غري ٍكاف يف البلدان النامية. ّفطبق الباحثون حتليل التحوف عىل معطيات لسلسلة زمانية حول الوفيات النامجة عن املرور والسكان واملركبات يف اإلمارات العربية 20.9 املتحدة واألردن وقطر، ثم َّف كيالباحثون املعطيات لتـتناسب مع نامذج أسية للتنبؤ بالوفيات، فاستنتجوا أن اخلطأ املطلق الوسطي مقداره % يف قطر و10.9% يف األردن و5.5%يف اإلمارات العربية املتحدة. لقد وجد الباحثون عالقة ّخطية قوية بني الناتج املحيل اإلمجايل وبني معدل الوفيات.

ABSTRACT Smeed’s equation is a widely used model for prediction of traffic fatalities but has been found inadequate for use in developing countries. We applied regression analysis to time-series data on vehicles, population and traffic fatalities in the United Arab Emirates (UAE), Jordan and Qatar. The data were fitted to exponential models for fatality prediction, producing an average absolute error of 20.9% for Qatar, 10.9% for Jordan and 5.5% for the UAE. We found a strong linear relationship between gross domestic product and fatality rate.

Décès par accidents de la circulation au Qatar, en Jordanie et aux Émirats arabes unis : estimation à l’aide de l’analyse de régression et relation avec la croissance économique

RÉSUMÉ L’équation de Smeed est un outil très courant pour prévoir les décès par accident de la circulation, mais s’est révélée inadaptée pour les pays en développement. Nous avons appliqué une analyse de régression aux données de série chronologique des véhicules, de la population et des décès par accident de la circulation aux Émirats arabes unis (EAU), en Jordanie et au Qatar. Les données ont été adaptées aux modèles exponentiels pour les prédictions relatives à la mortalité, générant une erreur moyenne absolue de 20,9 % pour le Qatar, de 10,9 % pour la Jordanie et de 5,5 % pour les EAU. Une forte relation linéaire entre le produit intérieur brut et le taux de mortalité a été observée.

1Department of Public Health, Weill Cornell Medical College of Qatar, Doha, Qatar (Correspondence to A. Bener:[email protected]; [email protected]). 2Healthy Lifestyle Promotion, Violence and Injury Prevention, World Health Organization Regional Office for the Eastern Mediterranean, Cairo, Egypt. 3General Administration of Public Security, Traffic and Patrol Department, Ministry of Interior, Doha, Qatar. 4Department of Finance and Economics, College of Business and Economics, University of Qatar, Doha, Qatar. 5National Health Authority, Doha, Qatar. 6Department of Civil Engineering, Faculty of Engineering, Al-Isra Private University, Amman, Jordan. Received: 12/07/07; accepted: 21/11/07

318 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Introduction Australia over a period of 20 years [13]. regression analysis and to compare the The observed fatalities were on average results with those derived using Smeed’s According to a World Health Organiza- 20% higher than estimated. equation for estimating fatalities. Ad- tion (WHO) report, it is estimated that It has long been known that annual ditionally, we aimed to identify the re- 5.8 million deaths are due to injuries, motor vehicle deaths vary in a similar lationship between economic growth making this the third leading cause of pattern to economic activity 7[ ]. A and traffic fatalities and to test the model death in the world [1]. Road traffic acci- biphasic relationship between traffic using the statistics available from the dents are a major cause of all injuries and fatalities and economic development United Arab Emirates (UAE), Jordan are known to be a factor in the majority has been recorded, with fatalities ris- and Qatar. of deaths, hence making road traffic inju- ing for the low income countries and ries (RTIs) the 9th leading cause of dis- falling for the high income countries ability adjusted life years (DALYs) lost [14]. Several cross-sectional studies Methods worldwide [2]. Road traffic fatality rates have been performed to identify the We collected the data on vehicles of a country are known to depend upon association between economic growth registered and fatalities from the traf- factors such as population, the number and traffic fatalities: the results were fic department and the population of motor vehicles in use, the total length more or less contradictory. A World estimate and GDP from the Annual of roads, the population density and Bank study showed that the economic Health Report [20] of Hamad Medi- economic conditions [3,4]. With over development of regions and nations cal Corporation and the Directorate of 1 million killed in car crashes annually, is associated with an increase in the Traffic, Ministry of the Interior, Qatar traffic injuries are projected to become number of injuries and deaths from for the period 1990–2006. the 3rd leading cause of DALYs lost by road traffic crashes15 [ ]. Increased 2020 [5]. In general, the total costs of road safety is related to socioeconomic Applications of Smeed’s road crashes and fatalities are a burden development [1]. These studies show equation that no uniform relationship can be for the country and costs vary from International comparisons of road derived between prosperity and road 2.8% to 5% of the gross national product safety can be very misleading and one of traffic crash mortality. [6,7]. Among developed countries, each the difficulties in this field was deciding year there are up to 500 000 deaths and Qatar experienced a rapid transi- on the rate to be used so that popula- 15 million injuries worldwide as a result tion in socioeconomic status after the tion size and number of motor vehicles of road traffic crashes. This represents discovery of oil in the mid 1990s, with a in a country would not contaminate 1400 fatalities and 41 000 injuries per dramatic rise in the national economy, the comparisons. To overcome this dif- day, and a quarter of these are in the expressed in terms of per capita income: ficulty, Smeed used data for road fatali- most highly developed countries [8]. the gross domestic product (GDP) per ties, vehicles and population for the year Smeed devised a formula that es- capita was estimated at US$ 67 000 in 1938 from 20 countries (the majority timates the road traffic fatalities of a 2007 [16]. This has led to road crashes of which were European) to derive a country by using the population and and the resulting fatalities being regard- rather complicated relationship which the number of registered vehicles of ed as a growing social and economic is expressed by the formula [9]: that country [9,10]. Later, Andreassen problem in Qatar. said that Smeed looked at deaths per It has been predicted that traffic D/V = 0.0003 (V/P)–0.67 vehicle plotted against vehicles per fatalities will be the 6th leading cause where D = number of road fatalities, V = capita and argued that the relationship of death worldwide and the 2nd lead- number of vehicles and P = population. found between these variables was not ing cause of DALYs lost in developing Smeed’s Law is an empirical rule valid, but did help to put crash rates into countries by the year 2020 [5]. The ef- relating traffic fatalities to motor vehicle perspective [11]. Referring to Smeed’s fects of some of the contributing factors registrations and country population. equation, Haight reported that when on traffic fatalities have been studied and Thus annually increasing traffic volume the formula disagrees with observations, correlations to predict fatalities were leads to a decrease in crashes per vehicle. certain authors tended to assume that developed [6,12,15,17–19]. However, It was posited after an analysis of figures the particular area under investigation these produced quite large deviations from a number of countries over several was “safer” or “less safe” than it “ought between the expected and the observed decades. fatalities. to be” [12]. Hampson used Smeed’s Smeed’s formula is expressed as: equation to estimate the number of The objective of this study was to traffic fatalities in New South Wales, estimate road traffic fatalities using D = 0.0003(np2)⅓

319 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

or, weighted per capita: Using regression analysis and data for international comparison should, from Qatar for the period 1990–2004 therefore, incorporate other relevant / = 0.0003 × ³√( / ) D p n p [22], a Smeed-type formula of the fol- variables. where D is annual road deaths, n is lowing form, which is statistically signifi- number of registered vehicles, and p is cant at the 95% level, was derived: Development of fatality population. Smeed showed this relation- prediction models t ship worked for 20 different countries. D/V = 0.0004079 (V/P) Smeed’s law has a plenty of critics who However, the validity of the formula has Where t refers to 95% CI (–0.746 to usually concentrate on the fact that also been disputed by several authors –0.254). the number of vehicles (V) appears [11,19,21,22]. Smeed showed that, al- The constants were restrained to be on both sides of the equation and that though originally derived from data for the same for comparison purposes. there is considerable deviation between 1938, this formula was equally applica- Bener and Ofosu [19] repeated the the expected and actual number of ble to data from 1960–67 or even later analysis done by Smeed for Saudi Ara- road fatalities. The latter is clearer when [10]. Further support for the formula bia, and the following relationship was the equation is applied to developing came from Preston [23] who carried derived: countries. In addition, the above review out a comparison of fatalities reported indicates that different countries might –0.72 in 1977 for 32 countries (developed D/V = 0.00021 (V/P) . need different constants in the Smeed equation, a conclusion suggested by An- and developing) and the expected Recent fatality prediction fatalities as calculated by Smeed’s for- models dreassen [11], Bener and Ofosu [19], mula. Despite a percentage deviation and Jadaan, Khalil and Bener [22] and The universality of the above model of the expected road fatalities from the demonstrated in the results of our study. by Smeed was tested on a number of observed ranged between –83% and These points indicated that a new pre- occasions. Cross-sectional data for 191%, Preston concluded that Smeed’s dictive model for road fatalities needs 32 countries showed that on average to be developed which fits the data of formula was still applicable and fitted the expected fatalities as estimated by developing countries and provides bet- data for 1977, which lends credence to Smeed’s formula exceeded the ob- ter estimates than Smeed’s equation. his suggestion that it represents a con- served by 21% [23]. When applied to sensus of the level of road fatalities that time-series data over 20 years for New countries are prepared to tolerate. South Wales, fatalities predicted us- Results Using methods similar to those used ing Smeed’s formula were on average by Smeed, an analysis of fatality rates in 20% lower than observed [13]. Smeed’s Time-series data of road fatalities, ve- developing countries was carried out by formula has therefore been subject to hicles and population for Qatar, UAE Jacobs and Bardsley [24] in 1977 for recent criticism. and Jordan are shown in Tables 1, 2 and 32 developing countries; the fatalities Given the relatively large deviations 3. It was believed that the data could per 10 000 persons and the number of between observed and expected values be fitted successfully to a regression fatalities per 10 000 vehicles were calcu- Andreassen reported “Smeed’s law has model. Two models were developed lated and compared. The relationship is proved to be an imperfect predictive using regression analysis (least squares expressed by the equation: tool,” then concluded that the formula method) and tested. These models had “cannot be extended to predict the –0.43 the following forms: D/V = 0.000742 (V/P) number of fatalities in any year in any which is statistically significant at the country” and “the equation D = con- Model 1 1.0% level. stant. (p)BI (V) B2 is poor” [11]. He D = 299.449 ln (V) –249.095 ln (P) In order to see whether this relation- also concluded that Smeed’s equation ship was stable over time, the analysis was not suitable for the estimation of or was repeated using 1971 data [24] pro- number of fatalities as it produces bi- D = ln (V 299.449 . p–249.095 ) ducing the following relationship: ased estimates which are very sensitive R2 = 0.801 to small variations in data. / )–0.43 D V = 0.000914 (V/P The above points are well under- Model 2 For the same level of vehicle owner- stood since the variation in a number of D = 1311.872 + 384.502 ln (V) – ship, the equation for 1971 gives a 24% countries is not accounted for by the use 420.486 (P) increase in fatality rate over the 1968 of only population and number of vehi- equation. cles. A universal model that can be used or

320 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 1 Estimated fatalities in the United Arab Emirates according to Smeed’s equation and regression analysis, 1990–2004 Year Vehicles Population Fatalities Error (%) Alla Regression Smeed’s Regression Smeed’s analysis estimate equation analysis equation estimate 1990 303 284 1 844 300 394 450 303 14.3 23.1 1991 309 539 1 908 800 490 454 312 –7.3 36.3 1992 344 850 2 011 400 510 486 335 –4.6 34.3 1993 398 788 2 083 100 567 539 360 –5.0 36.5 1994 428 149 2 230 000 600 563 386 –6.2 35.7 1995 437 945 2 377 453 563 568 406 0.8 27.9 1996 442 700 2 477 899 492 569 419 15.6 14.9 1997 468 440 2 624 000 619 590 443 –4.7 28.4 1998 539 407 2 759 000 646 657 480 1.7 25.6 1999 557 668 2 938 000 624 669 507 7.2 18.8 2000 575 929 3 108 000 673 682 532 1.3 21.0 2001 605 696 3 290 000 733 705 561 –3.8 23.4 2002 633 817 3 754 000 743 717 622 –3.5 16.2 2003 661 937 4 041 000 756 736 663 –2.7 12.3 2004 690 058 4 320 000 729 754 703 3.5 3.5

aTotal deaths reported to the Road Traffic Authority.

D = ln (e 1311.872 . V 384.502 . p–420.486 ) more than 80% of the road fatalities than Smeed’s equation. The average R2 = 0.840 in Qatar, the UAE and Jordan can absolute percentage errors were 23.9% where V, D and P are as above. be explained by population and the and 29.6% for Smeed’s equation for The regression output shows that number of vehicles. Comparing the the UAE and Qatar respectively. The all the models (Tables 1–3) are sta- actual and predicted number of fatali- corresponding figure for Jordan, the tistically significant, indicating that ties, all the models gave better estimates population of which is very much larger

Table 2 Estimated fatalities in Jordan according to Smeed’s equation and regression analysis, 1990–2004 Year Vehicles Population Fatalities Error (%) All Regression Smeed’s Regression Smeed’s analysis estimate equation analysis equation estimate 1990 254 617 3 453 000 379 345 434 –12.4 –10.3 1991 259 196 3 888 000 379 398 473 –18.7 3.5 1992 276 301 4 012 000 388 426 493 –16.5 3.3 1993 291 347 4 152 000 440 453 514 –20.1 9.4 1994 304 893 4 200 000 443 469 526 –21.9 12.4 1995 321 373 4 290 100 469 492 543 –12.7 3.6 1996 342 337 4 444 000 552 525 567 6.8 –15.3 1997 362 811 4 600 000 577 559 592 –9.8 4.4 1998 389 196 4 755 800 612 596 619 –7.7 4.1 1999 418 433 4 900 000 676 635 647 1.8 –3.7 2000 473 339 5 039 000 686 693 687 2.9 –2.1 2001 509 832 5 182 000 783 737 718 0.5 2.1 2002 542 812 5 329 000 758 779 747 4.8 –0.5 2003 571 498 5 480 000 832 818 774 8.1 –2.4 2004 612 330 5 650 000 818 868 808 19.1 –10.9

321 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Table 3 Estimated fatalities in Qatar according to Smeed’s equation and regression analysis, 1990–2006 Year Vehicles Population Fatalities Error (%) All Regression Smeed’s Regression Smeed’s analysis equation analysis equation estimate estimate 1990 161 262 422 145 96 63 92 –34.4 4.3 1991 177 082 435 658 96 70 97 –27.1 –0.8 1992 190 050 449 606 116 76 101 –34.5 12.7 1993 203 001 464 009 84 81 106 –3.6 –25.8 1994 207 912 476 402 52 84 108 61.5 –108.5 1995 217 802 494 225 99 88 113 –11.1 –14.0 1996 231 006 510 070 89 94 118 5.6 –32.0 1997 249 787 526 429 96 102 123 6.3 –28.3 1998 269 510 543 315 106 111 129 4.7 –21.7 1999 284 018 560 746 96 118 134 22.9 –39.7 2000 299 611 578 470 85 124 139 45.9 –64.0 2001 319 318 595 321 110 133 145 20.9 –31.9 2002 348 840 616 151 114 146 153 28.1 –34.1 2003 366 532 717 984 150 154 172 2.7 –14.8 2004 402 006 755 163 164 170 184 3.7 –12.0 2005 457 239 796 186 206 181 199 –12.1 3.6 2006 544 013 837 209 270 233 218 –13.7 19.4

than the populations of of the UAE and It has been reported that high per It has been indicated in a few indi- Qatar, was 5.8% (Table 4). capita economic growth in a country is vidual countries that increased prosper- no longer associated with additional traf- ity in itself is not sufficient to reduce the fic deaths [17,18], although the vehicle number of traffic deaths. A major public health challenge is to utilize this experi- Discussion population and number of crashes in- ence to avoid the predicted increase in crease with economic growth. We iden- The results obtained in our study are traffic-related mortality in less-devel- tified a different trend in Qatar which consistent with other reported studies oped countries [25]. For example, in deviates from this image. Although the Spain and Greece, the traffic crash mor- [6,12,15,17–19] that equation (2) ap- GDP per capita in Qatar was US$ 62 700 tality rate continued to rise after passing plies to both industrialized and develop- in 2006, fatalities increased with eco- the prosperity level that is expected to ing countries, although with different nomic growth. This positive association become protective. Both countries have values of α and β, safety and hazard between GDP and traffic deaths in Qatar so far been less successful in reducing indices, respectively, of a country [19]. may be due to the lack of improvement their fatal injury rates than most other Also, the ratio of deaths per vehicle in the traffic infrastructure and availability countries. This indicates that, regardless of the level of economic development, should not be used as a measure for inter- of improved trauma care. Improvements active policies and trauma care to re- national comparisons because deaths are in pre-hospital and emergency medical duce the number of traffic deaths should generally non-linearly related to vehicles. care could reduce death rates. never be neglected.

Table 4 Comparison of average absolute error for Smeed’s equation and Conclusion regression equation Country Average absolute error Regression analysis can be applied to es- Smeed’s equation Regression equation timate road fatalities in Qatar although United Arab Emirates 23.9 5.5 adjustment of bias is one of the weak- Jordan 5.8 10.9 nesses. Applying Smeed’s equation to Qatar 29.6 20.9 the data for Qatar produced a much

322 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

greater error than that produced by the number of deaths in any year in any actual road fatalities from the expected regression analysis. Smeed’s equation country. Smeed’s equation calculation number in developing countries. cannot be extended to accurately predict results in a considerable deviation of the

References

1. World health report 2001: Mental health: new understanding, 14. Van Beeck EF, Borshoom GJ, Mackenbach JP. Economic de- new hope. Geneva, World Health Organization, 2001. velopment and traffic accident mortality in the industrialised 2. Peden M et al. World report on road traffic injury prevention. world, 1962–1990. International journal of epidemiology, 2000, Geneva, World Health Organization, 2004. 29:503–9. 3. Bener A, Crundall D. Road traffic accidents in the United Arab 15. Kopits E, Cropper MI. Traffic fatalities and economic growth. Emirates compared to western countries. Advances in transport Washington DC, World Bank, 2003 (Policy Research Working Paper 3035). studies, 2005, 6:5–12. 16. About Qatar. Doha, Bureau of Public Affairs, 2009 (http:// 4. Paulozzi LJ et al. Economic development’s effect on road www.state.gov/r/pa/ei/bgn/5437.htm, accessed 29 July transport-related mortality among different types of road us- 2009). ers: a cross-sectional international study. Accident analysis & prevention, 2007, 39:606–17. 17. Bishai D. Traffic fatalities and economic growth.Accident analysis & prevention, 2005, 37:169–78. 5. Murray C, Lopez A, eds. The global burden of disease. Cam- bridge, Massachusetts, Harvard Press, 1996. 18. Bishai D et al. National road casualties and economic development. Health economy, 2006, 15:65–81. 6. Elvik R. Analysis of official economic valuations of traffic accident fatalities in 20 motorized countries. Accident analysis & 19. Bener A, Ofosu JB. Road traffic fatalities in Saudi Arabia.Journal prevention, 1995, 27:237–47. of the International Association of Traffic and Safety Sciences, 1991, 15:35–8. 7. Bener A et al. Strategy to improve road safety in developing countries. Saudi medical journal, 2003, 24:603–8. 20. Annual health report. Doha, Qatar, Hamad Medical Corpora- tion, 2006. 8. Hasson P. Rural road safety: a global challenge. Public roads, 21. Global economic prospects 2002: making trade work for the 1999 63(2) (http://www.tfhrc.gov/pubrds/septoct99/rural- world’s poor. Washington DC, World Bank, 2002 (http://pubrds.htm, accessed 2 August 2009). lications.worldbank.org/ecommerce/catalog/product?item_ 9. Smeed RJ. Some statistical aspects of road safety research. Jour- id=432634, accessed 29 July 2009). nal of the Royal Statistical Society, Series-A, 1949, 12:1–23. 22. Jadaan KS, Khalil R, Bener A. A mathematical model using 10. Smeed RJ. Methods available to reduce the number of road convex combination for the prediction of road traffic deaths. casualties. Traffic engineering & control, 1964, 6:509–12. Journal of computing & information, 1991, 2:139–57. 11. Andreassen DC. Linking deaths with vehicles and population. 23. Preston B. Road safety: international comparisons. Transport Traffic engineering & control, 1985, 26:547–9. reviews, 1980, 1:75–98. 12. Haight FA. Traffic safety in developing countries. Journal of 24. Jacobs, GD, Bardsley MN. Research on road accidents in safety research, 1980, 12:50–8. developing countries”, Traffic engineering & control, 1977, 13. Hampson G. The theory of accident compensation and the 18:166–70. introduction of compulsory seat belt legislation in New South 25. Soderlund N, Zwi AB. Traffic-related mortality in industrialized Wales. Proceedings of the Australian Road Research Board Con- and less developed countries. Bulletin of the World Health Or- ference, 1982, 11:135–40. ganization, 1995, 73:175–82.

323 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Drug prescription habits in public and private health facilities in 2 provinces in South Africa G. Mohlala,1 K. Peltzer,1,2 N. Phaswana-Mafuya 1,3 and S. Ramlagan 1

عادات وصف األدوية يف املرافق الصحية يف القطاعني العام واخلاص يف مقاطعتني يف جنوب أفريقيا غوردن مهالال، كال بلتزر، نانيس فسوانا مافيوا، شاندير رامالغان هدفت اخلالصـة:هذه الدراسة إىل استقصاء عادات وصف األدوية باستخدام املؤرشات املعيارية ملنظمة الصحة العاملية يف 15 مستشفى ًتابعا للقطاع العام و36من العيادات اجلراحية التابعة للقطاع اخلاص، وذلك يف مقاطعتني يف جنوب أفريقيا. واتضح من الدراسة ارتفاع العدد الوسطي لألدوية التي توصف للمرىض يف املستشفيات التابعة للقطاع العام ) (3.2 مقابل ما يصفه األطباء العامون ) (، 2.8وأن معدالت وصف األدوية اجلنيسة منخفضة لدى املستشفيات التابعة للقطاع العام )45.2%( مقابل ما يصفه األطباء العامون منها )24.5%(. أما معدالت وصف احلقن فقد كانت يف املستشفيات التابعة للقطاع العام 8.3% مقابل ما يصفه األطباء العامون منها 23.3%، ومعدالت وصف املضادات احليوية يف املستشفيات التابعة للقطاع العام 68.1% مقابل ما يصفه األطباء العامون منها 31.9%، ومعدالت وصف أدوية مدرجة ضمن قائمة األدوية األساسية يف املستشفيات التابعة للقطاع العام 92.6% مقابل ما يصفه األطباء العامون منها 68.5% واستنتج الباحثون أن هناك حاجة لتنظيم وصف األدوية يف كل من القطاع العام والقطاع اخلاص َّوالسياموصف املضادات احليوية واألدوية األساسية واألسامء اجلنيسة.

ABSTRACT The aim of this study was to explore drug prescription habits using WHO standard indicators in 15 public hospitals and 36 private surgeries in 2 provinces in South Africa. A high mean number of drugs were prescribed per patient (3.2 versus 2.8) in public hospitals and by general practitioners (GPs) respectively and generic prescribing rates were low (45.2% versus 24.5%). The rates of prescribing in public hospitals and by GPs were 8.3% versus 23.3% for injections, 68.1% versus 31.9% for antibiotics and 92.6% versus 68.5% for drugs from the essential drugs list. Drug prescribing in both sectors needs to be regulated, especially the use of antibiotics, essential drugs and generic prescribing.

Habitudes en matière de prescription de médicaments dans les établissements de santé publics et privés de deux provinces d’Afrique du Sud

RÉSUMÉ L’objectif de cette étude était d’examiner les habitudes en matière de prescription de médicaments à l’aide des indicateurs standards de l’OMS dans 15 hôpitaux publics et 36 cabinets privés dans deux provinces d’Afrique du Sud. Un nombre moyen élevé de médicaments était prescrit par patient (3,2 contre 2,8) dans les hôpitaux publics et par les médecins généralistes respectivement et le taux de prescription de médicaments génériques était faible (45,2 % contre 24,5 %). Le taux de prescription dans les hôpitaux publics et par les médecins généralistes était de 8,3 % contre 23,3 % pour les injections, 68,1 % contre 31,9 % pour les antibiotiques et 92,6 % contre 68,5 % pour les médicaments issus de la liste OMS de médicaments essentiels. Dans les deux secteurs, les prescriptions médicamenteuses ont besoin d’être régulées, en particulier en ce qui concerne l’utilisation d’antibiotiques, de médicaments essentiels et de génériques.

1Health Systems Development Unit, Human Sciences Research Council, Pretoria, South Africa (Correspondence to K. Peltzer: [email protected]). 2University of the Free State, Bloemfontein, South Africa. 3Nelson Mandela Metropolitan University, Port Elizabeth, South Africa. Received: 25/08/07; accepted: 09/12/07

324 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Introduction however, there is little or no knowledge Africa and the national Department of about rational drug use in the private Health website. These lists were col- The South African national drug policy health care sector as compared with lated to produce a sampling frame of (NDP) [1] was launched in January the public sector. The aim of this study 200 private GP surgeries (100 in each 1996 with the aim of ensuring an ad- was to investigate rational prescribing province). Over-sampling of private equate and reliable supply of safe, cost- indicators in private surgeries and pub- surgeries was necessary to compensate effective medicines of acceptable quality lic hospitals in Limpopo and Western for an expected high refusal rate. A target and encouraging rational use of these Cape provinces of South Africa. of 10 general practitioners (GPs) was drugs. Rational use of drugs requires set for each district. In districts with that patients receive medications ap- less than 10 GPs, all GPs were included propriate to their clinical needs, in doses Methods in the sample. For private surgeries, a that meet their own requirements, for an random sample of 15 patient files and adequate period of time and at the low- A cross-sectional design was employed 15 patients were selected. est cost to them and their community in this study carried out from August to Overall, 15 public hospitals and 36 [2]. Implementation of the NDP was, December 2005 based on a retrospec- GPs participated in the study. A total however, characterized by a mixed, and tive analysis of recent medical informa- of 733 exit interviews were conducted at times controversial, set of outcomes. tion from patient files and exit interviews with patients at public hospitals and 296 Several reviews of the NDP highlighted with patients at private surgeries and at selected private GP surgeries. All tar- important gains, notably the develop- public hospitals in 2 provinces. geted public hospitals agreed to partici- ment of an essential drugs list (EDL) Sample pate in the study. The low response rate and standard treatment guidelines for in private surgeries was attributable to various levels of public sector care [3]. Purposeful selection of 2 out of 9 prov- fewer patients booked for consultation Previous baseline and follow-up studies inces (Limpopo and Western Cape) on the day of the survey or simply a low conducted in Gauteng, Western Cape allowed for a comparative analysis of client turnout. Reasons for refusals from and Limpopo provinces between 1996 findings between a predominantly ur- some GPs to participate in the study and 2003 reported overprescribing of ban and a predominantly rural province. varied from fear of being investigated, antibiotics (> 50% of drugs prescribed) In addition, Western Cape and Lim- patient confidentiality, lack of free time, in public health facilities [4]. popo provinces were identified because the facility manager not being available The World Health Organization/ of previous reporting of antibiotic over- at the time of survey and unwillingness International Network for Rational Use prescribing (> 50%) in public health to participate. For most patients the of Drugs (WHO/INRUD) has set facilities [4]. Within each province, 2 reason for refusal to participate in the standards that should apply to prescrib- health districts (1 rural and 1 urban) study was lack of time, although this ing [5]. Under-prescribing can result were randomly selected. may have concealed a fear of divulging in subtherapeutic effects, secondary To select health facilities in the medical information to strangers. infections, a false sense of wellness and provinces, a list of public hospitals was delayed treatment. Over-prescribing, obtained from the provincial health Measures on the other hand, can lead to unwanted department offices and was used to up- WHO methods and guidelines for the drug interactions, adverse effects and date the 1998 list of public hospitals in evaluation of drug use with specific ultimately patient noncompliance. As the Human Sciences Research Council reference to rational drug use indicators a result, treatment failure usually leads database. The updated list was used to were used [10]. to the prescribing of newer treatment produce a sampling frame for hospitals. A retrospective analysis of recent regimens that are usually more costly An overall proportionate random sam- medical information from patients’ and less tolerable, thus reducing the ple was made of 15 public hospitals (9 in files allowed for the assessment of the chances of treatment success. Limpopo and 6 in the Western Cape). number of drugs prescribed per hospital- Several studies have been done in Random samples of 50 inpatient files, ization and number of drugs prescribed other countries to assess prescribing 50 outpatient files and 50 patients were per consultation day. A data collection habits but most of these were limited to set for each public hospital. form adopted from the WHO guide- drug handling in the public sector [6–8]. Lists of private surgeries in South lines [5] was used to collect information A study conducted in Zimbabwe com- Africa were obtained from MEDPages about patients’ demographic data (not pared prescription habits in practices of (a source for health care contact in- reported in this paper) and their most dispensing and nondispensing doctors formation in Southern Africa), the recent visit to the health facility. The in the private sector [9]. In South Africa, Health Professions Council of South medical history area had 11 items which

325 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

included among others, the patient’s age, In the Western Cape a mean of 1.9 suggest that there may be a problem of whether the patient’s recent visit was for drugs was prescribed in private surgeries over-prescribing in public hospitals in day consultation (outpatient) or for compared with 3.0 in public hospitals, the Western Cape compared with Lim- hospitalization (inpatient), the number and in Limpopo the mean was 3.7 in popo province where private surgeries of drugs prescribed and whether the private surgeries compared with 3.4 in prescribed slightly more drugs than drugs prescribed were generic. public hospitals. Analysis of data from public hospitals. Hogerzeil, in a study Patient exit interviews were con- patients’ files revealed that significantly of drug use in 12 developing countries ducted with the aim of assessing the fewer drugs were prescribed per day (including outlying values), found a percentage of drugs prescribed from consultation in private surgeries than in high average numbers of drugs per en- the EDL and the percentage of pre- public hospitals (mean 2.8 versus 3.2). counter in Indonesia and Nigeria (3.3 Comparing inpatients with outpatients scriptions with at least 1 injection and and 3.8) [12]. Hafeez et al. found the in public hospitals, a mean of 3.6 (SD 1 antibiotic. A group of 14 professional average number of drugs per prescrip- 2.3) drugs was prescribed per hospitali- tion was 2.7 in public sector facilities in nurses were trained as fieldworkers to zation (inpatient files) compared with a Pakistan [13]. Keohavong et al. found conduct interviews. mean 2.7 (SD 1.9) drugs prescribed per that an average 3 drugs were prescribed Informed consent was obtained day consultation (outpatient files). per encounter in the public sector in from GPs, public hospital managers and In the Western Cape 92.0% of all Lao People’s Democratic Republic [8]. patients. The study was approved by the prescribed drugs in public hospitals were Enwere et al. found the overall average Human Sciences Research Council’s drugs from the EDL compared with number of drugs prescribed was 3.2 in ethics committee. 68.0% in private surgeries. In Limpopo a medical outpatient clinic of a Nigerian Data analysis and 93.1% of prescribed drugs were drugs public tertiary hospital [14]. Compared management on the EDL compared with 69.0% in with our study, where GPs prescribed private surgeries. Public hospitals sig- 2.8 drugs per encounter, Moghadamnia The data were double entered and nificantly more often prescribed EDL et al. found an even higher rate (4.4) verified using Microsoft Access 2003. drugs than did GPs. among GPs in the Islamic Republic The database was designed to include In both Western Cape and Limpopo of Iran [15] and Trap et al. found range checks. The data were converted generic prescribing was significantly lower rates among GPs in Zimbabwe to SPSS, version 13.0 for analysis. The lower in private surgeries (27.1% and whereby dispensing doctors prescrib- descriptive statistics are reported with 21.9% respectively) than in public hos- ing significantly more drugs per patient frequencies and means and standard pitals (48.6% and 41.7% respectively). than nondispensing doctors (2.3 versus deviation (SD). Tests of significance In both Western Cape and Limpopo 1.7) [9]. for categorical variables were based on antibiotic prescribing was significantly Generic prescribing chi-squared tests and for means on Stu- higher in public hospitals (72.8% and dent t-tests. All P-values were derived 63.4% respectively) than in private sur- In this study the rate of prescribing from mean differences with Student geries (27.2% and 36.6% respectively). drugs by generic name was found to t-test. A P-value < 0.05 was considered In the Western Cape 13.7% of be low in public hospitals (45.2%) and statistically significant. Generic pre- patients who visited private surgeries by GPs (24.5%). This is similar to the scribing and EDL drugs were analysed received at least 1 injection compared outpatient clinic of a Nigerian public by a pharmacist (G.M.) using recent with 6.7% in public hospitals, and in hospital where the average percentage pharmacological textbooks, the South Limpopo the figures were 32.9% in of drugs prescribed by generic names African national standard treatment private surgeries and 9.8% in public was 49.5% [14], whereas in the Islamic guidelines and the EDL [11]. hospitals. Republic of Iran 98% of GPs prescribed by generic name [15]. If doctors wrote more prescriptions for cheaper branded Results Discussion generic drugs, both the government and consumers would achieve significant The findings for each drug use indicator Number of drugs per savings without any deterioration in by province and type of health facility encounter/prescription patient care [13]. However, a number are shown in Table 1. Statistics for each This study found a high mean number of factors have been attributed to the analysis were based on the cases with of drugs prescribed per prescription failure of private doctors to prescribe no missing or out-of-range data for any at public hospitals (3.2) and private generic medicines. Economic factors variable in the analysis. GP surgeries (2.8). Our findings also may play a role, as some pharmaceutical

326 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 1 Comparison of drug prescribing habits in public hospitals and private surgeries by province using WHO/INRUD indicators Province/facility No. of drugs per % of encounters % of antibiotics % of drugs % of drugs prescription with ≥ 1 injection prescribed per total prescribed prescribed prescribed responses from EDL generically Mean (SD) No. % No. % No. % No. % Limpopo province Public hospitals 3.4 (1.8) 429 9.8 1552 63.4 1409 93.1 1367 41.7 Private surgeries 3.7 (1.6) 152 32.9 630 36.6 415 69.0 147 21.9 Western Cape province Public hospitals 3.0 (2.0) 282 6.7 902 72.8 828 92.0 859 48.6 Private surgeries 1.9 (2.5) 124 13.7 275 27.2 181 68.0 153 27.1 Total Public hospitals 3.2 (2.3) 721 8.3 2454 68.1 2237 92.6 2226 45.2 Private surgeries 2.8 (1.3) 276 23.3 905 31.9 596 68.5 300 24.5

t- or χ2-value t = 9.42 χ2 = 18.55 χ2 = 34.34 χ2 = 23.38 χ2 = 29.35 P-value < 0.001 < 0.001 < 0.001 < 0.001 < 0.001

WHO/INRUD = World Health Organization/International Network for Rational Use of Drugs; SD = standard deviation; EDL = essential drugs list.

companies pay rewards to doctors treatment when there are several op- Prescribing drugs from the EDL who prescribe their products and this tions. Therefore a patient can choose if With regards to prescribing drugs from discourages generic prescribing. A they want to take an injectable drug or the national EDL, we found that public study conducted in Zimbabwe found not. Our results suggest that injection hospitals (92.6%) prescribed more that other factors, such as the desire use is not a general problem in South drugs from the EDL than did private to sustain income, play a role in the Africa. surgeries (68.5%). Other studies have prescribing and dispensing habits of found that the rate of medicines pre- private doctors [9]. Use of antibiotics scribed in the public sector conforming Our study found that antibiotic pre- to the national EDL was over 70% in Use of injections scribing was very high in the 2 provinces Tanzania [6], 84% in Lao [8] and 96% We found injections were prescribed in (68.1% in public hospitals and 31.9% by in Nigeria [14]. Rothberg and Walters 8.3% of encounters in public hospitals GPs) compared with more than 50% found in a large health maintenance and 23.3% in private surgeries. Other in public hospitals in the EDP survey in organization in South Africa that only studies have found higher rates of in- South Africa in 2003 [4]. Similar high 22.4% of current GP prescriptions jection prescribing in public health fa- rates of antibiotic prescribing of 1 or included EDL items; a further 19.6% cilities in developing countries, ranging more antibiotics have been reported in included “other forms of EDL” items from 36% to 48% in Uganda, Sudan and public health facilities in Uganda and [18]. Simply obtaining those EDL Nigeria [12], 18% in Lao [8], over 37% products that are currently prescribed Sudan (56% and 63% respectively) in Ethiopia [7] and 41% in the Islamic at state tender prices would reduce [12], 47% in Lao [8] and 60%–65% in Republic of Iran [16]. While our study costs by almost 20%, while extending Ethiopia [7], 58% in the Islamic Repub- found an injection prescribing rate of the use of EDL products might save 23.3% among GPs, a much higher rate lic of Iran [16] and 60.9% in Jordan [17]. in excess of 70% on private sector GP of prescribing of injections (58%) was Our findings suggest that antibiotic prescriptions. Compared with 1996, found among Iranian GPs [15]. Similar prescribing in the public sector needs there has been a significant increase rates have been found among doctors to be regulated. Tuberculosis and other from 22% to 69% in prescribing from in Zimbabwe, with dispensing doctors opportunistic infections related to HIV the EDL in the private health sector providing injections to more patients infection might have an influence on in South Africa. One reason why EDL (28.4%) than non-dispensing doctors high antibiotic prescribing. Today, the prescribing is not higher might be that (9.5%) [9]. In South Africa patients problem of antibiotic use is receiving in South Africa the private sector is only today have the right to accept or reject global attention as a result of increasing encouraged and not obliged to use the treatment and to choose the form of antimicrobial drug resistance. EDL. Furthermore, essential drugs in

327 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

the public sector only apply to certain highly scheduled and/or are used for Acknowledgements common ailments and it is therefore not rare diseases). possible to have 100% drug prescribing The team acknowledges with thanks the out of the EDL. Since the South African contribution and assistance of the follow- EDL does not contain all medications Conclusions ing: the Department for International for all illnesses or diseases but only Development for funding the study; the for most common ailments, prescrip- Our findings suggest that drug pre- provincial Departments of Health in tion of highly scheduled, more costly scribing by public and private prescrib- Limpopo and the Western Cape as well medications that do not appear on the ers needs to be regulated closely in as the private doctors who participated EDL is permitted but requires extensive South Africa. Use of injections was in the study for giving us permission to motivation on the part of the medical not found to be a problem in the 2 collect information from their health doctor or specialist who is prescribing provinces studied. With regards to the care facilities; Dr Khangelani Zuma, Dr (he/she has to be convinced or know prescribing of generic medicines, all Henry Fomundam and Adlai Davids about other medications not on the prescribers should be obliged to pre- for their input in the study and sampling EDL list and has to go through a lot of scribe generically to give the patient a design; Professor Andy Gray for earlier bureaucratic red tape to get permission choice of brand they want to use at the review comments; and the anonymous to prescribe the non-EDL drugs that are cost that suits them. reviewers for their useful comments.

References

1. National drug policy for South Africa. Pretoria, South Africa, De- 11. Standard treatment guidelines and essential drugs list for South partment of Health, 1997. Africa: primary health care, 3rd ed. Pretoria, South Africa, De- 2. Le Grand et al. Intervention research in rational use of drugs: a partment of Health, 2003. review. Health policy and planning, 1999, 14(2):89–102. 12. Hogerzeil HV. Field tests of rational drug use in twelve devel- 3. Gray A et al. Policy change in the context of transition: drug policy oping countries. Lancet, 1993, 342:1408–11. in South Africa 1989–1999. Johannesburg, South Africa, Centre 13. Hafeez A et al. Prescription and dispensing practices in public for Health Policy, School of Public Health, University of the sector health facilities in Pakistan: survey report. Journal of the Witwatersrand, 2002. Pakistan Medical Association, 2004, 54(4):187–91. 4. Impact of essential drugs programmes at primary health care level 14. Enwere OO, Falade CO, Salako BL. Drug prescribing pattern in South Africa: key findings and recommendations of surveys in at the medical outpatient clinic of a tertiary hospital in south- 2003. Pretoria, South Africa, Department of Health, 2003. western Nigeria. Pharmacoepidemiology and drug safety, 2007, 5. How to investigate drug use in health facilities. Geneva, World 16(11):1244–9. Health Organization, 1993. 15. Moghadamnia AA, Mirbolooki MR, Aghili MB. General 6. Massele AY et al. A study of prescribing patterns with special practitioner prescribing patterns in Babol city, Islamic Re- reference to drug use indicators in Dar es Salaam Region, Tan- public of Iran. Eastern Mediterranean health journal, 2002, zania. Tropical doctor, 1993, 23(3):104–7. 8(4–5):550–5. 7. Desta Z et al. Assessment of rational drug use and prescribing in primary health care facilities in North West Ethiopia. East 16. Cheraghali AM et al. Evaluation of availability, accessibil- African medical journal, 1997, 74(12):758–63. ity and prescribing pattern of medicines in the Islamic Re- public of Iran. Eastern Mediterranean health journal, 2004, 8. Keohavong B et al. Rational use of drugs: prescribing and dis- 10(3):406–15. pensing practices at public health facilities in Lao PDR. South East Asian journal of tropical medicine and public health, 2002, 17. Otoom S et al. Evaluation of drug use in Jordan using WHO 33(3):647–53. prescribing indicators. Eastern Mediterranean health journal, 9. Trap B et al. Prescription habits of dispensing and non-dispens- 2002, 8(4–5):537–43. ing doctors in Zimbabwe. Health policy and planning, 2002, 18. Rothberg AD, Walters L. Formulary and funding implications of 17(3):288–95. the gap between the national Essential Drugs List and current 10. Indicators for monitoring national drug policies: a practical prescribing in a large health maintenance organisation. South manual. Geneva, World Health Organization, 1999. African medical journal, 1996, 86(9):1084–90.

328 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Gender-specific oral health attitudes and behaviour among dental students in Palestine E. Kateeb 1

املواقف والسلوك لدى طالب طب األسنان يف فلسطني نحو صحة الفم اخلاصة بجنس معني إهلام طالب اخلطيب

اخلالصـة: ِّمهذه الدراسةتقي الفروق بني اجلنسني يف املعارف واملواقف واملامرسات بني طالب طب األسنان يف فلسطني حول صحة الفم. وقد 22 18 260 استوىف ًطالبا تتـراوح أعامرهم بني و ًعامايف جامعة القدس اإلخراجة اإلنكليزية مل ْ َسرد السلوك لطب األسنان يف جامعة هريوشيام. واتضح أن لدى اإلناث مواقف وسلوكيات أكثر إجيابية حول صحة الفم )فهن يقمن بزيارات منتظمة ألطباء األسنان، وأكثر ًثقافة حول استخدام فرشاة األسنان بطريقة مهنية، واعتنائهن أكثر باستخدامها(. وقد كان نصف الطالب من اجلنسني يظنون أهنم ال يستطيعون تفادي وضع أسنان صناعية عند بلوغهم ًسنامتقدمة. وقد كانت بعض جوانب السلوكيات واملواقف يف صحة األسنان خمتلفة بني الذكور واإلناث، فيام كان التدريب املهني يف جوانب أخرى للسلوكيات واملواقف ِّيعوض هذه االختالفات.

ABSTRACT This study assessed gender differences in oral health knowledge, attitude and behaviour among undergraduate dental students in Palestine. Students aged 18–22 years at Al Quds University (n = 260) completed the English version of the Hiroshima University–Dental Behavioral Inventory. Females had more positive dental health attitudes and behaviours (making regular visits the dentist, being more educated about professional toothbrushing and taking more care with brushing their teeth). Half the students of both sexes thought they could not avoid having false teeth when they were old. Some aspects of oral health behaviour and attitudes were different between males and females, but in other aspects professional training may have compensated for these differences.

Attitudes et comportements spécifiques à chaque sexe en matière d’hygiène bucco-dentaire parmi les étudiants en médecine dentaire en Palestine

RÉSUMÉ Cette étude évalue les différences selon les sexes en matière de connaissances de l’hygiène bucco- dentaire ainsi que les attitudes et les comportements des étudiants de premier cycle de médecine dentaire en Palestine. Âgés de 18 à 22 ans, les étudiants de l’université Al Quds (n = 260) ont rempli la version anglaise du questionnaire de l’Université d’Hiroshima sur l’hygiène dentaire (Hiroshima University–Dental Behavioral Inventory). Les attitudes et les comportements des femmes étaient meilleurs (visites régulières chez le dentiste, plus de connaissances en termes de brossage et plus de soin dans le brossage). La moitié des étudiants des deux sexes pensaient qu’ils ne pourraient pas éviter d’avoir de fausses dents lorsqu’ils seraient âgés. Certains aspects liés aux attitudes et comportements en matière d’hygiène bucco-dentaire étaient différents entre les hommes et les femmes, mais pour d’autres, la formation professionnelle peut avoir atténué ces différences.

1Department of Preventive and Community Dentistry, Al Quds University, East Jerusalem (Correspondence to E. Kateeb: [email protected]). Received: 19/11/07; accepted: 09/03/08

329 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Introduction attitudes between male and female den- was made to follow up students who tal students. were absent on the day of the survey. There is an increased interest in look- ing at gender differences in health and Statistical analysis disease, including dental health [1]. In Methods A total score was calculated based Lebanon, a survey of oral health prac- on the response to each statement. All tices in 1998 found poor flossing habits All undergraduate dental students at Al variables of the questionnaires were and unfavourable dietary habits among Quds University were invited to par- analysed by sex. Differences were as- students, and that variables such as sex, ticipate in the survey at the end of the sessed by the chi-squared test. SPSS, father’s education, exercise and dietary academic year 2006. version 10, was used to process and habits were significantly correlated with Data on oral health behaviour were analyse the data. The significance level how frequently the students brushed collected using the English version of the (P-value) was set at 0.05. their teeth [2]. According to the Report Hiroshima University-Dental Behav- on the Survey of Dental Diseases in ioral Inventory (HU-DBI) question- Tokyo, data have consistently shown naire, which was originally developed Results that the mean number of filled teeth in by Kawamura [8]. It consists of 20 items women is higher than in men and that primarily associated with toothbrush- Of 271 registered Palestinian students women tend to become edentulous ing behaviour, all with a dichotomous at Al Quds University Dental School, at a younger age [3]. These gender dif- response format (agree/disagree). A 260 students completed the question- ferences are not easy to explain; some quantitative estimate of oral health at- naire (175 females and 85 males), a researchers think the differences reflect titude and behaviour is provided by response rate of 94.1%. The age of the dentist–patient relationships more than the total of the appropriate responses. students ranged from 18 years in the disease occurrence [4]. Higher scores indicate better oral health 1st year to 22 years in the 5th year. The One study of self-reported dental attitude and behaviour 9,10[ ]. In a sam- distribution of the study sample over health attitudes and behaviour among ple of 517 Japanese university students, the 5 academic years and the response dental students showed that gender was the HU-DBI had good test–retest reli- rate by year and by sex are shown in not a major feature, but that favourable ability (0.73) over a 4-week period [11]. Table 1. attitudes/behaviour toward oral health The English version has also shown Table 2 shows the dental students’ at- appeared to reflect students’ clinical good test–retest reliability and transla- titudes and behaviour regarding dental training experience [5]. In another tion validity in a sample of 26 bilingual health and hygiene by sex. Significantly study, the gender difference was signifi- individuals [10]. The questionnaire has more male students than females were cant among dental students. In Jordan, been used to compare oral health at- not worried about visiting the dentist female students reported brushing their titudes and behaviours among dental (52.9% versus 32.6% respectively) (P < teeth more frequently than male stu- and dental hygiene students in different 0.01). More males than females agreed dents and believed in the necessity of countries [11–14]. that they put off going to the dentist using toothpaste during brushing more Students were asked by their faculties until they had a toothache (57.6% ver- often than did male students 6[ ]. So too to remain in class at the end of the final sus 46.9%) but the difference was not in the Islamic Republic of Iran, where examination of the academic year and significant. female senior dental students reported participate in the survey on a voluntary The same proportion of males and significantly higher frequencies of basis. The HU-DBI questionnaire was females (31.8% and 32.0% respectively) toothbrushing, fluoridated toothpaste distributed to all students. The English agreed that their gums bled when they use and flossing compared with male version of the questionnaire was used brushed their teeth. The majority of students [7]. in this survey as English is the language the students worried about the colour There is very little information from of instruction at the dental school at Al of their teeth and again there was no Palestine about gender differences in Quds University. Questions regarding significant difference between the sexes any aspect of general health or lifestyle the meaning of words in Arabic were (84.7% of males and 86.9% of females). practices that influence health. This allowed and answers to such questions Half of the dental students believed study therefore aimed to add to the were announced to all other students. that they could not avoid having false scarce literature documenting the oral The survey was completed anony- teeth when they were old (54.1% of hygiene practices and attitudes among mously and no personal, demographic, males and 54.9% of females). Females Palestinian dental students at Al Quds academic or performance data for the tended to care more about having bad University and to compare oral health students were collected. No attempt teeth despite more of them brushing

330 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 1 Demographic features of the sample of dental students in Palestine socioeconomic classes, based on their University year Response Total Males Females Age (years) high-school performance. The overall rate (n = 260) (n = 85) (n = 175) response rate in this study was high % No. No. No. Mean (SD) (94%). 1 100 71 32 39 18.63 (0.52) This study compared female and 2 94 61 13 48 19.65 (0.69) male dental students’ attitudes and 3 95 59 20 39 20.62 (0.92) dental behaviour. Since males and fe- 4 79 27 6 21 21.50 (0.53) males have different physiological and 5 93 42 14 28 22.16 (0.55) psychological behaviours, it is possible SD = standard deviation. that their oral health behaviour might be different as well. Researchers have found that females engage in better oral hygiene behaviour, possess a greater daily than males; 77.7% of females said Discussion interest in oral health and perceive their they brushed each of their teeth care- own oral health to be better than do fully compared with 61.2% of the males Dental students at Al Quds Univer- males [15]. (P < 0.01). More males (36.5%) than sity are a good representative sample of The present study showed differ- females (22.3%) agreed that they had dental students in Palestine. There are ences in dental health attitudes between not been taught how to brush their only 2 dental schools in the Palestinian the sexes, with significantly fewer males teeth professionally (P < 0.05). More of Authority and Al Quds University is the who agreed that they worried about vis- the male students (12.9%) used a child- only one with some financial support iting the dentist and more who put off sized toothbrush compared with female from the state, accepting students from dental visits until they had a toothache. students (8.6%). all over Palestine and from different These results agree in general with the

Table 2 Dental students’ agreement with statements about dental health and hygiene in the Hiroshima University–Dental Behavioral Inventory: distribution by sex Item Males (n = 85) Females (n = 175) χ2-value P-value No. % No. % I do not worry much about visiting the dentist 45 52.9 57 32.6 9.96 0.002** My gums tend to bleed when I brush my teeth 27 31.8 56 32.0 0.001 0.970 I worry about the colour of my teeth 72 84.7 152 86.9 0.22 0.638 I have noticed white sticky deposits on my teeth 38 44.7 60 34.3 2.65 0.104 I use a child-sized toothbrush 11 12.9 15 8.6 1.21 0.271 I think that I can not help having false teeth when I am old 46 54.1 96 54.9 0.01 0.911 I am bothered by the colour of my gums 47 55.3 89 50.9 0.45 0.502 I think my teeth are getting worse despite my daily brushing 27 31.8 75 42.9 2.95 0.086 I brush each of my teeth carefully 52 61.2 136 77.7 7.81 0.005** I have never been taught professionally how to brush 31 36.5 39 22.3 5.85 0.016* I think I can clean my teeth well without using tooth paste 19 22.4 25 14.3 2.65 0.104 I often check my teeth in a mirror after brushing 67 78.8 150 85.7 1.97 0.161 I worry about having bad breath 66 77.6 138 78.9 0.05 0.824 It is impossible to prevent gum disease with toothbrushing alone 46 54.1 110 62.9 0.11 0.177 I put off going to the dentist until I have a toothache 49 57.6 82 46.9 2.66 0.103 I have used a dye to see how clean my teeth are 17 20.0 27 15.4 0.85 0.356 I use a toothbrush which has hard bristles 27 31.8 34 19.4 4.85 0.028* I do not feel I have brushed well unless I brush with strong strokes 32 37.6 63 36.0 0.07 0.796 I feel I sometimes take too much time to brush my teeth 45 52.9 89 50.9 0.10 0.752 I have had my dentist tell me that I brush very well 36 42.4 77 44.0 0.03 0.802

*Significant at P < 0.05; **significant at P < 0.01.

331 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

results of Ostberg et al. and Fukai et al., with our results. In a sample of 375 difference between males and females. who found female dental students had dental students at the Jordan Univer- Also 52% of our sample were bothered better oral health attitudes and took sity of Science and Technology, female about the colour of their gums, again better care of their teeth than their male students reported brushing their teeth without a significant difference between colleagues [15,16]. In a study in Jordan, more frequently than did male students. males and females. However, female male students reported that they vis- Approximately 47% of the male stu- students in general tended to be less ited the dentist only when they had a dents brushed their teeth less than twice satisfied with the appearance of their toothache, which was more often than daily compared with 21% of the female teeth than their male colleagues despite their female peers; however, this was not students. Also, female students believed daily brushing. statistically significant [6]. in the necessity of using toothpaste dur- The finding that females had more Differences between the sexes were ing brushing more often than did male positive dental health attitudes and also seen in health behaviour. Signifi- students [6]. behaviours could be explained on the cantly more females said they brushed The above-mentioned studies re- basis that females usually care more their teeth carefully than did males. ported significant differences between about their body and appearance. They Similarly, a study conducted among men’s and women’s health attitudes and would thus be more concerned about new undergraduate students in Leba- health behaviours. In contrast, other visiting the dentist and would tend to non showed that females brushed their studies of the oral health behaviour of be more educated about their dentition teeth 4 times as often as males [2]. In senior dental students [18] and den- even before entering a course related to a study in Kuwait in 2001 and 2002 tists [19] found no gender differences, dentistry. female students reported twice-a-day probably because of the effect of profes- This study was the first formal assess- toothbrushing much more often than sional training. Contrary to our results, ment of dental attitudes and behaviour did male students, and use of fluoride Tseveenjav et al. found no differences among dental students in Palestine. The toothpaste more often than males. Oral between male and female Mongolian results were in agreement with studies health knowledge in that study was also dental students in toothbrushing fre- in other countries showing that some significantly higher among the female quency [18]. aspects of oral health behaviour and students than among the male students In Sweden the results from a ques- attitude are different between males and [17]. In another oral health attitude tionnaire-based study on 20–25-year- females, but that in other aspects, pro- and behaviour survey of senior Iranian old adults showed that about 59% of the fessional training of the dental students dental students in 2006, female students samples were satisfied with the appear- may compensate for these differences. reported significantly higher frequen- ance of their teeth [20]. In our study, Further studies are needed to evaluate cies of toothbrushing, use of fluoridated 86% of the sample (who were from a whether there is a difference in caries ex- toothpaste and flossing compared with similar age group) worried about the perience and in gingival health between their male colleagues [7]. This agrees colour of their teeth and there was no male and female dental students.

References

1. Zakrzewska JM. Women as dental patients: Are there any gen- 7. Khami MR et al. Oral health behaviour and its determinants der differences? International dental journal, 1996, 46:548–57. amongst Iranian dental students. European journal of dental 2. Kassak KM, Dagher R, Doughan B. Oral hygiene and lifestyle cor- education, 2007, 11:42–7. relates among new undergraduate university students in Leba- 8. Kawamura M. Dental behavioral science: the relationship be- non. Journal of the American College of Health, 2001, 50:15–20. tween perceptions of oral health and oral status in adults. Jour- 3. Dental Health Division of Health Policy Bureau, Ministry of nal of Hiroshima University Dental Society, 1988, 20:273–86. Health and Welfare, Japan. Heisei 5 nen shika shikkan jittai chosa 9. Kawamura M et al. Relationship between CPITN and oral hokoku [Report on the Survey of Dental diseases, 1993]. Tokyo, health behavior in Japanese adults. Australian dental journal, Oral Health Association, 1995:126–8. 1993, 38:381–8. 4. Kawamura M et al. An analytical study on gender differences in 10. Kawamura M et al. Dental behavioral science part IX: bilinguals’ self-reported oral health care and problems of Japanese em- responses to the dental behavioral inventory (HU–DBI). Journal ployees. Journal of occupational health, 1999, 41(2):104–11. of Hiroshima University Dental Society, 1992, 22:198–204. 5. Kawamura M, Iwamoto Y, Wright FAC. A comparison of self-re- 11. Kawabata K et al. The dental health behavior of university stu- ported dental health attitudes and behavior between selected dents and test–retest reliability of the HU–DBI. Journal of dental Japanese and Australian students. Journal of dental education, health, 1990, 40:474–5. 1997, 61:354–60. 12. Kawamura M et al. Comparison of United States and Korean 6. Al-Omari QD, Hamasha AA. Gender-specific oral health at- dental hygiene students using the Hiroshima University–Dental titudes and behavior among dental students in Jordan. Journal Behavioral Inventory (HU–DBI). International dental journal, of contemporary dental practice, 2005, 6(1):107–14. 2002, 52:156–62.

332 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

13. Kawamura Met al. A cross-cultural comparison of dental health 17. Al-Ansari JM, Honkala S. Gender differences in oral health attitudes and behavior among freshman dental students in Ja- knowledge and behavior of the health science college stu- pan, Hong-Kong and West China. International dental journal, dents in Kuwait. Journal of allied health, 2007, 36(1):41–6. 2001, 51:159–63. 18. Tseveenjav B, Vehkalahti M, Murtomaa H. Preventive practice 14. Kawamura Met al. Cross-cultural differences of self–reported of Mongolian dental students. European journal of dental educa- oral health behavior in Japanese and Finnish dental students. tion, 2002, 6(2):74–8. International dental journal, 2000, 50:46–50. 19. Tseveenjav B, Vehkalahti M, Murtomaa H. Oral health and its 15. Ostberg AL, Halling A, Lindblad U. Gender differences in knowl- determinants among Mongolian dentists. Acta odontologica edge, attitude, behavior and perceived oral health among ado- scandinavica, 2004, 62:1–6. lescents. Acta odontologica scandinavica, 1999, 57:231–6. 20. Stenberg P, Håkansson J, Åkerman S. Attitudes to dental 16. Fukai K, Takaesu Y, Maki Y. Gender differences in oral health health and care among 20 to 25-year-old Swedes: results behavior and general health habits in an adult population. Bul- from a questionnaire. Acta odontologica scandinavica, 2000, letin of Tokyo Dental College, 1999, 40(4):187–93. 58:102–6.

Reducing the burden of oral diseases

The burden of oral diseases and other chronic diseases can be decreased simultaneously by addressing common risk factors such as tobacco use and unhealthy diet:

• Decreased intake of sugars and well-balanced nutrition prevent tooth decay and premature tooth loss.

• Tobacco cessation and decreased alcohol consumption reduce risk for oral cancers, periodontal disease, and tooth loss.

• Fruit and vegetable consumption is protective against oral cancer.

• Effective use of protective sports and motor vehicle equipment reduces facial injuries.

Dental cavities can be prevented by a low level of fluoride constantly maintained in the oral cavity. Fluoride can be obtained from fluoridated drinking water, salt, milk, mouth rinse or toothpaste, as well as from professionally-applied fluorides. Long-term exposure to an optimal level of fluoride results in fewer cavities in both children and adults.

Information about the WHO Global Oral Health Programme can be found at: http://www.who.int/oral_health/en/

333 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Amalgam use and waste management by Pakistani dentists: an environmental perspective R. Mumtaz,1 A. Ali Khan,1 N. Noor 2 and S. Humayun 3

استخدام امللغم وتدبري نفاياته لدى أطباء األسنان الباكستانيني: وجهة نظر بيئية روبينة ممتاز، أياز عيل خان، نعامن نور، صدف مهايون اخلالصـة:لتقييم استخدام امللغم amalgam وبروتوكوالت تدبري نفاياته التي يامرسها أطباء األسنان الباكستانيون، أجريت دراسة عرضية شملت 239 من أطباء األسنان يف إسالم أباد وروالبندي، تم اختيارهم ًعىل بناءاملالئمة وعىل االعتيان وفق املجموعات. واتضح أن امللغم هي أكثر املواد التي تستخدم يف ترميم األسنان وأنه اختيار تفرضه العوائق املالية لدى املرىض. وفيام يعتقد 90.4% من أطباء األسنان أن امللغم من عوامل اخلطر فإن 46.4% منهم يعتربونه من األخطار التي تتهدد البيئة. وقلة قليلة منهم )5.9%( لدهيم أداة فاصلة للزئبق يف عياداهتم. إن بروتوكوالت تدبري نفايات امللغم وتدوير الزئبق ينبغي أن تدخل باكستان.

ABSTRACT To assess amalgam use and waste management protocols practised by Pakistani dentists, a cross- sectional study was made of 239 dentists in Islamabad and Rawalpindi, recruited by convenience and cluster sampling. Amalgam was the most frequently used restorative material, with the choice dictated by patients’ financial constraints. While 90.4% of dentists perceived amalgam as a health risk, only 46.4% considered it an environmental hazard. The majority disposed of amalgam waste in the trash, down the sink or as hospital waste. Very few (5.9%) had an amalgam separator installed in their dental office. Amalgam waste management protocols and mercury recycling should be introduced in Pakistan.

Utilisation de l’amalgame et gestion des déchets par les dentistes pakistanais : une approche environnementale

RÉSUMÉ Afin d’évaluer l’utilisation de l’amalgame et les protocoles de gestion des déchets mis en œuvre par les dentistes pakistanais, une étude transversale a été réalisée chez 239 dentistes d’Islamabad et de Rawalpindi, recrutés par échantillonnage de commodité et en grappes. L’amalgame était le matériau de restauration le plus fréquemment utilisé, ce choix étant dicté par les contraintes financières des patients. Alors que 90,4 % des dentistes estimaient que l’amalgame représentait un risque pour la santé, ils n’étaient que 46,4 % à le considérer comme un risque pour l’environnement. La plupart d’entre eux jetaient les restes d’amalgame dans la poubelle, dans l’évier ou le traitaient comme un déchet hospitalier. Un très faible nombre d’entre eux (5,9 %) avait un séparateur d’amalgame dans leur cabinet dentaire. Des protocoles de gestion des déchets d’amalgame et le recyclage du mercure devraient être introduits au Pakistan.

1Department of Community Dentistry, College of Dentistry, Riphah International University, Islamabad, Pakistan (Correspondence to R. Mumtaz: [email protected]). 2Pakistan Institute of Medical Sciences, Islamabad, Pakistan. 3Dr Rubina and Associates Dental Consultants, Islamabad, Pakistan. Received: 25/12/07; accepted: 28/02/08

334 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Introduction to document this aspect of dentists’ any central body, dental clinics within practices in Pakistan. each cluster were identified from the The dental profession is one of the local medical directories of Rawalpindi largest end-users of mercury [1]. and Islamabad respectively and sys- Amalgam manipulation and its waste Methods tematic selection of every 6th clinic was management in the dental office, if not made. Phase 1 was followed by phase 2 strictly regulated, contribute to the A cross-sectional study was conducted to minimize the risk of duplication since risk of occupational exposure as well over a period of 5 months from Febru- many dentists augment their morning as environmental pollution from this ary to June 2007 in the twin cities of employment in a teaching and/or pub- neuro- and nephrotoxic metal [1]. The Rawalpindi and Islamabad. Islamabad, lic dental hospital with evening private routes of mercury pollution from the the capital city, has a population of 1.04 practice on a part-time basis. Duplica- dental office include: unregulated dis- million, while Rawalpindi, its adjoining tion was avoided by asking dentists in posal of amalgam waste in the regular city, has a population of 3.04 million. phase 2 if they had already participated municipal waste or the domestic sewer- The 2 cities have 3 undergraduate teach- in the study. One dentist from each age wastewater; high-risk methods of ing dental hospitals and 3 tertiary-level clinic (the first one who came forward) amalgam manipulation [2]; disposal of government hospitals that have dental was asked to complete the question- extracted teeth in hospital waste that departments, and all 6 institutions also naire. The number of dentists/clinics is often incinerated; and autoclaving/ provide graduate level training in vari- identified in phase 2 was 96. heat sterilizing of amalgam-filling dental ous clinical dental specialities. The data were collected using was instruments [1]. The target population was full- a self-administered, 2-page structured Amalgam use in dentistry has been time or part-time practising licensed questionnaire developed based on embroiled in controversy for the past 3 dental practitioners in Islamabad and standard, validated questions gleaned decades, which has led to widely differ- Rawalpindi. According to the Pakistan from relevant publications [8–10]. The ing strategies. Scandinavian countries Medical and Dental Council (PMDC), questions were closed-ended and in the have begun to phase out the use of the number of registered, practising English language since the medium of amalgams completely [3–5], whereas dentists in these 2 cities was 524 [7]; dental education in Pakistan is English. organizations such as the American based on a confidence level of 95% and It was pilot tested on 13 dental prac- Dental Association, the US Centers confidence interval of 3, a sample size of titioners and revised according to the for Disease Control and Prevention, 352 was calculated. evaluation. The study was administered the US Public Health Service and the Participants were recruited by 2 sam- by a team of volunteer dental students, World Health Organization support pling techniques in 2 consecutive phases. who systematically visited the subjects’ the use of dental amalgam to fill cavities In phase 1, all the dentists working in the dental teaching institutes, teaching but with strict observance of amalgam 6 dental hospitals were approached to hospitals, public hospitals and private waste protocols [2]. participate in the study. The total number dental offices. Standard procedures of In 2004, the global anthropogenic of dentists working in these hospitals at informed consent were used, including release of mercury into the environment the time of the survey was 256. guarantees of anonymity and confiden- was more than 5000 metric tons, of In phase 2, cluster sampling was em- tiality. Some questionnaires were com- which about 50% originated from Asia ployed to access sufficient private practi- pleted on the spot and others were filled [6]. There are few data on the use and tioners to reach the target sample size of out at leisure and collected at a later visit. disposal of dental amalgam in Pakistan, 352. For this the cities were administra- No honorarium was offered. Dentists and the Pakistan Dental Association, tively divided into 15 sections: Islama- not returning the filled questionnaire the main national dental body, has no bad into 5 sections and Rawalpindi into after 2 recalls were considered as unwill- official consensus regarding amalgam 10. Out of the total number of registered ing to participate. Data collection was waste management in dental settings. dentists, 349 practised in Rawalpindi conducted over 3 months from March The aim of this study therefore was to and the rest in Islamabad [7]. Based to May 2007. assess the extent of amalgam use and on the population proportions, all 5 The study was reviewed by the In- waste management protocols practised sections of Islamabad and 5 from stitutional Review Board of the Human by Pakistani dentists. These baseline Rawalpindi, randomly selected by the Subjects Committee of Riphah Univer- data can support recommendations lottery method, were targeted in order sity, Islamabad and granted exemption for an amalgam waste-management to obtain an equal representation from status. protocol for the country. To the best both cities. As private dental practices in The answers to each question were of our knowledge, this the first study Pakistan are not obliged to register with numerically coded and the data were

335 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

entered into SPSS, version 10.0. Since The most common protocol of replaced with non-mercury based fill- the nature of this cross-sectional study amalgam manipulation was the mecha- ings, only 43 (18.0%) and 59 (24.7%) was descriptive, the results were ana- nized capsule system practised by 106 dentists strongly agreed and agreed lysed by descriptive statistics including (44.4%) dentists with 83 (34.7%) still somewhat respectively. However, 77 frequencies and percentages. using the manual method of elemental (32.2%) and 59 (24.7%) dentists disa- mercury and alloy in a pestle and mortar greed strongly or disagreed somewhat while 46 (19.2%) used both methods. to the same question. Results The dental assistant was the person mostly commonly in charge of the tritu- Frequency of amalgam waste generation Of the 352 dental practitioners invited ration according to 191 (79.9%) den- to participate in the study, 103 refused, tists, although 34 (14.2%) shared the To determine the monthly frequency of giving an overall participation rate of responsibility. amalgam waste-generating procedures, 70.7% [189/256 (73.8%) in phase 1 the respondents were asked to quantify and 60/96 (62.5%) in phase 2]. Ten Beliefs about health risks the average number of new amalgam questionnaires from suspected unli- An overwhelming 216 (90.4%) dentists fillings done, removal of old amalgam censed dental practitioners were dis- believed amalgam to be a health risk to restorations and the extraction of teeth carded. The results of this study were both dental personnel and patients alike, containing amalgam restoration in the therefore based on 239 completed but only 111 (46.4%) thought it to be past 3 months. The results are sum- questionnaires, although some dentists an environmental pollutant. Therefore marized in Table 2. Placement of new did not answer every question. when asked whether amalgam fillings amalgam restorations was the most Background characteristics should be completely phased out and frequently carried out procedure by the The characteristics of the study group are summarized in Table 1. The female Table 1 Characteristics of the studied dentists (n = 239 respondents) to male ratio was 1:1.2. The respondents Characteristic No. % were graduates from 14 out of the 19 Sex recognized dental colleges in Pakistan. Female 110 46.0 A majority of the respondents were Male 129 54.0 young graduates (56.5%) with fewer Year of graduation than 5 years of experience and a high Past 5 years 136 56.9 proportion had postgraduate qualifica- tions (13.8%). 5–10 years ago 49 20.5 > 10 years ago 37 15.4 Use and preparation of Level of dental qualification amalgam Bachelor/Doctor of Dental Surgery (BDS/DDS) 161 67.4 The dentists’ choices of material for fill- Postgraduate trainee 45 18.8 ings were decided primarily by patients’ Postgraduate specialist 33 13.8 financial constraints (210, 87.9%) and Clinical setting the clinical indications of the tooth to be Hospital practice only 117 49.0 restored (175, 73.2%), but also aesthetic Private practice only 58 24.3 demands (124, 51.9%) and patient’s Both hospital and private practice 59 24.7 choice (75, 31.4%). The self-declared Place of graduation frequency of use of amalgam was as Islamabad 88 36.8 follows: 147 (61.5%) dentists using it Peshawar 40 16.7 often/fairly often, 74 (31.0%) using Lahore 33 13.8 it always/almost always and only 18 Karachi 30 12.6 (7.5%) never/almost never using it. In comparison to other filling materials, Multan 20 8.4 amalgam was again most frequently Jamshor 10 4.2 used as the choice of filling material Abbotabad 7 2.9 (211, 88.4% of dentists), followed by Quetta 5 2.1 composite resin (140, 58.5%) and glass Foreign qualified 6 2.5 ionomer cement (128, 53.5%). Some frequencies do not add to 239 as respondents did not answer all questions.

336 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Table 2 Distribution of dentists by estimated number of amalgam waste-generating procedures per month in their practice (n = 239 respondents) Item 0–5 procedures 6–10 procedures 11–15 procedures > 15 procedures No. % No. % No. % No. % Placement of new amalgam restorations 21 8.8 12 5.0 39 16.3 165 69.0 Removal of old amalgam restorations 72 30.1 120 50.2 37 15.5 10 4.2 Extraction of teeth containing amalgam restorations 55 23.0 13 5.4 21 8.8 143 59.8

Some frequencies do not add to 239 as respondents did not answer all questions.

majority of the dentists (165, 69.0%), of it in the trash and/or down the the rapid growth of new private dental followed by extraction of teeth contain- sink. The option of disposal as part of colleges [11]. ing old amalgam fillings (143, 59.8%). hazardous hospital waste was practised Patients’ financial constraints dictat- Analysis by clinical setting showed mainly by hospital-based dentists (95, ed the choice of filling material by most that 34/58 private practitioners (58.6%), 81.2%). When questioned on the use of the dentists in this study (87.9%). 87/117 public hospital dentists (74.4%) of an amalgam separator, 14 dentists This naturally leads to amalgam being and 42/59 dentists working in both (5.9%) claimed their dental office had the most frequently used dental filling settings (71.2%) carried out > 15 new one installed in the drainage, while half material because it is inexpensive and amalgam fillings per month. For the the dentists (119, 49.8%) said they did more durable than other kinds of fill- removal of old amalgam fillings, 29 pri- not. Interestingly, 45 (18.8%) dentists ings. Consequently we found that the vate practitioners (50.0%) performed did not know whether they had an amal- number of amalgam waste-generating only 0–5 removals per month while 87 gam separator installed or not and 59 procedures in the average dental office public hospital dentists (74.4%) and (24.7%) did not know what an amalgam in Pakistan was high. This finding differs 38 dentists working in both settings separator was. from 2 cross-sectional surveys of den- (64.4%) performed 6–15 removals per tists from Saudi Arabia who favoured month. A similar trend was seen in ex- composite and glass ionomer restora- traction of amalgam restored teeth with Discussion tions over amalgam [12,13], a fact which only 1 private practitioner performing could be attributed to the differences in this procedure > 15 times a month, As the study sample included graduates gross national product between the 2 while 98 hospital dentists (83.7%) from all the major cities of the country countries [14]. and 48 practitioners in both settings (from 14 out of the 19 recognized den- Although over 90% of the dentists in (81.4%) performed this with the same tal colleges) and their characteristics our study thought amalgam was a health frequency. were congruent with PMDC statistics, risk, less than half believed it was an it can be assumed to be representative environmental pollutant and therefore Management of amalgam of the average Pakistani dentist. The protocols of amalgam manipulation waste sex distribution of 1:1.2 was very close were high-risk practices with little or no Self-rated knowledge regarding the best to the female to male dentist ratio of observance of standard mercury han- management of amalgam waste was 1:1.1 in Pakistan [7]. The proportion dling recommendations. This finding, limited for 135 (56.5%) and moder- of dentists with postgraduate qualifi- however, is consistent with the Saudi ate for 86 (35.9%) dentists. Only 4 cations (13.8%) was nearly twice the study [13]. (1.7%) dentists rated it as excellent and national average of 5.7% [7], but this Although more than half of the den- 11 (4.6%) as good. The ratings were is consistent with the fact that most of tists advocated the continued use of reinforced by the findings on protocols the dental specialists in Pakistan are amalgam, very few of them (5.9%) had practised for disposal of contact and concentrated in the larger cities. A ma- an amalgam separator installed in their non-contact amalgam waste (Figure 1). jority of the respondents were young dental offices. An overwhelming major- Although the disposal practices showed graduates (56.5%) with fewer than 5 ity disposed of amalgam waste in the a variation according to the category years of experience. This is consistent trash, down the sink or in the hospital of amalgam waste, only 6 dentists, all with the growth of dental professional hazardous waste that is eventually incin- private practitioners, claimed to store manpower in Pakistan, where the erated. While 6 dentists claimed to store it in a sealed container for recycling. number of new dental graduates has the amalgam waste in sealed containers The overwhelming majority disposed tripled in the past 5 years as a result of for recycling, on further questioning

337 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

they did not know of any mercury re- of these protocols in developed coun- nature of the survey. Since dental clinics cycling company and these containers tries [20,21]. Developing countries lag in Pakistan are not officially registered were inevitably thrown into the trash. far behind in this scenario and Pakistan with any central body, there was some The findings of our study are con- is no different. selection bias because dental clinics sistent with other studies in developing The issue of mercury pollution has that were not listed in the local medical countries in Asia. Studies in Palestine been taken up on a global platform by directories were not sampled. Addition- and Bangkok found that most dental the mercury programme of the United ally, Pakistan has nearly 70 000 unli- waste, including amalgam, was thrown Nations Environment Programme censed dental practitioners [11]—as in the regular trash [15,16], while 1 (UNEP), which is currently in collabo- compared to 7456 qualified dentists study reported that dentists at a teach- ration with different countries although [7]—whose role in amalgam use and ing hospital in New Delhi, India, were the list does not include Pakistan [22]. its waste management has not been not aware of biomedical waste manage- UNEP offers a full range of technical addressed. Further research on a larger ment and needed training [17]. New support and activities for reducing mer- scale and inclusive of unlicensed practi- Delhi releases 51 kg of mercury through cury use and release into the environ- tioners is called for. amalgam waste annually from hospital ment. We recommended that Pakistan and dental clinic disposal, which, ac- joins the UNEP mercury programme cording to a health care news report, is and begins with the establishment of Acknowledgements unregulated and reckless [18]. However, mercury recycling companies for the in contrast, a study in a dental teaching dental profession. At a local level, it is The authors would like to acknowledge hospital in Turkey showed that hazard- recommended that the Pakistan Dental the crucial role played by the Depart- ous waste collection rules were obeyed Association takes a lead role in training ment of Community Dentistry, Col- most of the time [19]. dental professionals to adopt the best lege of Dentistry, Riphah International Despite its acknowledged health practices for amalgam waste manage- University, Islamabad in the provision hazards, amalgam is popular among ment in the dental setting. Implemen- of administrative costs incurred in con- Pakistani dentists as a dental restora- tation of protocols can be begin with ducting this research. tive material due to its durability and training and awareness-raising and at a The authors would also like to ex- low cost; therefore phasing it out or later stage, involve legislative control. press their gratitude to all the 2007 2nd banning its use will be difficult. Encour- Although our study group can be year dental students of the College of aging best management practices for assumed to be representative of the Dentistry, Riphah International Univer- amalgam waste, from an environmental average Pakistani dentist, the generaliz- sity, who volunteered their time for data perspective, is a more viable option but ability of the results of this study was collection. Special thanks are due to Dr one which requires strict adherence for limited by the cross-sectional design Faisal Moeen, Assistant Professor, Col- maximum effectiveness. Legislation for and partial use of a convenience sample. lege of Dentistry, Riphah International occupational and environmental safety The reliability of the questionnaire was University for his invaluable input and have paved the way for implementation not checked due to the anonymous advice.

References

1. Hörsted-Bindslev P. Amalgam toxicity—environmental and oc- 5. Anderson BA et al. Dental amalgam: a report with reference to cupational hazards. Journal of dentistry, 2004, 32:359–65. the medical devices directive 93/42/EEC from an Ad Hoc Working Group mandated by DGIII of the European Commission. Angel- 2. A–Z topics: best management practices for amalgam waste. holm, Sweden, Nordiska Dental AB, 1998. American Dental Association [online article] (http://www. 6. Risher JF. Too much of a good thing (fish): methyl-mercury ada.org/prof/resources/topics/amalgam_bmp.asp, accessed case study. Journal of environmental, 2004, 67:9–12. 26 August 2009). 7. Statistics. Pakistan Medical and Dental Council [website] 3. Swedish Council for Planning and Coordinating Research (FRN) (http://dev.plexushosting.com/PMDC/Statistics/tabid/103/ Stockholm. Report No. 2652, 19 February 1998. Helsetilsynet Default.aspx, accessed 29 October 2009). [Norwegian Board of Health Supervision] (http://www.hel- 8. Burkel F et al. Amalgam and composite use in UK general den- setilsynet.no, accessed 26 August 2009). tal practice in 2001. British dental journal, 2003, 11:613–8. 4. Lundberg S. Swedish views on questions put at the stakeholder 9. Widstrom E, Forss H. Dental practitioners’ experiences on the meeting concerning the mercury export ban and storage obliga- usefulness of restorative materials in Finland 1992–1996. Brit- tions of mercury 8th September 2005. Brussels, European Com- ish dental journal, 1998, 185:540–2. mission, Ministry of Sustainable Development, Division for 10. Mercury source control policies and procedures manual. Colo- Eco-management and Chemicals, 2005. rado, Colorado Springs Utilities Industrial Pretreatment

338 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Program, 2007. (http://www.csu.org/business/services/ 18. Hospitals and dental clinics handle mercury callously: report. wastewater/industrial-pretreatment/mercury/item2301.pdf, Express healthcare management, 2004, 16–31 July (http://www. accessed 29 October 2009). expresshealthcaremgmt.com/20040731/hospinews01.shtml, 11. Oral health in Pakistan: a situational analysis. Islamabad, Paki- accessed 26 August 2009). stan, Ministry of Health and World Health Organization, 2004. 19. Ozbek M, Sanin FD. A study of the dental solid waste produced 12. Khairuldean N, Sadiq WM. Amalgam safety and alternative in a school of dentistry in Turkey. Waste management, 2004, restorative materials: a cross sectional amongst dentists. Saudi 24(4):339–45. dental journal, 1996, 8(1):1–7. 20. American Dental Association’s comments on FDA’s proposed rule 13. Sadig W. Preliminary study on dentists’ perception and safety and special control guidance on dental amalgam products. Chi- measures toward the use of dental amalgam in Riyadh private cago, Illinois, American Dental Association, 2000 (Docket No. clinics. Saudi dental journal, 2007, 19(3):164–70. 01N-0067) (http://www.ada.org/prof/resources/positions/ 14. Countries. World Health Organization [website] (http://www. statements/statements_amalgam_fda.pd, accessed 26 August who.int/countries/en/, accessed 26 August 2009). 2009). 15. Darwish R, Al-Khatib IA. Evaluation of dental waste manage- 21. Occupational exposure to hazardous chemicals in laboratories. ment in two cities in Palestine. Eastern Mediterranean health Toxic and hazardous substances. Regulations (standards–29 journal, 2006, 12(Suppl. 2):217–23. CFR). Washington DC, United States Department of Labor, Oc- 16. Punchanuwat K, Drummond BK, Treasure ET. An investigation cupational Safety and Health Administration, 2006 (Standard of the disposal of dental clinical waste in Bangkok. International No. 1910.1450) (http://www.osha.gov/pls/oshaweb/owadisp. dental journal, 1998, 48(4):369–73. show_document?p_table=standards&p_id=10106, accessed 26 August 2009). 17. Kishore J et al. Awareness about biomedical waste management and infection control among dentists of a teaching 22. Chemicals: mercury programme. United Nations Environmental hospital in New Delhi, India. Indian journal of dental research, Programme [website] (www.chem.unep.ch/MERCURY, ac- 2000, 11(4):157–61. cessed 26 August 2009).

Healthcare waste management

WHO Programme activities on healthcare waste management include developing technical guidance materials for assessing the quantities and types of waste produced in different facilities, creating national action plans, developing national healthcare waste management (HCWM) guidelines and building capacity at national level to enhance the way HCW is dealt with in low income countries.

Further information on WHO’s work in healthcare waste management is available from the WHO Centre for Environmental Health Activities (CEHA) (http://www.emro.who.int/ceha/index.asp).

339 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Case report Disseminated leishmaniasis caused by Leishmania tropica in HIV-positive patients in the Islamic Republic of Iran S. Jafari,1 M. Hajiabdolbaghi,1 M. Mohebali,2 H. Hajjaran2 and H. Hashemian1

Introduction and visceral involvement. This is the first were observed (Figure 2). The culture such report from the Islamic Republic was strongly positive 2 weeks after in- Both cutaneous and visceral forms of of Iran. oculation. To determine the species, fol- leishmaniasis are prevalent in some parts lowing mass production in RPMI1640 of the Islamic Republic of Iran [1,2]. The (Gibco), all of the promastigotes were Mediterranean type of visceral leishma- Case reports analysed by a random amplification of niasis (VL) is also reported sporadically polymorphic DNA polymerase chain throughout the country [3] with canines Case 1 reaction (RAPD-PCR) technique and being the main animal reservoir hosts A 32-year-old man, an intravenous drug the results compared with standard spe- [4]. The cutaneous form of leishma- user with many tattoos, known for 5 cies of L. infantum (MCAN/IR/96/ niasis is seen in 2 forms: anthroponotic months to be HIV-positive and with a LON49), L. tropica (MHOM/IR/99/ and zoonotic. Anthroponotic cutane- history of hepatitis C and incompletely YAZ1) and L. major (MRHO/IR/75/ ous leishmaniasis (ACL) is caused by treated pulmonary tuberculosis, was ad- ER) using 4 RAPD-PCR primers, in- Leishmania tropica and is prevalent in mitted to our hospital in June 2003. The cluding AB1-07 (5’ GGT GAC GCA many large- and medium-sized cities in patient also had tuberculous abscess G), A4 (5’ AAT CGG GCT G ), 327 the country [1,5]. Zoonotic cutaneous and many non-ulcerative leproid nod- (5’ ATA CGG CGT C) and 329 (5’ leishmaniasis (ZCL) is caused by L. ma- ules distributed all over his face and he GCG AAC CTC C) in the School jor and is endemic in many foci in the had deformed ears (Figure 1). His CD4 of Public Health, Tehran University north, east and south of the country [6]. 3 cell count was 180/mm . Serological re- of Medical Sciences [11,12]. Dr K.P. Some studies have implicated L. tropica sults for the detection of anti-Leishmania Chang from Chicago University con- as another agent of VL in humans and antibodies (DAT, IFA, rK39 dipstick) firmed the Leishmania identification dogs in reports from the north-west and were negative [2,10]. Examination of with PCR-RFLP. south of the country [4,7,8]. peripheral blood smears disclosed no The patient was treated with anti Few data are available, however, Leishman bodies. Blood culture in about host immunological response and monophonic and biphasic media was retroviral drugs (zidovudine, lamivu- parasite destruction when leishmaniasis negative, bone marrow aspiration and dine, nelfinavir) plus an antituberculosis is associated with immunosuppressants. direct microscopic examination on regimen. We also started pentavalent At present, the majority of cases of HIV– Giemsa-stained smears revealed scant antimony (Glucantime) 20 mg/kg/day leishmaniasis co-infection reported in Leishmania amastigotes. Culture of for a period of 1 month and the patient the Mediterranean basin were caused bone marrow aspirate in monophonic was followed up during the treatment by L. infantum [9]. and biphasic media was positive, and course. Disseminated leishmaniasis caused promastigotes were seen 10 days after The CD4 count of the patient by L. tropica in patients with HIV in- inoculation. reached 300/mm3 2 months after the fection is uncommon. Here, we report Smears prepared from the skin le- treatment and a considerable improve- on 2 patients with HIV–leishmaniasis sions showed a great number of intracel- ment in skin lesions was seen although co-infection caused by L. tropica with lular Leishman bodies within histiocytes, the lesions had not completely disap- generalized and multiple skin lesions and even extracellular flagellated forms peared.

1Department of Infectious Diseases, Imam Khomeini Hospital; 2Department of Medical Parasitology & Mycology, School of Public Health, Tehran University of Medical Sciences, Islamic Republic of Iran (Correspondence to M. Mohebali: [email protected] ; [email protected]). Received: 08/08/07; accepted: 31/12/07

340 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

No fever, weight loss or other constitu- tional symptoms were observed. Except for multiple, small, mobile lymph nodes (maximum diameter 1.5 cm) without tenderness detected in the cervical, axil- lary and inguinal chains, other physical findings were normal. Mild splenom- egaly and a few small lymph nodes in the liver hilum and paraceliac area were reported using abdominal sonography. 3 His CD4 cell count was 180/mm . Direct examination of the skin lesion biopsy showed Leishman bodies within histiocytes. Bone marrow aspiration and direct microscopic examination of Giemsa-stained smears revealed Leishmania sp. Culture of bone marrow aspiration in monophasic and biphasic Figure 1 Distribution of the skin lesions on the face in case 1: erythematous media was positive and promastigotes papules and nodules with central ulceration covered with haematic crusts were observed about 4 weeks after inoculation. Serological tests for the detection of anti-Leishmania antibodies (DAT, IFA, rK39 dipstick) were posi- Case 2 He reported suffering multiple insect tive. To determine the species, we per- A 49-year-old man, who was an intrave- bites in Bam. formed RAPD-PCR as described above nous drug user and known to be HIV- A few months after travelling, a small [11,12] and L. tropica was confirmed. positive, was admitted to our hospital in papular lesion appeared on his left ankle Treatment with antiretroviral drugs (zi- June 2006 with skin lesions distributed that gradually became larger. He had dovudine, lamivudine, nevirapine) and on his face and extremities. itching, erythematous, scaling plaque in pentavalent antimony (Glucantime) 20 Previously, he had a history of travel same area, extending to the knees, elbow, mg/kg/day was started. to Bam, a city in the south-east of the flexor surface of the forearm, left side of The CD4 count of the patient rose 3 country which is endemic for ACL [1]. the forehead, eyelid, nose and right ear. to 210/mm 3 months after treatment. A considerable improvement in skin lesions was also seen after treatment.

Discussion

Disseminated leishmaniasis is a granulomatous disease, clinically characterized by ulcerated skin and visceral involvement, whose clinical manifestations can progress spontaneously. Visceral involvement is often caused byL. donovani complex (L. donovani, L. infantum, and L. chagasi), endemic in the Middle East and Mediterranean regions [13]. VL is endemic in north-western and southern Islamic Republic of Iran, and mainly affects children;L. infantum is the domi- Figure 2 Leishman bodies in the Giemsa-stained smear, prepared from skin nant species [1,3,4]. ACL caused by L. lesions of case 1 (× 1000) tropica is endemic in some cities of our

341 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

country [1,5]. In the Persian Gulf War and they are positive in 34%–55% of to the relative clearance of skin lesions (1990–91), L. tropica was the causative HIV-positive patients [16]. The clini- but relapses in patients are common and agent for leishmaniasis in American cal picture of disseminated cutaneous a long follow-up period is needed. soldiers at bases in Bahrain and north- leishmaniasis in HIV-positive patients The last, and perhaps most interest- ern Saudi Arabia [14]. In recent years, sometimes differs significantly from ing, point about these cases is that they Tehran and Bam have been recognized that of classic cutaneous leishmaniasis. were caused by L. tropica. In previous as endemic foci of ACL and interest- In HIV-negative patients with classic reports, disseminated cutaneous forms ingly, Patient 1 was resident in southern cutaneous leishmaniasis, bone marrow, of leishmaniasis were usually caused Tehran and Patient 2 had a history of blood smear and culture are usually by L. donovani complex and they were travel to Bam. negative but in leishmaniasis-associated accompanied by visceral involvement There may be a very long incuba- HIV, bone marrow, blood smear and [18]. In a study carried out in southern tion period of many years, so although culture may be positive, and skin lesions Islamic Republic of Iran, 63 of the 64 the trip to Bam in Patient 2 may have may be present in unusual locations. children with VL had infection with L. been made a long time ago, he may Typically, 2 or 3 relapses per patient infantum: there was, however, a single have contracted leishmaniasis in that occur in spite of the correct treatment patient infected with L. tropica [7]. This endemic area. [17]. patient was not immunocompromised It is known that the pathogenesis of The course of the disease is often and did not differ in any way from the leishmaniasis depends on the interac- related to the host immune response. other patients. Our cases are the first tion between antigen-presenting cells The suppression of the immunological Iranian reports showing unusual cuta- and the parasite involving immuno- mechanisms of the host leads to facili- neous lesions with visceral involvement logical mechanisms [15]. Disseminated tating the multiplication of the parasite caused by L. tropica in HIV–leishmani- leishmaniasis has been described in and its dissemination [15,17]. Although asis co-infection. This report shows that many conditions with diminished im- the CD4 cell count decreased during an L. tropica can produce disseminated mune response, including malignant opportunistic infection in these 2 HIV- cutaneous leishmaniasis with visceral disease, especially lymphoma, systemic positive patients, the low immunity/ extension in HIV-positive patients. lupus erythematous, renal transplanta- low CD4 count seems to be related to We recommend blood samples col- tion anti-rejection therapy, during cor- HIV infection as the primary factor for lected from HIV-positive patients in the ticosteroid therapy, and, more recently, dissemination of leishmaniasis. Islamic Republic of Iran are tested for in patients with HIV infection [13]. Cutaneous lesions in our HIV- Leishmania antigens to determine the Physicians should be aware of the atypi- positive patients were the first sign of prevalence of the infection, particularly cal clinical manifestations of VL in HIV- visceral involvement. Skin lesions were in endemic foci of leishmaniasis. positive patients, and should consider quite variable: macules, papules, plaques this diagnosis in cases of unexplained nodules and ulcers with many amastig- pyrexia, anaemia or pancytopenia, otes, particularly in the first patient, were Acknowledgement especially in endemic areas. Anaemia observed. The presence of Leishmania with or without leukopenia and throm- amastigotes has previously been detected We thank Dr K.P. Chang for carrying out bocytopenia may be found but may in healthy skin in patients with VL and PCR-RFLP using the primers designed be by treatment such as zidovudine. HIV co-infection [17]. In our patients, from the nagt gene at the University of Serological tests have low sensitivity treatment with pentavalent antimony led Chicago, United States of America.

References

1. Nadim A. Leishmaniasis. In: Azizi F, Janghorbani M, Hatam 3. Edrissian GhH et al. Visceral leishmaniasis: the Iranian experi- H, eds. Epidemiology and control of common disorders in Iran, ences. Archives of Iranian medicine, 1998, 1(1):22–6. 2nd ed. Tehran, Endocrine & Metabolism Research Center, 4. Mohebali M et al. Epidemiological aspects of canine visceral Shaheed Beheshti University of Medical Sciences Press, 2000. leishmaniasis in the Islamic Republic of Iran. Veterinary parasi- 2. Mohebali M et al. Application of direct agglutination test (DAT) tology, 2005, 129(3–4):243–51. for the diagnosis and seroepidemiological studies of visceral 5. Yaghoobi-Ershadi MR et al. A new focus of cutaneous leish- leishmaniasis in Iran. Iranian journal of parasitology, 2006, maniasis caused by Leishmania tropica. Saudi medical journal, 1(1):15–25. 2002, 23(3):291–4.

342 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

6. Mohebali M et al. Characterization of Leishmania infection in 12. Hajjaran H et al. Identification of Leishmania species isolated rodents from endemic areas of the Islamic Republic of Iran. from human cutaneous leishmaniasis, using random amplified Eastern Mediterranean health journal, 2004, 10(4–5):591–9. polymorphic DNA (RAPD-PCR). Iranian journal of public health, 7. Alborzi AV et al. Leishmania tropica-isolated patient with vis- 2004, 33(4):8–15. ceral leishmaniasis in southern Iran. American journal of tropical 13. Gilles HM, ed. Protozoal diseases. London, Arnold, 1999:452. medicine and hygiene, 2006, 74(2):306–7. 14. Magill AJ et al. Visceral infection caused by Leishmania tropica 8. Hajjaran H et al. Leishmania tropica: another etiological agent in veterans of Operation Desert Storm. New England journal of of canine visceral leishmaniasis in Iran. Iranian journal of public medicine, 1993, 329(20):1503–4. health, 2007, 36(1):85–8. 15. Kharazmi A et al. T-cell response in human Leishmaniasis. Im- 9. Altés J et al. Visceral leishmaniasis: another HIV-associated op- munology letter, 1999, 65:105–8. portunistic infection? Report of eight cases and review of the literature. AIDS, 1991, 5:201–7. 16. El Safi SH et al. A comparison of the direct agglutination test and enzyme- linked immunosorbent assay in the sero-diagnosis of 10. Edrissian GhH et al. Evaluation of rapid Dipstick rk39 test in leishmaniasis in the Sudan. Transactions of the Royal Society of diagnosis and serological survey of visceral leishmaniasis in Tropical Medicine and Hygiene, 1989, 83:334–7. humans and dogs in Iran. Archives of Iranian medicine, 2003, 6(1):29–31. 17. Alvar J. Leishmaniasis and AIDS co-infection: the Spanish example. Parasitology today, 1994, 10:160–3. 11. Noys HA et al. Appraisal of various random amplified polymorphic DNA-polymerase chain reaction primers for Leishmania 18. Herrera E et al. The presence and significance of Leishmania identification. American journal of tropical medicine & hygiene, in mucocutaneous biopsies from HIV+ patients with visceral 1996, 55:98–105. leishmaniasis. European journal of dermatology, 1996, 6:51.

Joint EMRO/TDR Small Grants Scheme for Operational Research in Tropical and Other Communicable Diseases: 18th call for proposals 2010

The Eastern Mediterranean Regional Office (EMRO) of the World Health Organization (WHO) in collaboration with the UNICEF/UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR) is pleased to announce the 18th call for proposals of the Small Grants Scheme for Operational Research in Tropical and other Communicable Diseases for the year 2010. The scheme is co-funded by the WHO/EMRO and the UNICEF/ UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR). The deadline for application is 31 March 2010. The Research Proposal Form and further information about the Small Grants Scheme is available at: http://www.emro.who.int/TDR/pdf/TDR_Small_Grants_Scheme_010.pdf

343 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Case report Lupoid leishmaniasis due to Leishmania major with remaining large scars: report of 2 cases G. Sadeghian,1 H. Ziaei,1 L. Shirani-Bidabadi 1 and M.A. Nilforoushzadeh 1

Introduction several years and caused large scars on leishmaniasis in Afghanistan 10 years the face. before but some red infiltrative nodules Lupoid leishmaniasis, a clinical form of appeared at the margin of the remain- cutaneous leishmaniasis, was described ing scars 2 years before referring to our by Christopherson in 1923 [1]. It is Case reports centre (Figure 1). He complained of a chronic condition that typically fol- Case 1 new lesions at the margin of the old lows acute cutaneous leishmaniasis scars which were progressing. A 17-year-old Afghan man presented infection. In this clinical form, 1 to 2 Direct smear was negative for Leish- to the Skin Disease and Leishmaniasis years after healing of the acute lesion, man bodies, leishmanin test was posi- Research Center (SDLRC) in 2007 new papules and nodules appear at tive (8 mm induration) and Mantoux with 2 large atrophic plaques on both the margin of the remaining scar. The test was negative. papules have a granulomatous, lupoid sides of his face and 1 on his right ear. Polymerase chain reaction (PCR) appearance and are often associated He had been infected with cutaneous was used to identify the Leishmania spe- with scaling. They characteristically cies, which was found to be . present at the margin of the scarred L. major area. Less commonly lesions may The patient, who had not previously present months or years after the acute undergone any treatment for cutaneous lesion has healed [2]. Most reported leishmaniasis, was treated with systemic cases are associated with old world meglumine antimoniate (Glucantime, strains of leishmaniasis and Leishmania Aventis, France) 20 mg/kg per day in- tropica is responsible for the major- tramuscularly for 20 days. His response ity of such cases [3]. The incidence of to this treatment was good and the mar- lupoid leishmaniasis has been reported ginal nodules disappeared (Figure 2). in previous studies to range from 0.5% The patient was followed up for 1 year to 6.2% [4,5]. and showed no sign of recurrence. Histological features of this con dition include well-organized epithelioid granulomas, surrounded by lymphocytes, amastigotes are absent and culture for leishmaniasis is frequently negative [4]; however, the leishmanin test is usually positive [5]. Lupoid cutaneous leishmaniasis is prevalent in endemic area of leishmaniasis, particularly in the Middle Figure 1 Patient 1 showing large East and Afghanistan [6]. In this report atrophic plaques on both sides of his we describe 2 cases of lupoid leish- face. There were some red infiltrative nodules at the margin of the lesions. Figure 2 Patient 1 after treatment maniasis that were not diagnosed for

1Skin Disease and Leishmaniasis Research Center, Isfahan University of Medical Sciences, Isfahan, Islamic Republic of Iran (Correspondence to G. Sadeghian: [email protected]). Received: 16/10/07; accepted: 09/12/07

344 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

Case 2 Differential diagnosis is difficult and A 19-year-old Afghan man presented in may depend on the detection of a few 2007 with a large atrophic scar on one leishmania amastigotes in the histologi- side of his face and nose. There were cal sections, the growth of the promas- some red infiltrative nodules at the mar- tigotes in cultures, or the identification gin of the lesion which had appeared of amastigotes by other techniques. 3 years before referring to the SDLRC PCR has been used to detect leishma- (Figure 3).The patient complained nia amastigote DNA in tissue samples of recurrence and progression of his obtained from patients with chronic disease. lupoid leishmaniasis [5]. Direct smear was negative for Leish- In a study in the Islamic Republic of Figure 3 Patient 2 showing large Iran, L. major was identified in 63 patients man bodies and histological examination atrophic scar on one side of his face showed epidermal hyperplasia, severe and nose with red infiltrative nodules with chronic lupoid leishmaniasis [9]. infiltration of lymphoplasma cells and In our 2 reported cases, the patients had epithelioid granuloma without caseous not been diagnosed despite having had necrosis or Leishman bodies. Leishmanin the condition for several years because test was positive (10 mm induration) direct smears were negative. The delay and Mantoux test was negative. L. major in the diagnosis led to progression of was identified using PCR. the lesions, which caused the large scars This patient had also not been treated on their faces. It was the use of PCR that for leishmaniasis before and was treated allowed us to establish the diagnosis similarly with systemic meglumine anti- and identify the parasite responsible for moniate (Glucantime, Aventis, France) the lesions as L. major. Both the patients 20 mg/kg per day intramuscularly for were successfully treated with systemic 20 days. His response to treatment was meglumine antimoniate. also good and the marginal nodules We recommend that the diagnosis disappeared (Figure 4). He too was Figure 4 Patient 2 after treatment of lupoid leishmaniasis be considered followed for 1 year and showed no sign in all patients from endemic areas of of recurrence. leishmaniasis who present with chronic leishmaniasis cases [7]. Although not lesions, especially with facial lesions. Al- as destructive as lupus vulgaris, lupoid though cutaneous leishmaniasis is self- Discussion leishmaniasis may persist and spread limiting, if the lupoid form of cutaneous slowly for many years [8]. leishmaniasis is left untreated, it will Lupoid leishmaniasis, also known as The clinical and histopathological progress and destroy the skin leading chronic or relapsing leishmaniasis, is features closely resemble lupus vulgaris, to large scars. Early diagnosis and treat- widespread in the Middle East, where leading to diagnostic difficulty in distin- ment are vitally important to prevent it represents up to 5% of all cutaneous guishing between these 2 conditions. this complication.

References

1. Christopherson JB. Lupus leishmaniasis: a leishmaniasis of the 6. Gurel MS, U Lukanligil M, Ozbilge H. Cutaneous leishmaniasis skin resembling lupus vulgaris: hitherto unclassified. British in Sanliurfa: epidemiologic and clinical features of the last four journal of dermatology, 1923, 35:123–31. years (1997–2000). International journal of dermatology, 2002, 2. Bowling JCR, Vega-Lopez F. Clinicopathological case. Clinical 41(1):32–7. and experimental dermatology, 2003, 28(6):683–4. 7. Paradisi M et al. Un caso di leishmaniosi lupoide in eta pedi- 3. Asillian A et al. Carbon dioxide laser for the treatment of lupoid atrica diagnostica mediante PCR [A pediatric case of lupoid cutaneous leishmaniasis (LCL): A case series of 24 patients. leishmaniasis diagnosed by PCR]. Minerva pediatrica, 2001, Dermatology online journal, 2006, 12(2):3. 53(1):33–7. 4. Nilforoushzadeh MA et al. The therapeutic effect of combined 8. Even-Paz Z, Sagher F. Some basic medical problems illustrated cryotherapy, paramomycin, and intralesional meglumine anti- by experiments with cutaneous leishmaniasis. South African moniate in treating lupoid leishmaniasis and chronic leishmani- medical journal, 1961, 35:565–81. asis. International journal of dermatology, 2006, 45(8):989–91. 9. Momeni AZ et al. Chronic lupoid leishmaniasis. Evaluation 5. Aradeh S et al. Studies on chronic (lupoid) leishmaniasis. An- by polymerase chain reaction. Archives of dermatology, 1996, nals of tropical medicine and parasitology, 1980, 74(4):439–45. 132(2):198–202.

345 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Case report A case of Behçet disease with pulmonary artery pseudoaneurysm: long term follow-up M-H.S. Modaghegh,1 G. H. Kazemzadeh 1 and M. H. Jokar 2

Introduction vasculitis lead to occlusion of the vessels genital ulcerations, unilateral oedema or aneurysm formation [7]. Differential of the lower limb, blurred vision and Behçet disease (BD) is a chronic inflam- diagnosis of Behçet-induced vasculitis polyarthralgia. On physical examina- matory disorder of unknown etiology has been described by Kreuger, Hoff- tion vital signs were stable. Respiratory with a clinical spectrum that has greatly man and Merton [8]. sounds were decreased in the left lung. expanded since it was first described More than 200 cases of BD with Laboratory tests revealed erythrocyte in 1937 by Hulusi Behçet as a triple pulmonary involvement have been re- sedimentation rate = 152 mm/h and complex of recurrent oral and genital ported in the literature. The pulmonary anaemia. At the age of 22 years, follow- ulcers and uveitis [1]. arteries are the second most common ing amnesia for 19 days, she had been Although BD has a worldwide distri- site of arterial involvement after the aor- referred to a neurologist and examina- bution, it is mainly seen in Far East and ta. Aneurysms are more common than tion revealed bilateral papillary oedema Middle East countries [2,3]. The high- thrombosis [2]. Aneurysms associated and raised intracranial pressure. The est prevalence is reported from Turkey, with BD tend to be multiple and pul- patient was given oral acetazolamide, 80–370 per 100 000, and ranges from 2 monary artery aneurysms are relatively 125 mg twice a day. to 30 cases per 100 000 in other Asian common, with an associated rate of Regarding her past medical history, countries (16.7 per 100 000 in the Is- 1%–10% of patients with BD [9]. Hae- ophthalmologic consultation was per- lamic Republic of Iran) [2,3]. The age moptysis of varying degrees (up to 500 formed and a mild retinal vasculitis was of disease onset is usually in the second mL) is the most common and predomi- found via ophthalmoscopy. Extensive or third decade of life and the male to nant symptom [2]. Medical treatment deep vein thrombosis extending to the female ratio is reported to be almost is more beneficial when given in the inferior vena cava was demonstrated in equal. However, the disease runs a more early stage of the disease [10]. Surgical colour Doppler study. According to the severe course in men and in those with repair of pulmonary artery aneurysm in international criteria for diagnosis [4], onset before 25 years of age [2]. BD carries a high risk, with high rates of she was diagnosed as having BD (recur- Since BD does not have pathogno- morbidity and mortality [11]. Recently, rent mouth and genital aphthous ulcers monic symptoms or laboratory findings; endovascular management has been and eye lesion). She was prescribed 750 diagnosis is made on the basis of the attempted for treatment of vascular in- mg pulse cyclophosphamide monthly, criteria proposed by the International volvement. However, when aneurysms with warfarin 2.5 mg/day and methyl Study Group for Behçet disease in 1990 are multiple, endovascular treatment is prednisolone 1 g/day for 3 days, fol- [4]. According to the criteria, recurrent still difficult and challenging. lowed by oral prednisolone 1 mg/kg per oral ulceration must be present along To our knowledge, there is no report day, tapered gradually and continued at with at least 2 of the following: recur- on a patient with 4 or more arterial an- a low dose. Two weeks later, oedema of rent genital ulceration, eye lesions, skin eurysms treated solely by endovascular the lower limb had completely resolved, lesions, a positive pathergy test [2,4,5]. management [12,13]. thrombus size had substantially de- Many disorders, such as cutaneous creased and the patient was discharged small vessel vasculitis, inflammatory eye in good condition. disease, neurological disease, vascular Case report After 1 year, she was admitted again disease, arthritis, are associated with to the internal medicine ward, with a the presence of BD [6]. Pathological A 32-year-old woman was admitted 5-month history of dyspnoea, cough and features such as small- or large-vessel to the hospital with painful oral and haemoptysis. Purified protein derivative

1Mashhad Vascular and Endovascular Surgery Research Center; 2Department of Rheumatology, Mashhad University of Medical Sciences, Mashhad, Islamic Republic of Iran (Correspondence to M-H.S. Modaghegh: [email protected]). Received: 16/07/07; accepted: 25/02/08

346 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

into consideration her previous imaging, we decided to perform elective surgery. Three days later, the patient felt better and suffered less from haemoptysis. Under general anaesthesia with 1 lung intubation, left thoracotomy was performed. The upper lobe was col- lapsed and the lower lobe remained expanded because of the contained aneurysm. The left pulmonary artery was proximally controlled at the origin. Further exploration revealed a 10 × 10 cm aneurysm contained in the lower lobe and intact upper lobe branch. The aneurysm was opened. All lower lobe branches of the pulmonary artery were involved in the aneurysm. Resection of the lower lobe was performed and infla- Figure 1 Computerized tomography scan of the thorax with contrast showing a tion of the upper lobe was controlled large cavity at the inferior lobe with extension to the hill and adhesion to the chest wall, indicating a large pulmonary artery aneurysm before closure. The gross pathological examination of the specimen revealed a pseudo test and sputum smear were performed another attack of productive cough and aneurysm at the lower lobe in addi- 3 times but the results were negative. massive haemoptysis and was referred tion to bleeding into the adjacent lung No history of recent or remote trauma, to the Department of Vascular Surgery. parenchyma. Further histopathological drug abuse or infective endocarditis was Since she continued to suffer from hae- evaluation revealed focal coagulation elicited. On examination, the patient moptysis intermittently while tapering necrosis in addition to neutrophil–cell appeared chronically ill and pale; her the dosage of prednisone, and taking reaction in the wall of the pulmonary blood pressure was 100/70 mmHg and pulse rate was 70/min; renal and liver function tests were normal; S2 splitting was detected on chest examination; oral and genital lesions were noted. White blood count was 8.06 × 103/μL, haemo- globin was 10.0 g/dL, haematocrit was 32.1% and erythrocyte sedimentation rate was 41 mm/h. Chest X-ray showed a 6 × 6 cm opac- ity in the inferior lobe of the left lung. Computerized tomography (CT) scan of the thorax with intravenous contrast revealed an enhanced large cavity at the inferior lobe with extension to the hill and adhesion to the chest wall indicating a large pulmonary artery aneurysm (Figure 1). The diagnosis was further confirmed by a digital subtraction angiography (Figure 2). The patient started to receive 1 mg/kg prednisone daily and 750 mg cyclophosphamide intravenously. After 17 days, she experienced Figure 2 Digital subtraction angiography showing a cavity at the pulmonary artery

347 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

artery. Focal haemorrhagic necrosis and lymphoplasmocytic inflammatory reaction were detected in the lung parenchyma. The patient was discharged in good condition after 36 days. She has been under close observation for 50 months without suffering relapse of the symptoms or new complaints (Figure 3).

Discussion

BD is a systemic vasculitis affecting virtually all types and sizes of vessels. Pulmonary manifestations in patients are mainly related to vasculitis of the pulmonary arteries, veins and septal capillaries. Pulmonary vascular involve- Figure 3 Chest X-ray showing no abnormality 2 months after surgery ment can lead to aneurysm formation (pulmonary artery aneurysm), throm- botic occlusion, mainly in the vena cava, pulmonary infarction and pulmonary combination with steroids is most the aneurysm sac. Therefore consid- haemorrhage. Venous system involve- beneficial when given in the early ering the special characteristic of the ment is more common than arterial stage of the disease before irreversible aneurysm and the high risk of rupture, system involvement; however, rupture damage to the arterial wall develops resection was performed. The patient of an arterial aneurysm remains a major [7]. Surgery is available for resecting was discharged 36 days after the op- cause of mortality related to BD. Rup- arterial aneurysms and replacing grafts. eration, which was much longer than ture of an aneurysm with erosion into a Many attempts have been made to the expected hospitalization period; bronchus and the development of in situ manage these serious lesions by surgi- this was due to prolonged moderate thrombosis from active vasculitis have cal treatment. Despite several reports air leakage from the chest tube, which been suggested as explanations for the concerning recurrence after surgical was eventually sealed. Although some haemoptysis. Sudden hilar enlargement management in about half of cases, our or the appearance of polylobular and patient had no recurrent haemoptysis, articles mention fatal outcomes after round opacities on the chest radiograph productive cough or aneurysm in 50 lobectomy [15,17], our surgery was can represent pulmonary artery aneu- months of follow up [13,14]. completely successful and had a satis- rysms. As in our case, the most com- As in our patient, lung lobectomy factory result after a long term follow- mon symptom of pulmonary arterial has been reported in several articles up. aneurysm is haemoptysis. [15–18]. In our patient, the upper lobe In conclusion, pulmonary lobec- Medical treatment with im- of the left lung was intact and all of tomy can be performed safely as a treat- munosuppressive drugs alone or in the lower lobe branches arose from ment for pulmonary aneurysm in BD.

References

1. Hammami S et al. Intracardiac thrombus in Behçet’s disease: 5. Uzun O, Akpolat T, Erkan L. Pulmonary vasculitis in Behçet two case reports. Thrombosis journal, 2005, 3:9. disease. Chest, 2005, 127(6):2243–53. 2. Erkan F, Gül A, Tasali E. Pulmonary manifestations of Behçet’s 6. Barnes CG, Yazici H. Behçet’s syndrome. Rheumatology, 1999, disease. Thorax, 2001, 56(7):572–8. 38(12):1171–4. 3. Davatchi F et al. Behçet’s disease-Analysis of 3443 cases. APLAR 7. Çil BE et al. Embolization of a giant pulmonary artery aneurysm journal of rheumatology, 1997, 1:2–5. from Behçet disease with use of cyanoacrylate and the Bub- 4. International study group for Behçet’s disease. Criteria for diag- ble technique. Journal of vascular and interventional radiology, nosis of Behçet’s disease. Lancet, 1990, 335(8697):1078–80. 2005, 16(11):1545–9.

348 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

8. Krueger GRF, Hoffmann A, Merten UP. Systemic vasculitis, 14. Okada K et al. Surgical management of Behçet’s aortitis: primary and secondary differential diagnosis and classification. a report of eight patients. Annals of thoracic surgery, 1997, WASPaLM Global Pathology Supercourse, 2000 [internet 64(1):116–9. monograph] (http://www.dokkyomed.ac.jp/dep-k/cli-path/ 15. Tuzun H et al. Surgical therapy of pulmonary arterial aneu- a-super/vasculitis/super-vas.html, accessed 09 August 2009). rysms in Behçet’s syndrome. Annals of thoracic surgery, 1996, 9. Gebitekin C et al. Fatal hemoptysis due to pulmonary artery 61(2):733–5. aneurysm in Behçet’s disease. European journal of vascular & endovascular surgery, 1997, 13(2):233–6. 16. Ceyran H, Akcali Y, Kahraman C. Surgical treatment of vasculo-Behçet’s disease: a review of patients with con- 10. Tunaci M et al. CT findings of pulmonary artery aneurysms dur- comitant multiple aneurysms and venous lesions. Vasa, 2003, ing treatment for Behçet’s disease. American journal of roentge- nology, 1999, 172(3):729–33. 32(3):149–53. 11. Bozkurt AK. Embolization in Behçet’s disease. Thorax, 2002, 17. Yassine N et al. Anevrysmes e l’artere pulmonaire au cours 57(5):469–70. de la maladie de Behçet. A propos de 5 nouvelles observations [Aneurysms of the pulmonary artery in Behçet disease. 12. Cantasdemir M et al. Emergency endovascular management of pulmonary artery aneurysms in Behçet’s disease: report of two Apropos of 5 new cases]. Revue de pneumologie clinique, 1997, cases and a review of the literature. Cardiovascular and interven- 53(1):42–8. tion al radiology, 2002, 25(6):533–7. 18. Salamon F et al. Massive hemoptysis complicating Be- 13. Hama Y et al. Endovascular management of multiple arte- hçet’s syndrome: the importance of early pulmonary an- rial aneurysms in Behçet’s disease. British journal of radiology, giography and operation. Annals of thoracic surgery, 1988, 2004, 77(919):615–9. 45(5):566–7.

Eastern Mediterranean Approach to Noncommunicable Diseases

Chronic diseases such as cardiovascular, diabetes, cancer, renal, genetic and respiratory conditions are rising dramatically in the Eastern Mediterranean Region. Currently, 45% of the region’s disease burden is due to noncommunicable diseases. It is expected that this burden will rise to 60% by the year 2020. The impact of these conditions falls heavily on the region’s poor and marginalized populations.

The Eastern Mediterranean Approach to Noncommunicable Diseases has the overall objective to improve the health of populations of the Region by reducing mortality and morbidity from major NCDs through an integrated collaborative intervention programme of prevention and health promotion.

Further information about the work of EMRO in tackling noncommunicable diseases s is available at: http://www. emro.who.int/ncd/ncd_introduction.htm

349 EMHJ • Vol. 16 No. 3 • 2010 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale

Case report Ascaris lumbricoides infection: an unexpected cause of pancreatitis in a western Mediterranean country A. Galzerano,1 E. Sabatini 1 and D. Durì 2

Introduction effusion, pancreatic oedema, dilated Discussion gallbladder with a bile duct measuring Ascaris lumbricoides is a nematode 1.1 cm with no lithiasis, left pleural effu- Ascaris lumbricoides infestation is ac- parasite, endemic in the Middle East sion and basal atelectasis. Endoscopic quired through ingestion of eggs in raw and South America, especially in rural retrograde cholangiopancreatography vegetables. The human is the definitive countries. Ascariasis infection causes showed a dilated bile duct with a patent host. Ingested larvae penetrate the intes- about 20 000 deaths every year [1], ampulla with no lithiasis. tinal lymphatic and venous vessels and usually as a result of intestinal occlusion, The day after admission the patient through the portal vein reach the right and it contributes to infant malnutri- underwent cholecystectomy, cholan- heart, pulmonary circulation and the tion [2]. Poor sanitation is usually the giogram, positioning of Kher drainage alveoli. After alveolar rupture they pass most important risk factor for infection, and pancreatic necrosectomy. Due to into the trachea and the pharynx, are and women are more affected because haemodynamic instability and respira- then swallowed; after about 2 months progesterone plays a role in inducing tory failure the patient was then admitted they reach maturity. In the bowel Oddi’s sphincter relaxation, allowing to the intensive care unit. At admission nematodes can perforate the intestinal the nematode to access the biliary duct she was apyretic and microbiological wall, be ejected from the mouth or anus [3]. Although not common in devel- cultures from abdominal drainage spec- and penetrate the biliary ducts or the oped countries, ascariasis infection is imens were negative. After weaning and airways. The infestation can present as increasingly likely to be encountered by extubation the patient was transferred to a wide range of symptoms: intestinal clinicians because of the growing rates the surgical ward where she underwent perforation or occlusion, cholangitis, of travel to developing countries and an unremarkable recovery. obstructive jaundice, acute pancreatitis increased migration. About 20 days after admission she or appendicitis, pneumonia and respira- developed fever, nausea, vomiting, tory failure and allergic reactions to the marked eosinophilia that had not been ascaris antigen. In most cases, however, Case report noticed before (total leukocyte count patients present with unspecific symp- 11.8 × 103/μL, eosinophils 10%) and a toms and sometimes the diagnosis is We describe the case of a 78-year-old maculopapular rash. On the hypothesis incidental [3]. Italian woman who had never travelled of iatrogenic allergic dermatitis, ster- The diagnosis is usually made by abroad, who was admitted to the surgi- oid and antihistamine treatments were abdominal ultrasonography, revealing cal ward of A. Murri Hospital, Fermo, started, with no benefit. biliary duct dilation and the presence Italy, with fever (temperature 38 °C), One week later the patient vomited of the parasite, a hyperechoic linear leukocytosis (white blood cell count a 5 cm male ascarid nematode. Therapy structure with a hypoechogenic line in- 3 15.4 × 10 /µL), hyperamylasaemia with mebendazole 100 mg twice daily side, which is sometimes motile [3–5]. (serum amylase level 260 U/L) and was started with prompt resolution of Ultrasonography is also the gold stand- abdominal pain. the pancreatic oedema, as documented ard technique for follow-up. CT scan The patient underwent abdominal by CT scans. The patients’ subsequent and nuclear magnetic resonance im- ultrasonography and a computerized recovery was uneventful and she was aging can also be helpful. Endoscopic tomography (CT) scan of the abdomen discharged 48 days after initial admis- retrograde cholangiopancreatogra- and thorax, which revealed peritoneal sion. phy is the gold standard method for

1Department of Anaesthesia and Intensive Care, Santa Maria della Misericordia Hospital, Perugia, Italy. 2Department of Anaesthesia and Intensive Care, Sant Antonio Hospital, San Daniele del Friuli, Italy (Correspondence to D. Durì: dav.anestesia@ gmail.com). Received: 13/01/08; accepted: 09/03/08

350 املجلة الصحية لرشق املتوسط املجلد السادس عرش العدد الثالث

identifying and removing the nematode effusion at the time the examinations to any endemic areas, our hypothesis is a from the duodenal, biliary or pancreatic were made. contact with eggs through consumption tract [3]. Although ascariasis is the most of raw vegetables or contaminated soil. In this case neither CT scans nor common human worm infection in the The presence of eosinophilia should endoscopic retrograde cholangiopan- Mediterranean area, the development of have raised suspicion of the possibility creatography was able to reveal the a severe illness such as a pancreatitis due of a parasitic infection, even in a patient presence of the parasite, which probably to this infestation is unusual [6–9]. The not travelling or migrating from endemic had already migrated to the left lung, origin of the infestation was not estab- areas, but the rarity of this cause of acute causing basal atelectasis and pleural lished. As the patient had not travelled abdomen was certainly misleading.

References

1. Khuroo MS. Ascariasis. Gastroenterology clinics of North Ameri- langitis in our climate]. Gastroentérologie clinique et biologique, ca, 1996, 25:553–77. 1991, 15:660–1. 2. Villamizar E et al. Ascaris lumbricoides infestation as a cause of 7. Moulinier C, Battin J, Giap G. Evolution du taux de prevalence intestinal obstruction in children: experience with 87 cases. de quatre parasites intestinaux chez l’enfant [Development Journal of pediatric surgery, 1996, 31:201–4. of the prevalence rate of four intestinal parasites in children]. 3. Misra SP, Dwivedi M. Clinical features and management of biliary ascariasis in a non-endemic area. Postgraduate medical Pediatrie, 1990, 45:129–32. journal, 2000, 76:29–32. 8. Mosiello G et al. Ascaridiasi come possibile causa di addo- 4. Hoffmann H et al. In vivo and in vitro studies on the sonographi- me acuto anche in Italia: presentazione di un caso clinico cal detection of Ascaris lumbricoides. Pediatric radiology, 1997, [Ascariasis as a cause of acute abdomen: a case report]. La 27:226–9. pediatria medica e chirurgica, 2003, 25:452–4. 5. Ferreyra NP, Cerri GG. Ascariasis of the alimentary tract, liver, pancreas and biliary system: its diagnosis by ultrasonography. 9. De la Cruz Alvarez J et al. Ascariasis biliopancreática: una en- Hepatogastroenterology, 1998, 45:932–7. tida infrecuente en nuestro medio [Biliopancreatic ascariasis: 6. Petit A et al. L’ascaridiose: une cause d’angiocholite peu an infrequent disease in our environment]. Gastroenterología y banale sous nos climats [Ascariasis: an unusual cause of cho- hepatología, 1996, 19:210–2.

351

Members of the WHO Regional Committee for the Eastern Mediterranean املجلة الصحية لرشق املتوسط Afghanistan . Bahrain . Djibouti . Egypt . Islamic Republic of Iran . Iraq . Jordan . Kuwait . Lebanon هى املجلة الرسمية التى تصدر عن املكتب اإلقليمى لرشق املتوسط بمنظمة الصحة العاملية. وهى منرب لتقديم Libyan Arab Jamahiriya . Morocco . Oman . Pakistan . Palestine . Qatar . Saudi Arabia . Somalia السياسات واملبادرات اجلديدة ىف اخلدمات الصحية والرتويج هلا، ولتبادل اآلراء واملفاهيم واملعطيات الوبائية Sudan . Syrian Arab Republic . Tunisia . United Arab Emirates . Republic of Yemen ونتائج األبحاث وغري ذلك من املعلومات، وخاصة ما يتعلق منها بإقليم رشق املتوسط. وهى موجهة إىل كل أعضاء املهن الصحية، والكليات الطبية وسائر املعاهد التعليمية، وكذا املنظامت غري احلكومية املعنية، واملراكز املتعاونة مع منظمة الصحة العاملية واألفراد املهتمني بالصحة ىف اإلقليم وخارجه.

EMHJ is a trilingual, peer reviewed, open access journal and the full contents are freely available at its website: http://www/emro.who.int/emhj.htm

Cover 6.indd 2 5/24/2010 11:46:47 AM Contents

Letter from the Editor...... 241

املجلد السادس عرش / عدد Volume 16 / No. 3 3 آذار / مارس March / Mars 2010

Cover 3.indd 1 2/16/2010 11:26:44 AM