BMJ: British Medical Journal

20 January 2007 (Vol 334, No 7585 , pp. 103-162)

Editor's choice Editor's choice: A redesigned bmj.com Tony Delamothe BMJ 2007;334, doi:10.1136/bmj.39098.502836.47

US editor's choice: Milestones, tombstones, and sex education Douglas Kamerow BMJ 2007;334, doi:10.1136/bmj.39099.504583.3A

Editorials Sexual health in adolescents Trevor Stammers BMJ 2007;334:103-104, doi:10.1136/bmj.39087.374653.BE

Iron and zinc deficiency in children in developing countries Zulfiqar A Bhutta BMJ 2007;334:104-105, doi:10.1136/bmj.39094.513924.BE

Tuberculosis in resource poor countries Robert Colebunders, Ludwig Apers, Greet Dieltiens, William Worodria BMJ 2007;334:105-106, doi:10.1136/bmj.39092.388796.80

Surgery for cataract Christopher M Wood BMJ 2007;334:107, doi:10.1136/bmj.39093.388900.80

Renaming schizophrenia Jeffrey A Lieberman, Michael B First BMJ 2007;334:108, doi:10.1136/bmj.39057.662373.80

Letters This week's letters

Case management: Government should have respected evidence David Oliver BMJ 2007;334:109, doi:10.1136/bmj.39094.396123.1F

Case management: Community matrons do make a difference Martin J Howard BMJ 2007;334:109, doi:10.1136/bmj.39094.383796.1F

National Sickness Service: Fundamental problem is not lack of preventive care Richard J Lyus BMJ 2007;334:109, doi:10.1136/bmj.39094.381366.1F Surgical follow-up by GPs: Proposal strikes at the heart of medicine Andrew N Bamji BMJ 2007;334:109-110, doi:10.1136/bmj.39094.393912.1F

Smokers and surgery: The best time to die (to help fellow taxpayers) Robert A Da Prato BMJ 2007;334:110, doi:10.1136/bmj.39094.388588.1F

Anatomy crisis: Make surgeons more active in teaching anatomy at all levels Sanjay Purkayastha, Paraskeva Paraskevas, Ara Darzi BMJ 2007;334:110, doi:10.1136/bmj.39094.394606.1F

Child safety in cars: Seats must be secured when children are not using them Adam C Frosh, Jonathan Wernick BMJ 2007;334:110, doi:10.1136/bmj.39094.386215.1F

News Lack of hospital beds causes emergency departments to miss targets Susan Mayor BMJ 2007;334:111, doi:10.1136/bmj.39098.461968.DB

BMJ readers choose the "sanitary revolution" as greatest medical advance since 1840 Annabel Ferriman BMJ 2007;334:111, doi:10.1136/bmj.39097.611806.DB

UK regulatory body wants public consultation on human-animal hybrid research Susan Mayor BMJ 2007;334:112, doi:10.1136/bmj.39097.380313.DB

Democrats push for stem cell research in US Janice Hopkins Tanne BMJ 2007;334:112, doi:10.1136/bmj.39094.683056.94

Compulsive gamblers must get free NHS treatment, BMA says Lisa Hitchen BMJ 2007;334:113, doi:10.1136/bmj.39098.396563.DB

Government suffers its first defeat over Mental Health Bill Clare Dyer BMJ 2007;334:113, doi:10.1136/bmj.39097.637049.DB

In Brief: News BMJ 2007;334:114, doi:10.1136/bmj.39097.458044.4E

NHS is accused of "appalling lack of workforce planning" Sally Hargreaves BMJ 2007;334:114, doi:10.1136/bmj.39097.341944.DB

One in four trainee doctors have concerns about their NHS career, study shows Sally Hargreaves BMJ 2007;334:114, doi:10.1136/bmj.39097.605822.DB

Fertilisation authority raids controversial fertility clinics Lynn Eaton BMJ 2007;334:115, doi:10.1136/bmj.39097.746030.4E

Group asks US National Institutes of Health to reveal industry ties Janice Hopkins Tanne BMJ 2007;334:115, doi:10.1136/bmj.39097.388218.DB

High court upholds GMC's rejection of case against neurologist Clare Dyer BMJ 2007;334:116, doi:10.1136/bmj.39112.756262.E0

Plan to halve MRSA cases by 2008 is probably not achievable, memo says Michael Day BMJ 2007;334:116, doi:10.1136/bmj.39094.632697.DB

US health spending grew more slowly in 2005, but it's not all good news Janice Hopkins Tanne BMJ 2007;334:117, doi:10.1136/bmj.39097.487616.4E

GMC accuses doctor of handing out slimming pills "as if they were Smarties" Owen Dyer BMJ 2007;334:117, doi:10.1136/bmj.39098.554734.DB

Hospital withdraws letter to GPs saying how patients can jump queues Michael Day BMJ 2007;334:117, doi:10.1136/bmj.39098.337454.DB

Arabian Peninsula states launch plan to eradicate malaria May Meleigy BMJ 2007;334:117, doi:10.1136/bmj.39097.499641.4E

EU is urged to press for global ban on mercury Rory Watson BMJ 2007;334:117, doi:10.1136/bmj.39097.341076.DB

Shortcuts from other journals: Candidate drug for familial hypercholesterolaemia needs more work BMJ 2007;334:118, doi:10.1136/bmj.334.7585.118

Shortcuts from other journals: Adherence to drug treatments prolongs survival after heart attack BMJ 2007;334:118, doi:10.1136/bmj.39093.538542.80

Shortcuts from other journals: Syphilis returns to China BMJ 2007;334:118, doi:10.1136/bmj.334.7585.118-b

Shortcuts from other journals: Many Canadian immigrants are susceptible to measles, mumps, or rubella BMJ 2007;334:118-119, doi:10.1136/bmj.334.7585.118-c

Shortcuts from other journals: Race is a relentless and exhausting burden for some black American doctors BMJ 2007;334:119, doi:10.1136/bmj.334.7585.119

Shortcuts from other journals: US prisoners have high death rates in the weeks after release BMJ 2007;334:119, doi:10.1136/bmj.334.7585.119-a

Shortcuts from other journals: Patients with drug eluting stents do better on long term clopidogrel BMJ 2007;334:119, doi:10.1136/bmj.334.7585.119-b

Feature What have we learnt from Vioxx? Harlan M Krumholz, Joseph S Ross, Amos H Presler, David S Egilman BMJ 2007;334:120-123, doi:10.1136/bmj.39024.487720.68

Observations THE WEEK IN MEDICINE: No quick fix for the NHS Rebecca Coombes BMJ 2007;334:124-125, doi:10.1136/bmj.39097.690428.59

WHAT'S ON BMJ.COM: The Trouble with Medical Journals by Richard Smith: an alternative view Pritpal S Tamber BMJ 2007;334:125, doi:10.1136/bmj.39098.446065.59

Yankee Doodling: Today's doctor's dilemma Douglas Kamerow BMJ 2007;334:126, doi:10.1136/bmj.39097.516609.59

Analysis Graphical method for depicting randomised trials of complex interventions Rafael Perera, Carl Heneghan, Patricia Yudkin BMJ 2007;334:127-129, doi:10.1136/bmj.39045.396817.68 The in-between world of knowledge brokering Jonathan Lomas BMJ 2007;334:129-132, doi:10.1136/bmj.39038.593380.AE

Research Impact of a theoretically based sex education programme (SHARE) delivered by teachers on NHS registered conceptions and terminations: final results of cluster randomised trial M Henderson, D Wight, G M Raab, C Abraham, A Parkes, S Scott, G Hart BMJ 2007;334:133, doi:10.1136/bmj.39014.503692.55 (published 21 November 2006)

Effect of isoniazid prophylaxis on mortality and incidence of tuberculosis in children with HIV: randomised controlled trial Heather J Zar, Mark F Cotton, Stanzi Strauss, Janine Karpakis, Gregory Hussey, H Simon Schaaf, Helena Rabie, Carl J Lombard BMJ 2007;334:136, doi:10.1136/bmj.39000.486400.55 (published 3 November 2006)

Effects of fortified milk on morbidity in young children in north India: community based, randomised, double masked placebo controlled trial Sunil Sazawal, Usha Dhingra, Pratibha Dhingra, Girish Hiremath, Jitendra Kumar, Archana Sarkar, Venugopal P Menon, Robert E Black BMJ 2007;334:140, doi:10.1136/bmj.39035.482396.55 (published 28 November 2006)

Clinical review Syphilis Patrick French BMJ 2007;334:143-147, doi:10.1136/bmj.39085.518148.BE

Practice Redesign and modernisation of an NHS cataract service (Fife 1997-2004): multifaceted approach Adrian Tey, Barbara Grant, Dawn Harbison, Shona Sutherland, Patrick Kearns, Roshini Sanders BMJ 2007;334:148-152, doi:10.1136/bmj.39050.520069.BE

10-minute consultation: Collapse with loss of awareness Dougall McCorry, Angela McCorry BMJ 2007;334:153, doi:10.1136/bmj.39070.390961.DE

Change page: Don't use minocycline as first line oral antibiotic in acne Paul McManus, Ike Iheanacho BMJ 2007;334:154, doi:10.1136/bmj.39048.540394.BE

Views & reviews Personal views: Who is responsible for do not resuscitate status in patients with broken hips? Rahij Anwar, Azeem Ahmed BMJ 2007;334:155, doi:10.1136/bmj.39085.659248.59

Netlines Harry Brown BMJ 2007;334:156, doi:10.1136/bmj.39094.711238.94

Personal views: Referral management schemes are damaging patients' interests Peter Lapsley BMJ 2007;334:156, doi:10.1136/bmj.39094.602569.94

Shopping for patients on the high street Margaret McCartney BMJ 2007;334:157, doi:10.1136/bmj.39091.442720.59

FROM THE FRONTLINE: Law and disorders Des Spence BMJ 2007;334:158, doi:10.1136/bmj.39097.614977.94

THE BEST MEDICINE: What goes around, comes around Liam Farrell BMJ 2007;334:158, doi:10.1136/bmj.39094.508218.59

BETWEEN THE LINES: A historical whopper Theodore Dalrymple BMJ 2007;334:159, doi:10.1136/bmj.39066.561551.B7

Medical Classics: Doctor in the House James Owen Drife BMJ 2007;334:159, doi:10.1136/bmj.39091.730000.59

Obituaries This week's obituaries

William Ian McDonald Caroline Richmond BMJ 2007;334:160, doi:10.1136/bmj.39097.535093.FA

Moyna Gladys Clark Nicola Pritchard BMJ 2007;334:161, doi:10.1136/bmj.39094.614942.FA

Alexander ("Sandy") Gordon Elder Ewan B Macdonald, Ian S Symington, Richard Soutar BMJ 2007;334:161, doi:10.1136/bmj.39094.548056.FA

Gwilym Penrose Hosking Emma-Jane Hosking, Suzie Mitchell BMJ 2007;334:161, doi:10.1136/bmj.39094.641956.FA

Aloysius ("Lou") Michels Kenneth Lamb, Jonathan Falla, Josephine McGettigan BMJ 2007;334:161, doi:10.1136/bmj.39087.743310.FA

Imrich ("Emery") Sarkany Robin Graham-Brown BMJ 2007;334:161, doi:10.1136/bmj.39094.526991.FA

Peter Thomas John Christopher Plumbly Warner Iona Warner BMJ 2007;334:161, doi:10.1136/bmj.39094.564468.FA

Minerva Minerva BMJ 2007;334:162, doi:10.1136/bmj.39093.672697.801

Minerva Jason Y K Chan, Richard Oakley, Michael J Gleeson BMJ 2007;334:162, doi:10.1136/bmj.39093.672697.80

Fillers Endpiece: A historical view of the future BMJ 2007;334:132, doi:10.1136/bmj.39043.680347.F7

Endpiece: What is research? BMJ 2007;334:147, doi:10.1136/bmj.39043.685775.F7 bmjupdates+: Acupuncture may speed up the active phase of labour in women whose membranes rupture before the onset of labour BMJ 2007;334:152, doi:10.1136/bmj.39094.662708.DE

Corrections Minerva BMJ 2007;334, doi:10.1136/bmj.39097.580961.BE

Comparison of treatment effects between animal experiments and clinical trials: systematic review BMJ 2007;334, doi:10.1136/bmj.39097.585880.BE

Screening strategies for chronic kidney disease in the general population: follow-up of cross sectional health survey BMJ 2007;334, doi:10.1136/bmj.39098.405382.BE

For the full versions of these articles see bmj.com EDITORIALS

Sexual health in adolescents “Saved sex” and parental involvement are key to improving outcomes

w5 RESEARCH p133 Despite increasing provision of school sex education, children. Much evidence points to the effectiveness of teenage sexual health in the United Kingdom is in over- maternal involvement in sex education.6 However, the Trevor Stammers lecturer in healthcare ethics all decline, with increasing rates of terminations and presence of a father is also an important factor in teen- w6 St Mary’s University College, sexually transmitted infections in under 18s outweigh- agers saving sex until adulthood. In the Netherlands, Twickenham, London TW1 4SX ing recent modest reductions in conception rates in this many more mothers and fathers talk with their children [email protected] age group.1 2 w1 about sex than in the UK.7 The lower rate of single par- Competing interests: TS is a trustee of Family Education Counterintuitively, rather than improving sexual enthood in the Netherlands is an important factor in the Trust and Challenge Teams UK; health, sex behaviour interventions can make it worse.w2 lower rate of teenage pregnancy seen in that country.7 both charities provide abstinence Most studies on sex education programmes in schools Giving students homework to complete with their par- centred sex education packages to secondary schools in the UK. He is examine intermediate outcomes only, such as pupil ents greatly enhanced the effectiveness of one school also a (volunteer) web doctor for satisfaction or reported condom use. This often facili- based abstinence only programme.8 This study, how- Love for Life, the largest provider of tates premature false claims of success, whereas more ever, measured the usual intermediate outcomes only, abstinence centred sex education to schools in Northern Ireland, and robust outcome measures such as rates of terminations, such as self efficacy and intention to have sex, and more did paid consultation work for them unplanned conceptions, and sexually transmitted infec- randomised controlled trials of abstinence programmes in the past. tions show no benefit. are needed to look at effects on sexual debut, unplanned In this week’s BMJ Henderson and colleagues report conceptions, and sexually transmitted infections. BMJ 2007;334:103-4 doi: 10.1136/bmj.39087.374653.BE the effect of a theoretically based teacher delivered sex Much teenage sex has little to do with sex itself but education programme (SHARE) on registered concep- is connected with searching for meaning, identity, and tions and terminations.3 The trial found no significant belonging.w7 These wider sociocultural aspects influ- difference between the intervention and control groups encing sexual behaviour need greater attention.9 Their in National Health Service registered conceptions (per importance is highlighted by a Seattle community 1000 pupils: 300 SHARE v 274 control; difference 26, based programme for children up to age 12, designed to 95% confidence interval !33 to 86) or terminations (127 help avoid health risk behaviours in later adolescence. SHARE v 112 control; difference 15, !13 to 42) between Though good social decision making (including abstain- ages 16 and 20. ing) in areas such as conflict resolution or drug use was The results should stimulate urgently needed change encouraged, this programme contained no specific sex from previous ineffective approaches to school sex edu- education. Nevertheless, at nine year follow-up, peo- cation for which, as the authors admit, “evidence of ple in the programme were significantly older at first effectiveness is mixed.” Evidence shows that increased sexual experience and were significantly less likely to knowledge is a necessary but insufficient cause of change become pregnant by age 21 (38% v 56%; P <0.05, 95% in sexual behaviour.w3 It is also clear that strategies such confidence interval 0.27 to 0.93).10 A key factor in the as promoting availability and correct use of condoms and Ugandan success in reducing HIV rates so dramatically increasing use of the emergency pill do not necessarily during the 1990s was a community wide, mass media lead to a reduction in sexually transmitted disease rates,1 communication of messages to achieve the desired out- pregnancies,w4 or terminations.4 comes of abstinence and being faithful, in addition to So what needs to happen next? Henderson and condom use.w8 w9 colleagues stress the need for more comprehensive Blanket assertions that abstinence programmes approaches that incorporate the influence of parents on “don’t work” abound.w10 Ironically, however, the only sexual experience in teenagers, and to improve the future randomised trial of school sex education identified by life opportunities for vulnerable young people. The false the SHARE team to use clinical data on pregnancies assumption that “young teens will have sex anyway” is was of an abstinence only programme that resulted in a an insult to many young people who have the capacity significant reduction of pregnancies.11 With the introduc- to rise to a far more effective challenge than just “use a tion of its recent “Leave it till later” campaign, the UK condom every time.” government is belatedly recognising that saved sex is a Delaying first intercourse is recognised as an impor- vital component of truly safer sex. In the United States, tant outcome measure of sex education programmes, and teenage pregnancies are at their lowest level in more than teenagers are more likely to save sex (a more positive 30 years, and the teenage abortion rate has been halved and helpful term than “abstinence”) when their parents since its peak in 1988. The only peer reviewed studies of communicate the importance of doing so.5 Of course, reasons for these trends published up to 2006 attribute to be involved in this way, parents have to be with their two thirds of the teen pregnancy reduction in unmarried

BMJ | 20 JANUARY 2007 | VOLUME 334 103 EDITORIALS

teenage girlsw11 and just over half of the total reduction to final results of cluster randomised trial. BMJ 2007;334:132-6. 12 doi: 10.1136/bmj.39014.503692.55. delayed first sex. With a recent study showing that edu- 4. Glasier A. Emergency contraception. BMJ 2006;333:560-1. cation to save sex does not lead to decreased condom 5. Dilorio C, Kelley M, Hochenberry-Eaton M. Communication about sexual w12 issues: mothers, fathers and friends. J Adolesc Health 1999;24:181-9. use at first sex, it is time the evaluation of saved sex 6. McNeely C, Shew ML, Beuhring T, Sieving R, Miller BC, Blum RW. programmes had a share of UK funding for sex educa- Mother’s influence on the timing of first sex among 14 and 15 year-olds. J Adolesc Health 2002;31:256-65. tion research. However, if saving sex is seen as a way for 7. Social Exclusion Unit. Teenage pregnancy. June 1999. www. young people to avoid unplanned pregnancy more than socialexclusion.gov.uk/downloaddoc.asp?id=69. 8. Blake S Simkin S, Ledsky, Perkins C, Calabrese JM. Effects of parent- just an educational structure, then last week’s govern- child communications intervention on young adolescents’ risk for early ment announcement of plans to raise the school leaving sexual intercourse. Fam Plan Persp 2001;33:52-61. age to 18 should do more to reduce teenage pregnancy 9. Joseph Roundtree Foundation. “Planned” teenage pregnancy. 2006 www.jrf.org.uk/bookshop/eBooks/9781861348753.pdf. w13 rates than anything they have done so far. 10. Lonczak H, Abbott R Hawkins D, Kosterman R, Catalano RF. Effects of the Seattle social development project on sexual behaviour, pregnancy, 1. Paton D. Random behaviour or rational choice? Family planning, birth and sexually transmitted disease outcomes by age 21yrs. Arch teenage pregnancy and sexually transmitted infections. Sex Education Pediatr Adolesc Med 2002;156:438-47. 2006;6:281-308. 11. Cabezon C, Vigil P, Rojas I, Leiva ME, Riquelme R, Aranda W, et al 2. Wilkinson P, French R, Kane R, Lachowycz K, Stephenson J, Grundy C, et al Adolescent pregnancy prevention: an abstinence-centred randomized Teenage conceptions, abortions, and births in England, 1994–2003, controlled intervention in a Chilean public high school. J Adolesc Health and the national teenage pregnancy strategy. Lancet 2006;368:1879- 2005;36:64-9. 86. 12. Santelli JS, Abma J, Ventura, Lindberg L, Morrow B, Anderson JE, et al. 3. Henderson M, Wight D, Raab GM, Abraham C, Parkes A, Scott S, et al. Can changes in sexual behaviors among high school students explain Impact of a theoretically based sex education programme (SHARE) the decline in teen pregnancy rates in the 1990s? J Adolesc Health delivered by teachers on NHS registered conceptions and terminations: 2004;35:80-90.

Childhood iron and zinc deficiency in resource poor countries Fortification is beneficial, but the best strategy for delivery is unclear

RESEARCH p140 In this week’s BMJ Sazawal and colleagues report a trial findings from a recent large scale trial of iron and zinc sup- of milk fortified with multiple micronutrients (as a strat- plementation in a malaria endemic area in Zanzibar are Zulfiqar A Bhutta 1 11 Husein Lalji Dewraj professor and egy to deliver zinc and iron) in children in India. They more worrying. In this trial, which compared daily iron chairman found a significant reduction in severe illness and the and folic acid with iron and folic acid and additional zinc Department of Paediatrics and Child incidence of acute respiratory infections and diarrhoea. or placebo in infants, the iron and folic acid supplemen- Health, Aga Khan University, Karachi 74800, Pakistan Although data on the impact of the intervention on iron tation arms were stopped early because of a significantly [email protected] and zinc status are not presented, the functional benefits higher rate of mortality and hospital admission. Competing interests: None declared are consistent with the previously recognised benefits These data have raised concerns about the safety of

BMJ 2007;334:104-5 of zinc supplementation on the burden and severity of large scale iron supplementation programmes in malaria 2 doi: 10.1136/bmj.39094.513924.BE diarrhoeal diseases and respiratory infections. endemic areas, and highlighted the need for alternative Iron deficiency ranked ninth among 26 risk factors strategies to deal with deficiencies in iron and other micro- included in the global burden of disease study, and nutrients in children at risk. These include fortification accounted for 841 000 deaths and 35 057 000 disability of commonly consumed foods, condiments, and staples adjusted life years lost.3 Large sections of populations in such as wheat flour. These may be useful for adults, but Africa and Asia are at risk of dietary zinc deficiency and because of the limited consumption of some of these foods resulting high rates of stunting.4 Correcting micronutri- by young infants and children, they may not be the most ent deficiencies can help reduce child mortality,5 but effective means of delivering recommended amounts of it is unclear how these deficiencies can be dealt with micronutrients. at the population level. Although combinations of iron An alternative strategy might be to fortify commonly and zinc have been suggested as appropriate strategies consumed centrally processed complementary or wean- for rectifying deficiencies of multiple micronutrients in ing foods. However, the availability and cost of such children at risk, the benefits may not be additive because foods in developing countries precludes their wide- of potential interactions.6 spread use. Alternative strategies include the addition Despite promotion of micronutrient supplements,7 of microencapsulated forms of iron and other micro- evidence from many supplementation trials in children nutrients in the form of “sprinkles” (sachets of fortified is mixed. Some trials using dispersible tablets contain- powder) to commonly used foods available at home.12 ing multiple micronutrients have found no significant Although sprinkles are generally added to foods, much benefits on functional outcomes.8 One study in children effort may be needed to induce behaviour change and in Peru with persistent diarrhoea found higher rates of to monitor changes. diarrhoea, respiratory infections, and febrile episodes in Fortification of commonly used food items for the children who received multiple micronutrients and children, such as milk, offers an attractive alternative. zinc compared with those given zinc supplementation Sazawal and colleagues’1 finding of the acceptability alone.9 However, in Karachi daily micronutrient and of fortified milk as a delivery vehicle are supported by zinc supplementation reduced rates of diarrhoea.10 The similar findings from Mexico.13 In the Mexican study

104 BMJ | 20 JANUARY 2007 | VOLUME 334 EDITORIALS

milk fortified with ferrous gluconate and zinc oxide children in developing countries: pooled analysis of randomized controlled trials. Zinc Investigators’ Collaborative Group. J Pediatr reduced anaemia and iron deficiency. 1999;135:689-97. Notwithstanding the benefits, milk fortification is 3 Stoltzfus RJ. Iron deficiency: global prevalence and consequences. expensive, and central processing and commodity Food Nutr Bull 2003;24(suppl 4):S99-103. 4 International Zinc Nutrition Consultative Group (IZiNCG). Assessment costs are high. The usefulness, cost effectiveness, and of the risk of zinc deficiency in populations. Food Nutr Bull 2004;25: sustainability of this approach must be evaluated at S130-62. 5 Jones G, Steketee RW, Black RE, Bhutta ZA, Morris SS; Bellagio Child scale in representative populations. A few examples are Survival Study Group. How many child deaths can we prevent this available of efforts to improve complementary feeding year? Lancet 2003;362:65-71. 6 Fischer Walker C, Kordas K, Stoltzfus RJ, Black RE. Interactive practices as a way of increasing micronutrient intake. effects of iron and zinc on biochemical and functional outcomes in The remarkable benefits on growth seen in Peru after supplementation trials. Am J Clin Nutr 2005;82:5-12. a nutrition and health education intervention delivered 7 Shrimpton R, Shrimpton R, Schultink W. Can supplements help meet the micronutrient needs of the developing world? Proc Nutr Soc by the health system suggest that this may be feasible 2002;61:223-9. in communities where food availability and choice are 8 Smuts CM, Lombard CJ, Benade AJ, Dhansay MA, Berger J, Hop le T, 14 et al; International Research on Infant Supplementation (IRIS) Study not a problem. However, this approach may not work Group. Efficacy of a foodlet-based multiple micronutrient supplement in places like India, where poverty and religious beliefs for preventing growth faltering, anemia, and micronutrient deficiency preclude adequate intake of micronutrient containing of infants: the four country IRIS trial pooled data analysis. J Nutr 2005;135:631S-8S. foods such as meat and poultry products. If the avail- 9 Penny ME, Marin RM, Duran A, Peerson JM, Lanata CF, Lonnerdal B, et ability of food can be ensured through appropriate al. Randomized controlled trial of the effect of daily supplementation with zinc or multiple micronutrients on the morbidity, growth, and financial support and social insurance schemes, these micronutrient status of young Peruvian children. Am J Clin Nutr interventions are the most logical way to provide iron 2004;79:457-65. and zinc to young children. 10 Sharieff W, Bhutta Z, Schauer C, Tomlinson G, Zlotkin S. Micronutrients (including zinc) reduce diarrhoea in children: the Pakistan sprinkles Availability of fortified milk poses a risk to programmes diarrhoea study. Arch Dis Child 2006;91:573-9. for the support of exclusive breastfeeding in such coun- 11 Sazawal S, Black RE, Ramsan M, Chwaya HM, Stoltzfus RJ, Dutta A, et al. Effects of routine prophylactic supplementation with iron and folic tries, so their use and promotion must be strictly targeted acid on admission to hospital and mortality in preschool children in and monitored. There is therefore a pressing need to a high malaria transmission setting: community-based, randomised, placebo-controlled trial. Lancet 2006;367:133-43. evaluate such interventions in large scale community 12 Zlotkin S, Antwi KY, Schauer C, Yeung G. Use of microencapsulated iron studies before they can be recommended. (II) fumarate sprinkles to prevent recurrence of anaemia in infants and young children at high risk. Bull World Health Organ 2003;81:108-15. 1 Sazawal S, Dhingra U, Hiremath G, Kumar J, Dhingra P, Sarkar A, et 13 Villalpando S, Shamah T, Rivera JA, Lara Y, Monterrubio E. Fortifying al. Effects of fortified milk on morbidity in young children in north milk with ferrous gluconate and zinc oxide in a public nutrition India: community based, randomised, double masked placebo program reduced the prevalence of anemia in toddlers. J Nutr controlled trial. BMJ 2007; 334: 140-2. doi: 140-2./10.1136/ 2006;136:2633-7. bmj.39035.482396.55. 14 Penny ME, Creed-Kanashiro HM, Robert RC, Narro MR, Caulfield LE, 2 Bhutta ZA, Black RE, Brown KH, Gardner JM, Gore S, Hidayat A, et al. Black RE. Effectiveness of an educational intervention delivered Prevention of diarrhea and pneumonia by zinc supplementation in through the health services to improve nutrition in young children: a

Tuberculosis in resource poor countries Better access to antiretroviral therapy and isoniazid prophylaxis offer new opportunities for control

RESEARCH p136 In most countries with limited resources the epidemics cal models have suggested it has limited potential to of HIV and tuberculosis continue to grow.1 Even with reduce the burden of tuberculosis within the general Robert Colebunders 3 head optimal treatment of active tuberculosis, the absolute population. However, these models did not take into HIV/STD Unit, Clinical Sciences number of tuberculosis cases will continue to rise if the account the potential effect of antiretroviral therapy on Department, Institute of HIV epidemic is not controlled.1 In this week’s BMJ, Zar the health seeking behaviour of populations with a high Tropical Medicine 2000 Antwerp, Belgium and colleagues report a randomised controlled trial per- seroprevalence of HIV. [email protected] formed in South Africa on the effect of isoniazid prophy- Better access to antiretroviral therapy makes people Ludwig Apers laxis on mortality and the incidence of tuberculosis in more willing to be tested for HIV.6 This has resulted clinician 2 Epidemiology Unit, children infected with HIV. The results of this study in a greater awareness of HIV and a reduction in the 7 Public Health Department, suggest that isoniazid prophylaxis may be an effective stigma associated with the disease. Moreover, voluntary Institute of Tropical Medicine public health intervention to reduce mortality in HIV counselling and testing sites are suitable places to screen Greet Dieltiens epidemiologist infected children in settings with a high prevalence of for tuberculosis and initiate tuberculosis chemoprophy- 2 8 William Worodria tuberculosis. laxis. clinician Today, antiretroviral therapy programmes also offer HIV seropositive adults with no clinical evidence of HIV/STD Unit, Clinical Sciences new opportunities to control tuberculosis.3-5 Highly tuberculosis benefit from isoniazid prophylaxis.9 This Department, Infectious Diseases Institute, active antiretroviral therapy was shown to decrease protection is less effective in tuberculin skin negative Kampala, Uganda the incidence of tuberculosis in HIV positive people patients, probably not because they are not latently Competing interests: None declared by 70% in South Africa4 and 80% in Brazil.5 Although infected with tuberculosis, but because they have such

BMJ 2007;334:105-6 antiretroviral therapy is likely to reduce the incidence serious cellular immunodeficiency that even chemo- doi: 10.1136/bmj.39092.388796.80 of tuberculosis in people infected with HIV, mathemati- prophylaxis cannot protect them. The study by Zar and

BMJ | 20 JANUARY 2007 | VOLUME 334 105 EDITORIALS

colleagues shows that isoniazid prophylaxis in children and IMAI training modules.13 These are currently (most of whom were not taking antiretroviral therapy) implemented in a growing number of settings, such as reduced mortality compared with placebo (median Uganda and Senegal, which will report back on their follow-up 5.7 months: 11 (8.3%) v 21 (16%); hazard ratio effectiveness in operational circumstances. 0.46, 95% confidence interval 0.22 to 0.95) and reduced Antiretroviral therapy programmes could also favour the incidence of tuberculosis. Antiretroviral therapy reac- the transmission of tuberculosis. Firstly, nosocomial tivates the immune system and therefore may increase transmission of tuberculosis could occur, especially in the protective effect of tuberculosis chemoprophylaxis. large scale and centralised HIV treatment centres that A study in Brazil showed that a combination of highly lack proper preventative measures. Secondly, the total active antiretroviral therapy and isoniazid prophylaxis number of people living with HIV and at increased risk reduced the incidence of tuberculosis to 0.6 per 100 for developing tuberculosis in the region will increase. person years of follow-up.10 Finally, if patients do not adhere to their treatment, if Effective case finding and treatment are crucial to inappropriate retroviral regimens are used, or if a reli- controlling tuberculosis. However, in most instances able supply of antiretroviral therapy is not maintained the diagnosis of tuberculosis in people with HIV is some patients may remain severely immunodeficient delayed, which may foster its transmission in the com- and susceptible to developing tuberculosis and drug munity. Many factors contribute to this delay such as resistance. insufficient awareness about tuberculosis, stigma asso- The increased access to antiretroviral therapy in ciated with tuberculosis and HIV, insufficient access resource poor settings offers important opportunities for to good health care, lack of confidence in healthcare controlling tuberculosis. Access to antiretroviral therapy services, and difficulties in diagnosing tuberculosis, seems to be changing the health seeking behaviour of the particularly in people with HIV. Increased access to population. This could lead to earlier diagnosis and treat- antiretroviral therapy could positively influence some ment of HIV and tuberculosis. Collaboration between of these factors. Indeed, in many countries where free those implementing HIV and tuberculosis control pro- antiretroviral therapy is available, treatment centres are grammes, needed to scale up the implementation of overwhelmed by patients. Moreover in countries with antiretroviral therapy, will hopefully lead to the strength- a high seroprevalence of HIV and increasing access to ening of the healthcare infrastructure in general. antiretroviral therapy, community efforts to fight both HIV and tuberculosis are growing. 1 Currie CS, Williams BG, Cheng RC, Dye C. Tuberculosis epidemics driven by HIV: is prevention better than cure? AIDS 2003;17:2501-8. Before initiating antiretroviral therapy patients should 2 Zar HJ, Cotton MF, Strauss S, Karpakis J, Hussey G, Schaaf HS, et be evaluated for signs or symptoms of tuberculosis. If al. Effect of isoniazid prophylaxis on mortality and incidence of tuberculosis is diagnosed, treatment for the disease tuberculosis in children with HIV: randomised controlled trial. BMJ 2007; 334:136-9. doi: 10.1136/bmj.39000.486400.55. should probably be started before initiating antiret- 3 Currie CS, Floyd K, Williams BG, Dye C. Cost, affordability and cost- roviral therapy, as this would decrease the risk of the effectiveness of strategies to control tuberculosis in countries with high HIV prevalence. BMC Public Health 2005;5:130. tuberculosis immune reactivation inflammatory syn- 4 Badri M, Wilson D, Wood R. Effect of highly active antiretroviral therapy drome.11 Thus, an increase in the capacity to diagnose on incidence of tuberculosis in South Africa: a cohort study. Lancet tuberculosis is needed for the successful implementation 2002;359:2059-64. 5 Santoro-Lopes G, de Pinho AM, Harrison LH, Schechter M. Reduced of antiretroviral therapy. risk of TB among Brazilian patients with advanced HIV infection In countries with limited resources, antiretroviral ther- treated with HAART. Clin Infect Dis 2002;34:543-6. 6 Wanyenze R, Kamya M, Liechty CA, Ronald A, Guzman DJ, Wabwire- apy is usually not started until patients have advanced Mangen F, et al. HIV counseling and testing practices at an urban HIV disease. Consequently, many people with HIV hospital in Kampala, Uganda. AIDS Behav 2006;10:361-7. 7 Levy NC, Miksad RA, Fein OT. From treatment to prevention: the already have tuberculosis. For example, at the Infec- interplay between HIV/AIDS treatment availability and HIV/AIDS tious Diseases Institute in Kampala, Uganda, the mean prevention programming in Khayelitsha, South Africa. J Urban Health CD4 lymphocyte count of patients starting antiretroviral 2005;82:498-509. 6 8 Wambua N, Odhiambo J, Njoroge A, Chakaya JM, De Cock KM. therapy is 65!10 /l and 14% of these patients have a his- Implementing measures to reduce the burden of HIV/TB in dually tory of tuberculosis.12 With increasing access to antiret- infected patients in a Nairobi slum [abstract]. 14th international conference on HIV/AIDS and sexually transmitted infections in Africa roviral therapy, more asymptomatic patients with HIV (ICASA). Abuja, Nigeria 2005: ThPoC00416. may have access to such treatment. To facilitate early 9 WHO global tuberculosis programme/UNAIDS. Policy statement on antiretroviral therapy, CD4 lymphocyte counting should preventive therapy against tuberculosis in people living with HIV. 1998; Document WHO/TB/98.255; UNAIDS/98.34. www.eldis.org/ be available wherever HIV testing is performed. static/DOC7104.htm. Decentralising CD4 lymphocyte counting will lead 10 Golub JE, Saraceni V, Cavalcante S, Pacheco AG, King B, Moore RD, et al. Tuberculosis (TB) incidence by HAART and isoniazid prophylactic to earlier initiation of antiretroviral therapy. This will therapy (IPT) in HIV-infected patients in Rio de Janeiro, Brazil decrease the risk of patients developing tuberculosis [abstract]. XVI international AIDS conference, 13-16 August 2006: and potentially reduce the incidence of tuberculosis in MOPE0395. 11 Narita M, Ashkin D, Hollender ES, Pitchenik AE. Paradoxical worsening communities where HIV and tuberculosis are prevalent. of tuberculosis following antiretroviral therapy in patients with AIDS. However, this decentralisation will require extra staff, Am J Respir Crit Care Med 1998;158:157-61. 12 John L, Baalwa J, Kalimugogo P, Nabankema E, Castelnuovo B, including qualified nurses and counsellors, and even lay Muhindo G, Colebunders R, Kambugu A. Response to ‘Does immune counsellors if not enough health staff are available. reconstitution promote active tuberculosis in patients receiving highly active antiretroviral therapy?’ AIDS. 2005 Nov 18;19(17):2049-2050. As a way of achieving this objective the World Health 13 WHO. Briefing package. Integrated approaches to HIV care, ART and Organization developed the integrated management prevention: IMAI and IMCI tools. Geneva: WHO. www.who.int/3by5/ of adolescent and adult illnesses (IMAI) guidelines publications/documents/imai/en/.

106 BMJ | 20 JANUARY 2007 | VOLUME 334 EDITORIALS

Surgery for cataract Reorganisation of in-house services is an efficient way to improve quality and increase volume

PRACTICE p148 Cataract is the most common cause of visual impair- and increased training opportunities for junior ophthal- ment throughout the world.1 In the United Kingdom mologists.9 This model can be set up at relatively low Christopher M Wood clinical director and consultant the prevalence of visually significant cataract is 30% in cost, it increases throughput, and more importantly it ophthalmologist people over the age of 65.2 Modern cataract surgery is fully integrated with the local service.9 Long term Eye Infirmary, Sunderland SR2 9HP rapidly improves vision, can be performed as a day case results from a similar treatment centre show that the [email protected] procedure, and has a low rate of complications. The effect can be maintained, with rates of access to cataract Competing interests: None declared demand for cataract surgery in the UK exceeds its avail- surgery among the highest in the country.10 11 During the BMJ 2007;334:107 ability, and the best way to organise services to meet the past three years the number of people having cataract doi: 10.1136/bmj.39093.388900.80 demand is unclear. In this week’s BMJ a study by Tey surgery has stabilised in England and Wales and waiting and colleagues reports on how reorganisation of their times have shortened—median waiting times are around existing National Health Service ophthalmic service 70 days.6 Currently 95% of UK cataract procedures are increased the quality and volume of cataract surgery.3 still performed in (NHS) hospitals, but should more The demand for cataract operations has increased, cases be performed in treatment centres?7 Similar and the number of procedures performed annually in problems with cataract surgery waiting times occurred the UK increased by 50% between 1990 and 1997.4 in Canada, and were resolved by using high volume However, the rate of cataract surgery for older peo- cataract centres. Waiting times have reduced but there ple in the UK remained disproportionately low. In are concerns that this may be at the expense of funding 1997 it was still fewer than 2000 per 100 000 for over for other ophthalmic treatments, and ultimately there 65 year olds compared with a government target of will be excessive unused capacity.12 3200 per 100 000, and waiting times were longer than The study by Tey and colleagues suggests that treat- 200 days.5 The government responded to the deficit ment centres within hospitals provide value for money by increasing funding for cataract surgery.4 The main and can deal with the surgical backlogs efficiently and aims were to improve referral to secondary care, reduce effectively. However, the UK government is keen to the number of patient visits before and after surgery, extend patient choice and increase the number of inde- and encourage development of efficient, high volume, pendent treatment centres. Independent sector treat- cataract only operating lists. Cataract surgery would ment centres certainly have a role in improving access be performed not only in general ophthalmic operat- to all forms of surgery, but only if they can integrate into ing theatres, but also in cataract treatment centres that local services and take into account local needs. could be part of the normal local ophthalmic service or 1 Thylefors B, Negrel A-D, Pararajasearam R, Dadzie K. Global data on in the independent sector. By 2003 annual targets for blindness. Bull World Health Organ 1995;73:115-21. cataract operations were exceeded,6 and by 2005 the 2 Reidy A, Minassian DC, Vafidis G, Joseph J, Farrow S, Wu J, et al. maximum wait for surgery was less than three months Prevalence of serious eye disease and visual impairment in a north 6 7 London population: population-based, cross sectional study. BMJ for all age groups. 1998;316:1643-6. Despite these promising results the role of the inde- 3 Tey A, Grant B, Harbison D, Sutherland S, Kearns P, Sanders R. Redesign and modernisation of an NHS cataract service (Fife 1997-2004): pendent sector, specifically the contribution of independ- multifaceted approach. BMJ 2007;334:148-52. doi: 10.1136/ ent sector treatment centres in increasing capacity, has bmj.39050.520069.BE. 4 Department of Health. Action on cataracts: good practice been questioned. A report by the House of Commons guidance. London: DOH, 2000. www.dh.gov.uk/ Heath Committee concluded that they were poorly inte- assetRoot/04/01/45/14/04014514.pdf. grated into the NHS, and that the decision to go ahead 5 Department of Health. NHS drive to end long waits for eye operations. Press release. London: DOH, 2003. www.dh.gov.uk/ with phase one of the independent sector treatment cen- PublicationsAndStatistics/PressReleases/PressReleasesNotices/fs/ tre programme was “a leap into the dark.”8 en?CONTENT_ID=4046841&chk=j%2B1XKd. 6 Department of Health. Hospital activity and episode statistics. An independent sector treatment centre operating in London: DOH. www.dh.gov.uk/PublicationsAndStatistics/Statistics/ an area with no capacity problem is a prime example of HospitalEpisodeStatistics/fs/en. 7 Department of Health. Faster access and more choice for cataract poor integration. It may result in little or no reduction patients. Press release. London: DOH, 2005. www.dh.gov.uk/ in the waiting time for cataract surgery at the expense of PublicationsAndStatistics/PressReleases/PressReleasesNotices/fs/ financial destabilisation of local NHS hospitals.8 Also, en?CONTENT_ID=4102616&chk=fMQsuJ. 8 House of Commons Health Committee. Independent sector treatment there will be a loss of training opportunities for junior centres. Fourth report of session 2005-6. Vol. 1. www.publications. doctors in the long term, which may lead to a shortage parliament.uk/pa/cm200506/cmselect/cmhealth/934/934i.pdf. 9 9 Au L, Saha K, Fernando B, Ataullah S, Spencer F. “Fast track” cataract of appropriately trained surgeons. services and diagnostic and treatment centre: impact on surgical Separating elective and emergency care does have training. Eye 14 Jul 2004 [Epub ahead of print]. 10 Cresswell PA, Allen ED, Tomkinson J, Chapman FM, Pickering S, benefits. For hospital cataract services, much can be Donaldson LJ. Cost effectiveness of a single function treatment center for gained from having an in-house treatment centre, cataract surgery. J Cataract Refract Surg 1996;22:940-6. where only cataract operations are performed.3 10 The 11 Bailey K, Deane M, MacKnight N, Chappel D, Wilkinson J. Cataract surgery in the North East of England. North East Public Health study of Tey and colleagues describes such a model, Observatory. Paper 04, 2004. www.nepho.org.uk/index.php?c=634. which has improved the efficiency and quality of care 12 Bellan L. Cataract surgery in Canada. Can J Ophthalmol 2004;41:539-41.

BMJ | 20 JANUARY 2007 | VOLUME 334 107 EDITORIALS

Renaming schizophrenia Diagnosis and treatment are more important than semantics

Recent reports in the media have called for schizophre- applied, the patient will receive the wrong treatment nia to be “abolished as a concept” because it is scien- and potentially have the stigma of having a mental Jeffrey A Lieberman 1 chairman and director tifically meaningless. This is not the first time that the illness. For example, if a patient with a toxic (such [email protected] validity of this diagnostic entity has been challenged, as phencyclidine induced) psychosis is misdiagnosed Michael B First and it will not be the last until the cause of the disorder with schizophrenia, he or she may be given a long and professor of clinical psychiatry Department of Psychiatry, Columbia and its precise pathophysiology are known. unnecessary course of antipsychotic drugs. To avoid University College of Physicians The current system of psychiatric diagnosis can- this situation, psychiatric diagnoses have built-in safe- and Surgeons, New York State not describe definitive disease entities because of our guards in the form of exclusion criteria that prevent Psychiatric Institute, New York, NY 10032, USA inability to demonstrate “natural” boundaries between a diagnosis from being made if certain conditions are Competing interests: JAL receives disorders. However, as Kendell and Jablensky point present (for example, a diagnosis of schizophrenia is research grant support from out, “thoughtful clinicians have long been aware that not permitted unless psychotic symptoms persist for a Acadia, Bristol-Myers Squibb, diagnostic categories are simply concepts, justified only substantial period of time after the person has stopped GlaxoSmithKline, Merck, Organon, and Pfizer. MBF has received by whether they provide a useful framework for organ- using the drug in question). consultant fees from AstraZeneca, ising and explaining the complexity of clinical experi- Concerns about potential stigma associated with Abbott Laboratories, and ence in order to derive inferences about outcome and having a serious mental illness have resulted in pro- GlaxoSmithKline and has been an 2 expert witness regarding diagnosis to guide decisions about treatment.” In this context, posals to change the name of schizophrenia. “Integra- of schizophrenia the charge that schizophrenia does not define a specific tion disorder” and “dopamine dysregulation disorder” illness is clearly unwarranted. Although the validity of have been suggested as possible alternatives.10 Unfor- BMJ 2007;334:108 doi=10.1136/bmj.39057.662373.80 the diagnosis remains to be established, its diagnostic tunately, changing the name of the condition (or even reliability and usefulness are indisputable. abolishing the concept) will not affect the root cause For more than 100 years schizophrenia has been of the stigma—the public’s ignorance and fear of peo- an integral part of our nosology and has facilitated ple with mental illness. Renaming may even have the research and treatment of people affected by this dis- unintended effect that the person, rather than the ill- ease.3 4 People qualify for the diagnosis if their clini- ness, is blamed for the symptoms.11 cal signs and symptoms conform to the operational Ultimately, we must gain a more complete under- diagnostic criteria that define schizophrenia. Many standing of the causes and pathophysiological mecha- studies have shown that these diagnostic criteria can nisms underlying schizophrenia. Only then can we be applied reliably and accurately by trained mental replace the way we characterise schizophrenia with health professionals.5 6 Although a diagnosis of schizo- a diagnosis that more closely conforms to a specific phrenia depends on the presence of a pattern of symp- brain disease. In the meantime, we can be confident toms (such as delusions, hallucinations, disorganised and grateful that the benefits conferred by the concept speech, disorganised or catatonic behaviour, and nega- of schizophrenia far outweigh any perceived disad- tive symptoms such as lack of motivation), evidence vantages. shows that these are manifestations of brain pathol- 7 ogy. Schizophrenia is not caused by disturbed psy- 1 Boseley S. Call to wipe out schizophrenia as catch-all tag. Guardian, chological development or bad parenting. Compared 10 October 2006. 2 Kendell R, Jablensky A. Distinguishing between the validity and utility with normal controls, people with schizophrenia have of psychiatric diagnoses. Am J Psychiatry 2003;160:4-12. abnormalities in brain structure and function seen on 3 Kraepelin E. Psychiatrie: Ein Lehrbuch für Studirende und Ärzte. 4th neuroimaging and electrophysiological tests. In addi- ed. (Psychiatry: a textbook for students and physicians. 4th ed.) Leipzig, Germany: Abel, 1893. tion, the evidence that vulnerability to schizophrenia 4 Bleuler E. Dementia praecox, oder die Gruppe der Schizophrenien. is at least partly genetic is indisputable.8 (Dementia praecox, or the group of the schizophrenias.) Leipzig, Germany: Franz Deuticke, 1911. Once a diagnosis of schizophrenia is made, the treat- 5 Jakobsen KD, Frederiksen JN, Hansen T, Jansson LB, Parnas J, Werge ing clinician has a wide array of treatment options T. Reliability of clinical ICD-10 schizophrenia diagnoses. Nord J available, which have been tested empirically on simi- Psychiatry 2005;59:209-12. 6 McCormick LM, Flaum M. Diagnosing schizophrenia circa 2005: how lar groups of people. Furthermore, the doctor will also and why? Curr Psychiatry Rep 2005;7:311-5. have access to the huge body of empirical data that 7 Andreasen N. The broken brain: the biological revolution in psychiatry. New York: Harper and Row, 1984. characterises this condition including its course, treat- 8 Lewis DA, Lieberman JA. Catching up on schizophrenia: natural history ment response, outcome, and family history. This is and neurobiology. Neuron 2000;28:325-34. important because evidence shows that early interven- 9 Perkins DO, Gu H, Boteva K, Lieberman JA. Relationship between duration of untreated psychosis and outcome in first-episode 9 tion may improve outcome. The diagnosis also helps schizophrenia: a critical review and meta-analysis. Am J Psychiatry when explaining to the patient and their family the 2005;162:1785-804. 10 Sugiura T, Sakamoto S, Tanaka E, Tomoda A, Kitamura T. Labelling nature of the problem, the range of treatments and effect of Selshin-bunretsu-byou, the Japanese translation for outcomes, and the assistance available from support schizophrenia: an argument for re-labelling. Int J Soc Psychiatry groups. 2001;47:43-51. 11 Penn DL, Nowlin-Drummond A. Politically correct labels and Of course, diagnostic labels have potential disad- schizophrenia. A rose by any other name? Schizophrenia Bull vantages. If a diagnosis of schizophrenia is mistakenly 2001;27:197-203.

108 BMJ | 20 JANUARY 2007 | VOLUME 334 We select the letters for these pages from the rapid responses posted on bmj.com favouring those received within five days of publication of the article to which they LETTERS refer. Letters are thus an early selection of rapid responses on a particular topic. Readers should consult the website for the full list of responses and any authors’ replies, which usually arrive after our selection.

who did their best to perform a quasi- 1 Electronic responses. Impact of case management (Evercare) on frail elderly patients. BMJ 2007; www.bmj. experimental evaluation given the com/cgi/eletters/334/7583/31 constraints of the project handed to them. 2 Gravelle H, Dusheiko M, Sheaff S, Sargent P, Boaden R, Pickard S, et al. Impact of case management David Oliver senior lecturer, Institute of Health Sciences, (Evercare) on frail elderly patients: controlled before University of Reading, Reading RG1 5AQ and after analysis of quantitative outcome data. BMJ [email protected] 2007;334:31-4. (6 January.) Competing interests: None declared. 1 Gravelle H, Dusheiko M, Sheaff S, Sargent P, Boaden R, Pickard S, et al. Impact of case management NATIONAL SICKNESS SERVICE (Evercare) on frail elderly patients: controlled before and after analysis of quantitative outcome data. BMJ 2007;334:31-4. (6 January.) Fundamental problem is not CASE MANAGEMENT 2 Kane RL, Keckhafer G, Flood S, Bershadsky B, Siadatty MS. The effect of Evercare on hospital use. J Am Geriatr lack of preventive care Soc 2003;51:1427-54. Government should have 3 Singh D. Transforming chronic care. Evidence about I do not recall from my training in the people with long term conditions. University of United Kingdom meeting as many respected evidence Birmingham Health Services Management Centre. 2005. www.hsmc.bham.ac.uk/news/transformingchroniccare patients who are utterly disenfranchised, Gravelle et al report the impact of case 4 Hutt R, Rosen R, McCauley J. Case-managing long term disempowered, and uninterested in their management (Evercare) on frail elderly conditions. What impact does it have in the treatment of health, as so many of my patients are older people? London: King’s Fund, 2004. 1 patients. 5 National Library for Health. Primary Care Information here in the US. I work in a clinic that sees The Evercare pilot study was based on Service. Answer to question: “What is the best and most almost entirely uninsured or underinsured recent evidence for case management in primary care the findings of one quasi-experimental as a means of reducing admission rates to hospital?” patients. It seems that years of being unable study of case management in US nursing 17 August 2005. www.clinicalanswers.nhs.uk/index. to afford care, of worrying about paying homes, albeit one with impressive results. cfm?question=976 rent, or buying their children clothes, leaves It was imported to the United Kingdom people too exhausted to be concerned with in 10 pilot sites and applied to a very Community matrons do make even their current state of health, let alone different group of patients (community a difference their wellbeing in 10 or 20 years. I agree dwelling), in a very different health with Heath that the solution to healthcare economy, with different nursing skills and Respondents on bmj.com are right to disparity across classes is the eradication of information system. Even the author of the wonder whether the tail wagged the dog social and economic disparity.1 original paper expressed surprise that the when policy on case management emerged On the Venn diagram of preventive intervention had been implemented in this three years before the publication of the medicine and a National Sickness Service, way.2 This cost the UK taxpayer over £4m research from Gravelle et al.1 2 However, the where the two overlap sits primary care. (€6m; $7.8m)—not to mention the cost of emphasis on research is misleading. Local My experience in the US has convinced me “backfill” for community nursing posts implementation at the level of the primary that this is the most essential component vacated by the new advanced practitioner care trust was never intended to be a of a utilitarian health system because it has nurses. research project. Instead, it was a bold local cost effective, evidence based methods of If the Department of Health had any decision to invest to save, at a time when the controlling the course of common diseases regard for evidence it would not have Bristol health community was in financial in chronically sick patients. ignored two excellent systematic reviews3 4 crisis. Richard J Lyus resident in family medicine that showed no consistent evidence for Policy making is an illogical world. Swedish Providence Family Medicine, 550 16th Avenue, the effectiveness of case management Meanwhile, back in the real world, despite Seattle, WA 98122, USA [email protected] Competing interests: None declared. in preventing hospital attendances or the research that tells us that it shouldn’t admissions, reducing health costs, or work, the approach does what we want it to: 1 Heath I. In defence of a National Sickness Service. BMJ 2007;334:19. (6 January.) improving function. Nor would it have patients are alive and well and still living at repeatedly touted the “Castlefields” home and avoiding hospital; and the trust’s model5 (never published in a peer board is pleased with the local evidence SURGICAL FOLLOW!UP BY GPS reviewed journal) as good evidence showing that our community matrons for chronic disease management. Most more than cover their costs in emergency Proposal strikes at the heart importantly, it would have commissioned admissions saved (unpublished data triaged of medicine a good randomised controlled trial of by a nurse and medical consultant). the intervention in the UK instead of Martin J Howard service improvement manager To abandon follow-up appointments after commissioning an evaluation almost as an King Square House, Bristol BS2 8EE surgery strikes at the fundamentals of the [email protected] afterthought. philosophy of medicine.1 2 That surgeons Competing interests: The author is an NHS employee, and The only parties to emerge with credit has no financial or other connection with United HealthCare will be mere technicians whose only contact from this exercise are the research team Group or its subsidiaries. with the patients is, more or less, in the

BMJ | 20 JANUARY 2007 | VOLUME 334 109 LETTERS

operating theatre, is appalling. Follow-up is non-smoker, responsible eater lived to be as an adjunct to traditional methods. essential for many reasons, not just for audit 85. Since age 65 she received 20 years’ If more time and money is not spent on of the surgery or to deal with complications. retirement as a government worker, 20 years anatomy for students then their knowledge It is part of the doctor-patient relationship; social security payments, had Medicare will always be weak. A shift in the medical it allows assessment of things other than paid pharmaceutical charges for almost school curriculums needs to happen, and technical success; it gives an opportunity to two decades of diabetes, hypertension, and anatomical knowledge should be retested discuss questions the patient may have; and, raised cholesterol, outpatient office visits, at clinical levels, perhaps through end it provides some satisfaction from seeing a and hospitalisation costs for a pneumonia, of specialty examinations and a surgical happy patient with a good result. All of these fractured hip, and sub-endocardial anatomy component of finals. Until there constitute “clinically important” follow-up infarction. Her brother, a heavy smoker and is one standardised examination for all and to deny it suggests that surgeons do not drinker, literally dropped dead at 60. Up medical students in the United Kingdom, need to behave as doctors, but as robots. to that point he paid for medical expenses this will not be possible. If the UK is to And what do the patients think of all this? out of his own pocket. One sibling cost produce high quality surgeons of the future Have they been asked, and if not, why not? the taxpayers probably close to a million then the surgeons of today must be more How sad it will be if patients who wish dollars (far greater that she ever paid into actively involved in teaching, reinforcing, for, or need, any of the above have to be the system), the other nothing (in fact it was and planning anatomy teaching at both re-referred (and wait the 11 weeks for an a great financial gain since he had a high undergraduate and postgraduate level. appointment, which in all probability will paying job and paid taxes for years.) Sanjay Purkayastha specialist registrar, general surgery have been made through Choose & Book When I am asked, the advice I give to Academic Surgical Unit, St Mary’s Hospital, London W2 1NY and be with the wrong surgeon). Assuming smokers is this: “As a physician I can tell [email protected] that half all surgical patients do this, then you that if you smoke you will probably Paraskeva Paraskevas senior lecturer in surgery Ara Darzi professor of surgery there will be no cost saving at all. Surgeons die of heart disease or cancer. If you Department of Biosurgery and Surgical Technology, Imperial who do not see their postoperative results don’t smoke you will probably die of College, 10th Floor QEQM Building, London W2 1NY will not learn either from their successes or heart disease or cancer, but usually some Competing interests: None declared. failures, and the potential for another Bristol years later. During the extra time non- 1 Dobson R. Anatomy teaching in United Kingdom is in cardiothoracic surgery debacle will be smoking gives you, you may develop the crisis, new report says. BMJ 2007;334:12. (6 January.) enormously increased. infirmities, disabilities, illnesses, and aches Andrew N Bamji consultant rheumatologist and pains which go along with old age, CHILD SAFETY IN CARS Queen Mary’s Hospital, Sidcup, Kent DA14 6LT and then die. My overall recommendation [email protected] is not to smoke, however, because it Seats must be secured when Competing interests: ANB is a secondary care physician really does increase the risk of dying from with a long term follow-up cohort of patients with chronic children are not using them diseases. emphysema which is a very unpleasant 1 Kmietowicz Z. GPs to follow up patients after surgery. way to die. As a taxpayer, however, I We are concerned that the European BMJ 2007;334:9-a. (6 January.) applaud your decision to smoke since you compliant law of 18 September 2006 2 Electronic responses. GPs to follow up patients after surgery. BMJ 2007; www.bmj.com/cgi/ will probably be much less of a financial regarding safety regulations for children eletters/334/7583/9-a burden to taxpayers because you will die over 31 may contribute towards increased sooner. You might even be a financial asset numbers of injuries and fatalities from to them if you time it right.” unsecured car safety seats acting as potential Robert A Da Prato physician projectiles when the child is not in the car. Portland, OR 97229, USA [email protected] Unsecured projectiles in cars can be lethal Competing interests: None declared. in the event of a crash, typically increasing 1 Glantz L. Should smokers be refused surgery? BMJ 2

M the effective weight of the item by 20 times.

O 2007;334:21. (6 January. C . This concern led to legislation in Holland, S O

T enforcing that all heavy objects in a vehicle O H

P ANATOMY CRISIS be secured. Unfortunately, this issue was not factored SMOKERS AND SURGERY Make surgeons more active in into the risk assessment when the legislation teaching anatomy at all levels was made. Currently, statistics of in-car The best time to die (to help injuries from unsecured projectiles in the fellow taxpayers) That anatomy teaching is in crisis has been UK are not collected. We suggest an urgent felt by clinicians for many years.1 Surgeons risk assessment to ensure that the new law Not too long ago an accountant told me should be more involved in the teaching of does not cause overall more death and that, from a financial perspective only, a anatomy and be integrated with anatomists injury than it is designed to prevent. patriotic citizen should die the moment he and other clinical specialists to provide a Adam C Frosh consultant ENT surgeon, Lister Hospital, or she changes from a net taxpayer (a net multispecialty approach. The practicalities Stevenage, Hertfordshire SG3 6ST, [email protected] tax asset of the state) to a net tax consumer of maintaining large numbers of cadavers Jonathan Wernick compliance consultant (a net tax liability of the state).1 For many have to be balanced with cost, utilisation, J W Compliance, Uppingham, Rutland LE15 9QJ people this means one should drop dead on and usefulness of the teaching methods Competing interests: None declared. the way home from one’s retirement party. adopted. New technology (such as web 1 Hayes M. Child safety in cars. BMJ 2006;333:1183-4. (9 December.) A brother and sister acquaintance illustrate based learning, 3-D anatomical packages, 2 Lethal weapons. 1 January 2007. www.openroad.com. this quite well. The sister, a non-drinker, and virtual prosections) should be embraced au/motoring_roadsafety_lethalweapons.asp

110 BMJ | 20 JANUARY 2007 | VOLUME 334 For the full versions of articles in this section see bmj.com UK NEWS Fertilisation authority raids controversial fertility clinics, p115 NEWS WORLD NEWS Democrats push for stem cell research, p112 bmj.com Hospital withdraws letter to GPs saying how patients can jump the queue Lack of beds leads to missed government targets

Susan Mayor LONDON ment did not reach this target. “Not enough Doctors from more than a third of accident available in-patient beds” was the main and emergency departments in hospitals reason given for departments not meeting in England said they were not reaching the the emergency access target, and 88% of government’s target of dealing with patients respondents reported it as a major contribu- within four hours. The main reason for this tory factor. finding, from a survey published this week, Don MacKechnie, chairman of the BMA’s was a shortage of available hospital beds, mak- emergency subcommittee and a consultant ing it difficult for patients to be admitted. in emergency medicine at Pennine Acute Only 49% of the nearly 500 members Hospitals NHS Trust, said that financial of the British Association for Emergency problems in the NHS had meant that the Medicine—all consultants and middle grade availability of hospital beds for patients need- doctors—responding to the survey reported ing admission from accident and emergency that their departments met the emergency departments was an important problem. He access target for 98% of all patients to be said, “In certain trusts, financial recovery seen, treated, admitted, or discharged within measures have included closing beds pre- four hours. viously used as ‘safety valves’ before, such

Although only 49% considered that their as beds in rehabilitation wards. This has a JOHN COLE/SPL department actually met the emergency knock-on effect to A&E [accident and emer- Patients are sometimes “moved” on the computer access target, 67% said that the figures gency departments], with recovering patients but not in reality reported by their department indicated that that might have previously been moved the four hour target can be met.” He said the target had been met. Respondents’ com- remaining in acute beds.” this meant ensuring that sufficient beds were ments on what had happened in their depart- Mr MacKechnie said it was important that available to admit patients from emergency ments included, “Middle managers revising emergency access targets were seen as hav- departments and ensuring that doctors on figures before being submitted,” “Minor ing broader application than simply affect- call in each specialty are free to see patients injury unit figures used to help achieve ing accident and emergency departments. in accident and emergency departments department’s 96.9% target,” and “Patients “Sometimes, other specialties forget that this within the target of one hour after being often ‘moved’ on the computer tracking but is not just an A&E target. It is an emergency referred. still kept in department to meet targets.” access target, with a responsibility from all Emergency Medicine: Report of a National Survey of Forty one per cent said that their depart- specialties in an acute trust to ensure that Emergency Medicine is at www.bma.org.uk.

BMJ readers choose sanitation as greatest medical advance since 1840

Annabel Ferriman BMJ pioneered the introduction of of the general public, and against health hazards is More than 11 300 readers of piped water to people’s homes one in seven were students. often the best way to improve the BMJ chose the introduction and sewers rinsed by water, Another tenth were academic population health. of clean water and sewage attracted 15.8% of the votes, researchers. Almost two fifths “The original champions of disposal—“the sanitary while antibiotics took 15%, of the voters were from the the sanitary revolution were revolution”—as the most and anaesthesia took 14%. The United Kingdom, and a fifth John Snow, who showed that important medical milestone next two most popular were the were from the United States. cholera was spread by water, since 1840, when the BMJ introduction of vaccines, with Johan Mackenbach, professor and Edwin Chadwick, who was first published. Readers 12%, and the discovery of the of public health at Erasmus MC came up with the idea of were given 10 days to vote on structure of DNA (9%). Medical Center, Rotterdam, sewage disposal and piping a shortlist of 15 milestones, A total of 11 341 people who championed the cause of water into homes. and sanitation topped the voted on the shortlist, which sanitation, said, “I’m delighted “Inadequate sanitation is poll, followed closely by the was chosen by a panel of that sanitation is recognised still a major problem in the discovery of antibiotics and the experts from a list nominated by so many people as such developing world.” development of anaesthesia. by readers. Almost a third an important milestone. The The Medical Milestones The work of the 19th century of the voters were doctors, general lesson which still supplement is distributed with lawyer Edwin Chadwick, who while a fifth were members holds is that passive protection this week’s BMJ.

BMJ | 20 JANUARY 2007 | VOLUME 334 111 NEWS

UK body wants consultation on human-animal hybrid research

Susan Mayor LONDON the authority to adopt a policy supportive “of and are needed for treating infertility. They The Human Fertilisation and Embryology embryo research which involves the mixing— pointed out that the House of Commons Authority (HFEA), the independent regula- for research purposes only—of human and ani- Science and Technology Committee recently tor of in vitro fertilisation (IVF) treatment and mal cells and DNA, subject to its own strict concluded, after a long inquiry, that such embryo research in the United Kingdom, has licensing requirements and to the usual 14 research was ethically acceptable and should called for a public debate before deciding day limit which applies to human embryos.” be regulated by the authority. whether to grant licences for research involv- In a letter to the Times on 10 January (www. Stephen Minger, director of the stem ing hybrids of human and animal cells. timesonline.co.uk/article/0,,59-2538977,00. cell biology laboratory at King’s College The authority recommended further public html) they argued that there were “clear London and one of the signatories to the let- discussion after it met last week to discuss the potential benefits to human health from, for ter to the Times, said, “We are disappointed broad principles for handling any research example, being able to grow stem cells with that the HFEA didn’t just recommend that the proposals involving animal-human hybrids specific genetic abnormalities, improving the applications go through the usual process for or chimeras (organisms or organs consisting efficiency of therapeutic cloning techniques embryo research. However, we are gratified of two or more tissues of different genetic and establishing cell lines for the testing of that the authority stood up and showed itself composition). It has already received two new treatments for diseases such as motor to be an independent regulator with backbone applications from groups wanting to carry out neurone disease, Alzheimer’s disease and and did not just follow the government’s white research using human cells and animal eggs to spinal muscular atrophy.” paper, which could have just banned this type produce stem cells. The group argued that these advances could of research now.” After discussion the authority decided that be achieved by using human-animal hybrids Dr Minger’s group is one of the two that these techniques were not prohibited by cur- without having to rely on the use of human have applied for a licence to use human- rent law and would potentially fall within its eggs, which they said are in very short supply animal hybrid techniques to study degen- remit to regulate and license. However, it erative neurological diseases, including delayed making a final decision until Parkinson’s disease, Alzheimer’s disease, after public consultation. and spinal muscular atrophy. Angela McNab, the authority’s Dr Minger said that animal eggs, such chief executive, said: “The issues as cow eggs, which his laboratory plans around hybrid and chimera research to use, are freely available as a byproduct are unique and different from main- of the food industry. Their plentiful stream human embryo research. After supply meant that high quality eggs could weighing up the scientific, legal, and ethi- be selected for research purposes. The cal issues presented, the authority decided cells created would be used only as cell that there needs to be a full and proper pub- lines to enable understanding of the lic debate and consultation as to whether, in pathology underlying the neurologi- principle, licences for these sorts of research Cow eggs are in plentiful cal diseases being investigated and to could be granted.” supply as a byproduct of develop new approaches to their treat- A group of more than 40 leading UK doc- the food industry ment and would never be implanted to

tors, scientists, ethicists, and politicians urged create embryos. LEDNER/STONE/GETTY CATHERINE

Democrats push for stem cell research despite opposition from Bush Janice Hopkins Tanne NEW YORK But the White House last week President Bush restricted The US House of said it opposed the Bill, and federal funding for embryonic Representatives, controlled President George Bush used stem cell research to 21 cell since the November elections the first veto of his presidency lines that were created before 9 by the Democrats, has passed to kill an identical bill last year August 2001. Many are thought a bill allowing embryonic stem (BMJ 2006;333:216). to be too contaminated for use cell research using unwanted, A presidential veto can be in humans (BMJ 2005;330:214). donated frozen embryos from overturned by a two thirds President Bush opposes fertility clinics. The Senate is majority vote in both the the destruction of unwanted expected to pass the bill in the House of Representatives and embryos donated from fertility next few weeks. the Senate. Last week’s vote in clinics to produce stem cells. Polls have shown that most the House fell short of the 290 Last year his spokesman Tony Americans favour stem cell votes needed for a two thirds Snow said, “The simple answer

RON EDMONDS/AP/EMPICS research, which may hold cures majority, being 253 to 174. In is he thinks murder’s wrong” President Bush’s spokesman Tony for Parkinson’s disease, spinal the Senate 66 or 67 senators (of (Register, www.theregister.co.uk, Snow (above) said the president cord injuries, Alzheimer’s 100 senators) are in favour of 11 Jan, “Stem cells face second thought murder was wrong disease, and other disorders. the bill. Bush veto”).

112 BMJ | 20 JANUARY 2007 | VOLUME 334 NEWS

Government suffers its first defeat over Mental Health Bill Clare Dyer BMJ The government has suffered what is expected to be the first of a series of defeats over its Mental Health Bill, which is intended to strengthen powers of compulsory treatment for psychiatric patients in England and Wales. An alliance of Liberal Democrats, Conservatives, and cross-bench peers plans to press a series of amendments to a vote that, if successful, will see the bill reach the House of Commons significantly changed from the form in which the government introduced it. Last week’s defeat for the government

COMSTOCKCOMPLETE.COM came when an amendment was passed, by a majority of 106, that would allow mentally ill people who are compulsorily detained Compulsive gamblers must get the same right to refuse treatment as other patients if their decision making capacity is unimpaired. Baroness Barker, a Liberal free NHS treatment, BMA says Democrat spokeswoman, said the right had been an omission from the current legislation, Lisa Hitchen LONDON of the report and professor of gambling studies the 1983 Mental Health Act. The NHS must be given the resources to at Trent University, Nottingham. The Mental Health Bill applies to anyone treat the increase in the number of problem If £10m proved not to be enough, 1% of with “any disorder or disability of the mind” gamblers that is likely to result from legal and the profits from gambling could be ploughed who needs to be detained for his or her technological changes in gambling activities, back into research, prevention, and treatment, own health or safety or the protection of says the British Medical Association. Professor Griffiths added. others. It will abolish the “treatability” test In a new report published this week the In particular, studies should look at inter- for detention under the 1983 Mental Health BMA says that gambling addiction must be net gambling, as well as the effect of the 24 Act, which requires that treatment given recognised as being as serious and complex a hour licensing laws on gambling problems, the under detention should be “likely to alleviate medical problem as other addictions and be BMA report recommends. or prevent a deterioration” in the patient’s able to be treated on the NHS. New laws coming into force this September condition. Instead the bill simply requires that Research from 1999 estimates that the are also likely to have an effect on gambling “appropriate medical treatment” be available United Kingdom has as many as 300 000 behaviour and must be monitored, it says. for the patient. This includes “nursing, problem gamblers, but this is likely to be an The 2005 Gambling Act allows the provision psychological intervention, and specialist underestimate of the problem, as it does not of a further 17 casinos to add to the 140 that mental health habilitation, rehabilitation take into account developments in the indus- already operate in the UK. One of these—a and care.” try since then, the report notes. “super-casino”—will have 1250 jackpot slot The bill has been described as “flawed” Doctors lack awareness of how to treat prob- machines. by the Mental Health Alliance, an umbrella lem gamblers, and research into which treat- People with low incomes, those with comor- body for 78 organisations, and is opposed by ments work is needed, it says. bidities such as drug or alcohol addiction or legal and medical groups, including the BMA, The BMA is calling on gaming operators mental health problems, children and adoles- the Law Society, and the Royal College of and anyone who makes money through gam- cents, and a growing number of women are all Psychiatrists. bling to pay into a £10m (€15m; $20m) a year at high risk of developing gambling addictions, Lord Carlile, the Liberal Democrat fund for research and treatment programmes; Professor Griffiths said. This is partly because peer who chaired the joint committee that this amount would be three times that currently gambling was partly destigmatised when the scrutinised the draft bill, said peers did not being provided through the Responsibility in National Lottery was introduced in 1994. But press for an amendment last week requiring Gambling Trust, an independent body funded interactive technologies, which allow gambling that some therapeutic benefit must be shown by the industry to raise awareness of prob- through mobile phones, televisions, and the for anyone who is compulsorily detained only lem gambling and to promote research into internet, have also allowed greater and easier because the debate reached that stage late in treatment. access to gambling. the evening. “Three million pounds equates to £10 per Gambling Addiction and its Treatment within the “I can assure you there will be a vote on year per problem gambler, which I think is NHS: A Guide for Healthcare Professionals can be treatability at a later stage. I would expect us totally inadequate,” said Mark Griffiths, author seen at www.bma.org.uk. to win substantially on therapeutic benefit.”

BMJ | 20 JANUARY 2007 | VOLUME 334 113 NEWS

IN BRIEF NHS has “appalling lack of workforce planning”

CDC warns of dangers of cough Sally Hargreaves LONDON thought to the future balance between GPs mixtures for young children The government faced renewed criticism and specialists. In some specialties, in some Three deaths of infants in 2005 were last week of its strategies for the training and areas of the country, there is a possibility that linked with cough and cold medicines, the US Centers for Disease Control and recruitment of NHS staff, which may in the there will be more people with specialist quali- Prevention has warned. The infants ranged future result in more doctors being trained fications than there are traditional consultant in age from 1 to 6 months, and all had high than the NHS can afford to pay. posts. blood concentrations of what seemed to The chairman of the BMA, James Johnson, “One reason for this is that medical care be pseudoephedrine. told reporters at a press conference in London and the way the NHS treats patients is con- The survey of deaths was last week that the NHS’s situation “demon- stantly evolving.” based on email queries strates an appalling lack of workforce plan- He added: “Increasingly, there will be a to medical examiners ning.” He called for the reinstatement of more need to encourage medical graduates to con- and a review of news and journal reports, effective planning strategies to avoid wasting sider options other than progression to tradi- and because the response was low it may millions of pounds of public money. tional consultant posts. This is in line with our underestimate the number of cases, it “In 2008 the year on year significant rise vision for services to be delivered closer to says. See www.cdc.gov/mmwr/preview/ in additional NHS resources will fall back patients rather than within secondary care.” mmwrhtml/mm5601a1.htm. dramatically to figures around the 2.5% level. Research shows that outcomes among Despite the extra money NHS trusts all over patients are better the higher the number of MMR payments to lawyers top £14m the country are in deficit, clinics cancelled, local doctors there are locally per patient and The Legal Services Commission has authorised payments totalling more than wards closed, operating theatres being under- that patients want a consultant led service, £14m (€21m; $27m) to lawyers and used, and staff made redundant or posts not Ian Gilmore, president of the Royal College experts involved in cases involving the advertised,” said Mr Johnson. of Physicians, said. measles, mumps, and rubella vaccine, “At the same time we hear that the govern- He warned that a shortage of consult- says the UK Department for Constitutional ment believes that by 2010-11 we will have ant posts for doctors in the near future “has Affairs. Solicitors’ costs came to just over an excess of 3200 consultants,” he added, implications not only for patient care but for £8m and those of barristers to £1.7m. “alongside a shortage of 1200 GPs and 1100 the education and training of future doctors, Total fees and expenses paid to experts too few junior and staff grade doctors” (BMJ as there will be fewer consultants to deliver came to £3.4m, and eight experts each received more than £100 000. 2007;334:61, 13 Jan). this.” “To add to the problems we know there “Workforce planning is difficult at the best Why Irish bar workers are smiling will shortly be a huge bulge in the number of of times but becomes virtually impossible in Irish pubs and bar workers are healthier junior doctors chasing training jobs due to the the context of the rapid implementation of so as a result of the 2004 ban on smoking abolition of the senior house officer grade,” many NHS reforms,” he said. in workplaces, concludes a report in he said. He warned that doctors, who cost a Richard Brooks, an associate director at the American Journal of Respiratory and the minimum of £250 000 (€380 000; $500 000) Institute for Public Policy Research, said that Critical Care Medicine (doi: 10.1164/ to train in the United Kingdom, may be forced tackling problems in the training and recruit- rccm.200608-1085OC). Bar workers’ exhaled breath carbon monoxide fell by to go abroad for work. ment of NHS staff was bound to pose a chal- 79% after the ban, their salivary cotinine A Department of Health spokesman said: lenge for the government, given the sheer fell by 81%, and they had significantly “Analysis so far suggests we need to give more scale and complexity of the NHS. improved pulmonary function.

Somali fighting helps spread deadly virus One in four trainees are worried about career An outbreak of Rift Valley fever that has Sally Hargreaves LONDON istrars from a range of disciplines and asked killed more than 80 people in Kenya could spread uncontrollably in neighbouring Specialist trainee doctors in England are wor- them to complete an online questionnaire Somalia, given the current insecurity and ried about the effect on their job prospects of survey. Thirty of the 127 respondents (24%) the worst flooding in 50 years, aid agencies government plans to treat more patients out- considered that their future employment warn. Emergency control measures have side hospital, concludes research published prospects as consultants were “poor or very been hampered by recent US air attacks on this week. worrying.” Seventy seven (61%) considered fleeing Islamist militants, they say. Ongoing NHS reforms will see a shift in that in their future role as consultants they healthcare delivery into the community and would need to work directly in the commu- Living with children increases fat intake a reduction in the need for hospital based nity, and half of them considered that to be Adults living with children tend to eat more specialists, says the report in the Postgradu- a bad development. fat than those in homes without children, ate Medical Journal (www.postgradmedj.com, A total of 102 (80%) believed that they says a study in the Journal of the American doi: 10.1136/pgmj.2006.054320). “For many needed training on the delivery of specialty Board of Family Medicine 2007;20:9-15. trainees both the immediate and long term care in the community; 96 (76%) wanted this Their cupboards are more likely to be uncertainty generated by these changes have in the form of a university degree delivered stocked with convenience foods high in caused alarm,” say the study’s authors, from by a distance learning programme, although fats, such as biscuits, cheese, peanuts, the Leicester General Hospital. only half thought that this would enhance and processed meats, the researchers say. The researchers approached specialist reg- their prospects of becoming a consultant.

114 BMJ | 20 JANUARY 2007 | VOLUME 334 NEWS

clinics run by Mr Mohamed Taranissi. While Fertilisation authority we have tried to work with his clinics to get the information we require to do our job, we raids controversial have been challenged by this clinic and its lawyer at almost every step of the way. This fertility clinics has greatly lengthened the time it has taken Lynn Eaton LONDON to tackle these issues.” The United Kingdom’s Human Fertilisation The Reproductive Genetics Institute ceased and Embryology Authority (HFEA) raided to have a licence at the end of 2005 and has two private fertility clinics in London this not had one since that date, she said. Yet week. The raids took place hours before the authority had received information that the BBC Panorama television documentary treatment had been given from that centre, broadcast allegations that the clinics were without it having a licence. offering unproved fertility treatments and “That is a matter that is very serious indeed that one was carrying out treatments without and needed investigation,” said Ms McNab. a valid licence. A spokesman for the authority said that after Investigators from the authority took the licence for the clinic expired Mr Taranissi police officers with them to the Assisted applied for another one but did not give the Reproduction and Gynaecology Centre in authority enough information for one to be Upper Wimpole Street, London, and the granted. WALTER WHITE/GETTY IMAGES WALTER nearby Reproductive Genetics Institute in Mohamed Taranissi Ms McNab said the issues of concern Weymouth Street. included the reporting of information and the Both clinics are run by Mr Mohamed court to gain the necessary warrant to search information that had been given to patients Taranissi, a consultant gynaecologist, whose the premises. “This will allow us to gain about the types of treatment available. wealth, the BBC says, may amount to as much unimpeded access to the clinics,” she said. Mr Taranissi said that he had no objections as £38m (€58m; $75m). She denied allegations that the authority to being inspected by the authority. “That is, Mr Taranissi denied that the fertility had failed to regulate the clinics, insisting after all, their job,” he said. “My only issue is treatments he offered were unproved and that they had been following the appropriate the timing and the manner in which it was emphasised on the programme that the processes all along. done.” women he treated gave informed consent. “For some time the HFEA has been taking The fact that the authority announced He admitted on the programme that he regulatory action against clinics run by Mr the raids in a press release just hours before had continued to use the Weymouth Street Taranissi. This is a matter we have been the Panorama programme was broadcast clinic after its HFEA licence was revoked, treating very seriously.” suggested that the programme makers were explaining that he was putting his patients’ She said the process had taken a long time in collaboration with the authority, he said. interests first. because the authority had had to follow the “Why do they [the authority] have to make The authority’s chief executive, Angela correct regulatory procedures. this an issue in a press release. Why not just McNab, said the authority had gone to “We have faced unique difficulties with the regulate?” asked Mr Taranissi. Group asks US institutes to reveal industry ties Janice Hopkins Tanne NEW YORK The draft letter, which was was important because doc- infection to their babies. Treat- The US Center for Science in shown to the BMJ by a US tors are encouraged to follow ment might involve the use of the Public Interest is calling on scientist, is addressed to Elias evidence based guidelines on drugs to prevent neonatal infec- organisations and researchers Zerhouni, director of the NIH, clinical practice, “but who writes tion, such as GlaxoSmithKline’s to sign a letter asking the US and to Anthony Fauci, direc- the evidence [based guidelines]?” valaciclovir (Valtrex). National Institutes of Health tor of the National Institute of The centre’s draft letter calls The draft letter says that (NIH) to reveal ties to industry Allergy and Infectious Diseases, on the NIH to “adopt the neonatal herpes is rare and that among scientists on its advi- Carolyn Deal, chief of that standard long in place at the worries about transmission are sory committees. The centre’s institute’s sexually transmitted Office of Medical Applications “highly contentious.” The letter “integrity in science” project is infections branch, and Walla of Research [OMAR] inside the mentions a front page story in starting by focusing on speakers Dempsey, clinical trials pro- Office of the NIH director.” the Wall Street Journal (2006 Dec at a conference next month on gramme officer at the institute. It continues, “It is our under- 13, p A1) reporting that screening for neonatal herpes. Copies will be sent to the standing that OMAR prohibits GlaxoSmithKline had paid doc- The centre is asking for signa- chairmen of the health related physician-scientists with conflicts tors for speaking engagements tures to a letter protesting at the committees in the House of Rep- of interest from serving on its and continuing medical educa- fact that some of the speakers resentatives and the Senate. consensus panels.” tion sessions promoting the at the NIH conference on 20 Merrill Goozner, director of The NIH conference will screening of all pregnant women February have undisclosed ties the integrity in science project, make recommendations about for herpes. The article said that to companies that make antiviral said that the letter was still being screening pregnant women for more screening could increase drugs to treat herpes. finalised. He said that the issue their risk of transmitting herpes sales of the firm’s drug.

BMJ | 20 JANUARY 2007 | VOLUME 334 115 NEWS

High court upholds GMC’s rejection of case against neurologist Clare Dyer LEGAL CORRESPONDENT The General Medical Council has compre- hensively defeated a challenge in the High Court by the Singapore Medical Council to the GMC’s decision to drop a disciplinary case against the British neurologist Simon Shorvon over a research project he headed in Singapore (BMJ 2003;326:839). The Singapore council found Professor Shorvon guilty of professional misconduct

in February 2004 over the $S10m (£3.3m; LOUNATMAA/SPL DR KARI €5m; $6.5m) project, which was halted amid allegations that researchers breached confi- dentiality, failed to obtain informed consent, Plan to halve MRSA cases by and disregarded the best interests of vulner- able patients. The GMC investigated the allegations, 2008 will not be achieved but when the case reached the investigation committee of the fitness to practise panel Michael Day LONDON ing all the hospital infection statistics so that the chairman decided to discontinue it in The government will fail to meet its pledge of “superbug” infections are not singled out. But September 2005. halving the numbers of infections of methi- it admits that some of these measures will be The Singapore regulatory body sought cillin resistant Staphylococcus aureus (MRSA) seen as a “cop out” or “fiddling” the figures. a High Court ruling quashing the decision, in UK hospitals by next year, says a memo The Liberal Democrat’s health spokesman, which it argued was reached by a process of leaked from the Department of Health. Norman Lamb, said that the leaked memo flawed reasoning. David Pannick QC, for the The memo also says that the target may showed that ministers were losing the fight Singapore council, also contended that the prove unachievable and that “a certain level against hospital acquired infections. He said, decision was procedurally flawed because of MRSA is unavoidable and we don’t know “The government’s strategy is failing.” the GMC was in breach of a duty to inform what that level is.” A spokeswoman for the health department it before taking a final decision to drop the It discusses how the government might said, “Progress towards a 50% reduction of case. avoid the resulting flak and find its way out of MRSA blood poisoning has been slower than But Mr Justice Davis, who described the trouble. And it warns that another potentially anticipated, and though numbers of infections litigation as “unfortunate,” rejected both argu- deadly bacterial infection, Clostridium difficile, are coming down, faster progress is needed to ments. He said the chairman of the investi- which causes around 50 000 infections and meet the target.” gation committee had “reached a conclusion around 1000 deaths a year, is now “endemic” She added that the government “remained which was rational and sustainable and one in NHS wards. It says that some trusts now committed to getting on top of the problem of he was entitled to reach.” “see C difficile as a fact of life.” MRSA and other hospital infections.” The chairman had decided to drop the case In 2004 the then health secretary, John However, the epidemiologist Mark Enright, after an independent expert consulted by the Reid, pledged to halve the number of cases of Imperial College London, said, “I haven’t GMC supported the views of four experts of MRSA blood poisoning by 2008. met anyone in the health service who ever who produced reports on behalf of Professor However, the leaked memo, written by the thought the target was achievable. Shorvon that his role in the research did not health department’s director of health protec- “Currently hospitals are struggling to amount to professional misconduct. tion, Liz Woodeson, and circulated in Octo- contain outbreaks of MRSA and C difficile, The judge added: “The reasoning was logi- ber, says that the NHS “is not on course to and this will only get worse if, as seems likely, cal and supportable and stands up to analy- hit that target and there is some doubt about nursing staff are made redundant due to sis.” Professor Shorvon said, “It has been a whether it is in fact achievable.” financial pressures on NHS trusts.” protracted struggle against the Singaporean The document, which was passed to the Hugh Pennington, emeritus professor of authorities, and I am pleased to have been Health Service Journal, shows that the problems bacteriology at Aberdeen University, said that exonerated by the GMC and that the High are widespread. It says that 116 acute NHS only the adoption of the rigid, zero tolerance Court upheld this decision. I have maintained trusts, or one in three, are “underperforming” policies as successfully used by Scandinavian all along that I acted in good faith in conduct- in relation to the target and that overall the countries would bring about radical falls in ing the research project and followed inter- NHS is off course by 27%. rates of infection. He noted, however, that nationally accepted professional and ethical It says that ministers might quietly do away extra resources in terms of staff and isolation standards.” with the target or fudge the figures by merg- facilities would be needed.

116 BMJ | 20 JANUARY 2007 | VOLUME 334 NEWS

US health spending grew more slowly in 2005

Janice Hopkins Tanne NEW YORK states, which pay for part of Medicaid, tried ANNUAL GROWTH IN SPENDING ON Spending on health care in the United States to cut or freeze payments, instituted cost HEALTH CARE AND IN GDP IN THE US grew more slowly in 2005 than in any year control measures, increased efforts to fight Recession since 1999, but it still grew at twice the rate fraud and abuse, and made greater use of Spending on health care Gross domestic product (GDP) of inflation and faster than the rise in wages. disease management programmes. 16 In their annual report on healthcare Consumers spent more on health care in spending, officials from the US Centers for 2005 ($249bn) than in 2004 ($236bn). Just 12 Medicare and Medicaid Services reported over 20% of consumer spending was on 8 in the journal Health Affairs (2007;26:142-53) prescription drugs. Consumers with health Annual increase (%) that, for the third year in a row, healthcare insurance are required to pay part of the cost 4 spending grew at a slower rate than in the of a prescription, and health insurers have 0 previous year. In 2005 it grew by 7%, increased the amounts that consumers pay 1981 1985 1990 1995 2000 2005 reaching nearly $2000bn (£1000bn; for brand name drugs, to encourage them to Year €1500bn). use generic drugs. Source: Health Affairs The authors said it wasn’t clear whether Spending on hospital care was the largest the slowdown in spending was temporary or item in healthcare spending, and the increase Americans do not have health insurance. represented a long term trend. remained steady at 7.9% between 2001 Karen Davis, a health economist and head In 2005 the US spent 16% of its gross and 2005. The second biggest category was of the non-profit Commonwealth Fund, said, domestic product on health care, up only spending on physician and clinical services, “We do not get good value for our healthcare slightly from 15.9% in 2004. The US spent which slowed since 2000. dollar, compared with other countries, or almost $6700 per person on health care Meanwhile, spending on home health care even with the best examples in the US.” in 2005, more than double what most and nursing homes grew by 6% from 2004 to She said that the US needed “to undertake a industrialised countries spent. 2005. This is up from a 4.1% increase from major drive toward value and efficiency.” The reason for the slowdown in growth 2003 to 2004. Spending on home health care Meanwhile, states are working to offer was slowed spending on drugs, which grew was the fastest growing category between health insurance to their citizens. Last only 5.8% in 2005. Drug spending by the 2003 and 2005. Almost two thirds of nursing week Governor Arnold Schwarzenegger of Medicaid programme for poor people also home care was paid by public programmes, California announced a plan to cover all the grew more slowly. This was because the such as Medicaid. About 47 million state’s 36 million residents.

GMC says doctor gave out slimming pills “as if they were Smarties”

Owen Dyer LONDON £100 (€150; $200) a week He took the collected He is also accused of A doctor at a London slimming on drugs prescribed by Dr evidence to the police. Three indecently assaulting a female clinic doled out pills “as if they Bondyopadhyay. She had months later, inspectors from the patient while working as a locum were Smarties,” the General developed depression after being Healthcare Commission raided GP at the Smallthorne Health Medical Council heard last forced by psoriasis to quit her Dr Bondyopadhyay’s clinic. Centre in Stoke-on-Trent, in week. job as an intensive care nurse. They found loose prescription January 2004. Telling her he Imon Bondyopadhyay, Her regular GP was never drugs on a desk and in drawers; wanted to listen to her chest, who ran the London West told she was taking the drugs, these included expired drugs he attempted to remove her Slimming Clinic in Townsend said Mr Kark. He added: “Dr and drugs not licensed in the bra before “feeling around her Road, Southall, west London, Bondyopadhyay was doling out United Kingdom, Mr Kark told chest area” and “tweaking her is accused before the GMC’s pills as if they were Smarties, the hearing. nipples,” Mr Kark told the panel. fitness to practise panel of with little or no regard for the A plaque at the front of Dr Bondyopadhyay is also prescribing excessive quantities consequences of his actions.” the slimming clinic listed Dr accused of threatening a local of addictive drugs, including Ms A was found dead at her Bondyopadhyay’s wife as a healthcare manager after being benzodiazepines, phentermine, home in February 2004. Her doctor, although she is not removed from a list of doctors and buprenorphine. brother, a registered GP, told registered to practise in Britain. allowed to practise as a GP. Mr One patient, referred to the GMC how, on visiting her The clinic was not properly Kark told the hearing that Dr as Ms A, died after taking a Brentford flat two days after the registered, the GMC charges. Bondyopadhyay threatened “therapeutic dose” of lorazepam discovery, he found dozens of Dr Bondyopadhyay is Peter Savege, head of healthcare and diazepam combined with padded envelopes containing also charged with providing services at Enfield Primary Care alcohol, leading to a coroner’s drugs that had been delivered phentermine to an overweight Trust, that he would instruct verdict of misadventure. by taxi or by hand. Credit card man, although the patient’s heart “powerful non-white solicitors” The GMC counsel Tom slips in the bags identified Dr disease and diabetes should have to accuse him of racism. Kark said Ms A spent about Bondyopadhyay’s clinic. precluded the drug’s use. The case continues.

BMJ | 20 JANUARY 2007 | VOLUME 334 117 NEWS

SHORT CUTS WHAT’S NEW IN THE OTHER GENERAL JOURNALS Alison Tonks, associate editor, BMJ [email protected]

Candidate drug for familial results but say their new compound is To test their hypothesis further the hypercholesterolaemia needs unlikely to get much further until they sort researchers also looked for associations more work out its adverse effects on the liver. Concen- between poor adherence to the heart drugs trations of liver enzymes went up substan- and admissions for cancer—a link you might EFFECT OF BMS-201038 ON LIVER ENZYME AND LIVER FAT VALUES tially in four of the six patients, and the same expect to find if poor adherence was associ- number developed a fatty liver. By the end ated with unhealthy lifestyles. They found Serum alanine aminotransferase 800 of the study, two patients had a hepatic fat no associations between adherence and can- Patient 1 content of 30%. cers of the lung, breast, or prostate. Patient 2 600 N Engl J Med 2007;356:148-56 JAMA 2007;297:177-86 Patient 3 Patient 4 400 Patient 5 Adherence prolongs survival after Syphilis returns to China Patient 6 200 heart attack

Alanine aminotransferase (IU/l) Syphilis has made a dramatic comeback in 0 DRUG ADHERENCE AND MORTALITY China. According to national surveillance ACCORDING TO MORTALITY RISK Liver fat figures, the incidence of primary or second- 50 Adherence ary infection with Treponema pallidum has

40 Low Intermediate High increased more than 25-fold since 1993, from

Liver fat (%) High risk 0.2 to 5.7 cases per 100 000. These numbers 50 30 dwarf the more modest resurgence of syphi- 40 lis already reported in the US, Canada, and 20 30

Mortality (%) Europe. They are accompanied by an even 20 10 more striking upswing in the incidence of 10 congenital syphilis, from 0.01 cases per 100 0 0 00123 41234123412341234 000 live births in 1991 to 19.7 such cases Intermediate risk Week 40 in 2005. 0.03 0.10 0.30 1.00 Washout 30 Commentators blame China’s explosive BMS-201038 (mg/kg/day) 20 economic growth for the re-emergence of a N ENGL J MED Mortality (%) 10 disease that was successfully eliminated by People with severe familial hypercholestero- 0 the communist regime during its first few Statins  blockers Calcium channel laemia have non-functioning receptors for blockers decades in power. But more recent social low density lipoprotein (LDL) cholesterol. JAMA changes including mass migration of young This defect leads to sky high serum concen- People who keep taking their heart disease working men, the reappearance of a sex trations of total and LDL cholesterol, early tablets as directed tend to do better than industry, and the rising cost of health care cardiovascular disease, and death. Treat- people who don’t. Is this improved progno- have fuelled an epidemic. China’s pros- ment options are limited because conven- sis a result of the drugs or do people who perous southeastern seaboard is the worst tional drug treatments don’t work, and LDL diligently take tablets tend to be health- affected region. The highest incidence is in aphoresis—a kind of cholesterol dialysis—is ier in general than people who won’t, or Shanghai, where there are 55.3 cases per intensive, expensive, and not widely avail- can’t? To find out, researchers studied a 100 000. The researchers who collated the able. Production of LDL cholesterol can be large cohort of survivors of heart attack latest figures say the situation is likely to stopped at its source, however, by inhibiting from Ontario who were discharged from be worse than it looks. China has a good the hepatic protein that transfers triglycer- hospital with prescriptions for a statin, a  surveillance infrastructure, but up to three ides on to apolipoprotein B to make very blocker, or a calcium channel blocker. quarters of sexually transmitted diseases go low density lipoprotein, the precursor of After a median follow-up of nearly two unreported in some areas. LDL. Researchers recently tested such an and half years, the authors found a clear link Lancet 2007;369:132-8 inhibitor (BMS-201038) in six adults with between worsening adherence and increasing homozygous familial hypercholesterolaemia, risk of death for statins and  blockers, drugs and it had the desired effect. Their mean that improve survival after a heart attack. No Many Canadian immigrants serum concentrations of LDL cholesterol fell such link was seen between death and adher- need MMR from 15.9 to 7.8 mmol/l, a significant fall of ence to calcium channel blockers, which more than 50% (P<0.001). BMS-201038 had don’t improve survival. The authors infer A survey of immigrants and refugees living a similar effect on total cholesterol during 16 that patients who take their tablets live longer in one culturally diverse area of Montreal weeks of treatment at an increasing dose. because of the biological action of the drugs in Quebec found that 36% of people tested The researchers were pleased with the not the so called “healthy adherer effect.” were susceptible to measles, mumps, or

118 BMJ | 20 JANUARY 2007 | VOLUME 334 NEWS

rubella. This prompted calls for catch-up token gesture to “diversity.” Overt preju- tion of the US was in prison, on probation, vaccination programmes or at the least dice from patients is indulged, without or on parole. heightened surveillance by primary and discussion, by division chiefs. N Engl J Med 2007;356:157-65 secondary healthcare facilities. “It is a drain to carry this burden,” says The researchers surveyed 1480 adults one respondent. who had come to Canada from six devel- “My burden is to deal with the pressure Patients with drug eluting stents oping regions including south Asia, Latin of whatever stereotypes people may have do better on long term clopidogrel America, Africa, and eastern Europe. All about race... and it is a daily stress at work. were recruited from two hospitals and It’s exhausting.” CLOPIDOGREL AND MORTALITY three primary care clinics in Montreal. Ann Intern Med 2007;146:45-51 AND HEART ATTACK Almost half were refugees. They were Mortality 60 well educated and most were in the mid- Drug eluting stent With clopidogrel dle or high incomes brackets in their home US prisoners have high death Without clopidogrel 40 countries. rates after release Bare metal stent Susceptibility to measles, mumps, or With clopidogrel Without clopidogrel rubella varied between 22% and 54% MORTALITY IN FORMER PRISONERS 20

depending on age, sex, and country of Cumulative incidence (%) 3000 origin. Immigrant or refugee women were 0 significantly more likely than men to be 2500 susceptible to measles (odds ratio 2.1, 95% 2000 Death or myocardial infarction 60 1500 CI 1.2 to 3.8) and rubella (1.7, 1.2 to 2.6). Adjusted mortality rate for residents The authors say their results are 1000 worrying and probably reflect the poor 40 500 global coverage of the measles, mumps, 0 and rubella (MMR) vaccine compared Deaths per 100 000 person years Overall 1-2 3-4 5-6 7-8 9 20

with coverage in Canada and the US. Pub- Weeks after release Cumulative incidence (%) N ENGL J MED lic health authorities need to close the gap 0 or risk repeated outbreaks of these child- Convicts released from Washington State 12 15 24 hood diseases. prisons are nearly 13 times more likely to Months JAMA Ann Intern Med 2007;146:20-4 die during their first two weeks of freedom than other residents of the state (relative Drug eluting stents work well in the short risk 12.7, 95% CI 9.2 to 17.4), according term, but longer follow-up of patients with Race is an exhausting burden for to a new study. these stents suggested they were associated some black American doctors Among the 30 237 inmates released dur- with an increased risk of catastrophic resten- ing the study, 38 died within two weeks, osis several years after a heart attack. No one American doctors of African descent 27 of them from a drugs overdose. In total, knows exactly why. It is possible, however, that reported recently that issues of race domi- 443 died in the first two years after release. antiplatelet agents such as clopidogrel could nate their working lives, affect their rela- This mortality was more than three times prevent restenosis if taken for long enough. tionships with patients and colleagues, the expected rate in a population of simi- In the absence of any clinical trials of and create a silent burden that takes its lar age, sex, and cultural heritage. Again, extended treatment with clopidogrel, toll on their careers. The 25 doctors, who a drugs overdose was the leading cause of researchers looked for clues in a cohort were all from New England, took part in death (103 deaths, relative risk 12.2, 95% of 4666 adults given drug eluting or bare a qualitative study to explore the impact CI 10.2 to 14.9), and cocaine was the most metal stents at one specialist centre in the of race on the working lives of African, common drug. Homicide (55 deaths, 10.4, US. They found a significant association African American, or African Caribbean 8.0 to 13.6), suicide (40 deaths, 3.4, 2.5 to between long term treatment with clopi- doctors working in diverse sectors of US 4.7), road traffic incidents (35 deaths, 3.4, dogrel (six or 12 months) and a lower risk health care. 2.4 to 4.8), and cardiovascular disease (57 of death over two years in patients with They described being constantly aware deaths, 2.1, 1.6 to 2.7) were other impor- drug eluting stents. Patients on extended of their racial minority status at work, an tant causes of death in recently released treatment also had a lower risk of death awareness that extended to others but was prisoners. or heart attack over two years. Long term never discussed. One general surgeon These values are consistent with find- clopidogrel seemed to make no difference said, “When I walk on to a ward or on ings from similar studies outside the US. to patients with bare metal stents. the floor, I’m a black guy before I’m the They confirm that the transition period The authors used propensity scoring to doctor. I’m still a black guy before I’m the immediately after release is stressful, and adjust for any systematic differences between guy in charge, before I’m the attending of ex-prisoners are at high risk as they strug- patients who stopped treatment and those record, so that permeates everything.” gle to reintegrate with their families and who continued with clopidogrel for six or Others described being held to higher communities, find work and housing, and 12 months. But they, and a linked editorial performance standards than their non- access health care—including mental health (pp 209-11), are still cautious about the find- black colleagues, being excluded from care—write the authors. ings, which rely on patients remembering leadership roles or top academic jobs, and Nationally, millions of people are at risk. accurately the length of their treatment. being press ganged on to committees as a At the end of 2004, 3% of the adult popula- JAMA 2007;297:159-68

BMJ | 20 JANUARY 2007 | VOLUME 334 119 the rofecoxib case

What have we learnt from Vioxx? In October UK patients who had cardiovascular events while taking rofecoxib lost the right to fight Merck in the US for compensation. But researchers and journals can still benefit from this case if they learn from the mistakes, write Harlan Krumholz and colleagues

Rofecoxib (Vioxx) was intro- volunteers by about half.w2 In internal emails cardiovascular risks, despite FDA concern,5 duced by Merck in 1999 as an made public through litigation,3 Merck offi- and disseminated the results to promote the effective, safer alternative to cials sought to soften the academic authors’ drug’s cardiovascular safety to doctors in its non-steroidal anti-inflamma- interpretation that cyclo-oxygenase-2 (COX “cardiovascular card,”6 7 a marketing device tory drugs for the treatment 2) inhibition within the vascular endothe- cited by US Congressman Henry Waxman of pain associated with osteoarthritis. It was lium may increase the propensity for throm- for falsely minimising cardiovascular risks8 subsequently found to increase the risk of bus formation, the basis of what became and never approved by the FDA. cardiovascular disease and withdrawn from known as the FitzGerald hypothesis.w3 The the worldwide market. Merck now faces academic authors changed the manuscript at The VIGOR study legal claims from nearly 30 000 people who Merck’s request—for example, they changed In January 1999, Merck launched its larg- had cardiovascular events while taking the “systemic biosynthesis of prostacyclin ... was est study yet of rofecoxib, the Vioxx gastro- drug.1 The company has stated that it will decreased by [rofecoxib]” to “Cox-2 may intestinal outcomes research (VIGOR) study. fight each case, denying liability.2 Our recent play a role in the systematic biosynthesis of The study was intended to expand the drug’s participation in litigation at the request of prostacyclin.”3 w2 To the authors’ credit, they approved indications by showing that it plaintiffs provided a unique opportunity to continued to investigate the effects of COX would have fewer gastrointestinal side effects thoroughly examine and reflect on much of 2 inhibition and ultimately provided much than naproxen for the treatment of rheuma- the accumulated court documents, research, of the basic science knowledge that clarified toid arthritis. The study of over 8000 patients and other evidence. This story offers impor- the pathways by which rofecoxib probably was initiated without a standard operating tant lessons about how best to promote con- leads to cardiovascular events.w4-w7 procedure for collecting information on car- structive collaboration between academic However, despite Merck’s knowledge that diovascular events and without a cardiologist medicine and industry. rofecoxib might increase thrombus forma- on the data safety monitoring board. Data tion, none of the intervention studies that safety monitoring boards are independent Early suspicion of cardiovascular risk constituted its new drug application to the committees whose purpose is to monitor Since the early development of rofecoxib, Food and Drug Administration in 1998 were the results of an ongoing trial to ensure the some scientists at Merck were concerned designed to evaluate cardiovascular risk. safety of trial participants.w8 The study was that the drug might adversely affect the car- The nine studies were generally small, had designed to continue until a predetermined diovascular system by altering the ratio of short treatment periods, enrolled patients at number of confirmed uncomplicated or prostacyclin to thromboxane, which act in low risk of cardiovascular disease, and did ­complicated gastric perforations, ulcers, or opposition, balancing blood flow and clot- not have a standardised procedure to collect bleeds had occurred. ting.w1 A study sponsored by Merck during and adjudicate cardiovascular outcomes.4 The first non-endpoint safety analysis was 1996-7 reported that rofecoxib reduced uri- Moreover, Merck seemingly pooled data presented to the safety board in November nary metabolites of prostacyclin in healthy from these studies and others for analysis of 1999, at which time a 79% greater risk of

120 BMJ | 20 jANUARY 2007 | Volume 334 the rofecoxib case

Vioxx in the dock: lawyer Mark Lanier holds up a sample packet of rofecoxib as he speaks during proceedings against Merck in New Jersey in March 2006

­contained data from an interim analysis that had different termination dates for cardio­ vascular and gastrointestinal events (gastro­ intestinal events were counted for one month longer than the cardiovascular events). This highly irregular procedure was not described in the publication and had the effect of favour- ing the drug’s effect on gastrointestinal events while understating the risk of cardiovascu- lar events.w9 The published cardiovascular risk was not accurate because three addi- tional myocardial infarctions occurred in the rofecoxib group in the month after the researchers stopped counting cardiovascular events (none had occurred in the naproxen group). The potential harm was further mini- s c i

p mised by a post hoc subgroup analysis based m on “indication for aspirin prophylaxis”; had /ap/e s Merck included the three cases, the subgroup analysis would have shown an increased

erlevan ­cardiovascular risk in both groups.w10 m The publication concealed the cardio­ death or serious cardiovascular event was naproxen, it could be used to petition the vascular risk even further by presenting the found in one treatment group compared with FDA for a new indication. However, if the hazard of myocardial infarction as if naproxen the other (P=0.007).9 The board allowed the study raised concerns about cardiovascular was the intervention group (relative risk 0.2, study to continue and planned to review sub- harm, the billion dollar drug franchise 0.1 to 0.7) and without reporting the absolute group analyses in December, at which time would be threatened. The study showed number of cardiovascular events, even though the analyses again showed higher cardio- that rofecoxib was not more effective in all other results were presented appropriately vascular risk in one group. On this basis the relieving symptoms of rheumatoid arthritis with rofecoxib as the intervention group.w11 board recommended that an analysis plan be but did halve the risk of gastrointestinal Finally, the authors proposed a naproxen developed to examine serious cardiovascular events. However, there was also evidence hypothesis, suggesting that rofecoxib had not events and that the study continue until it of an increased risk of myocardial infarction been harmful but that naproxen had been reached its gastrointestinal endpoint target ­(relative risk 5.00, 95% confidence interval protective, despite there being no accepted (expected March 2000). 1.68 to 20.13). When this result was circu- evidence that naproxen had a strong cardio- Matters were complicated by the existence lated internally at Merck, Edward Scolnick, protective effect. of conflicts of interest among board mem- the company’s chief scientist, Merck strongly promoted bers. According to Merck policies, the board wrote in an email to colleagues The published VIGOR the VIGOR study, purchas- is supposed to be independent, without finan- about the cardiovascular risk: ing nearly 1 million reprints study obscured the cial or emotional stake in the trial being mon- “It is a shame but it is a low to circulate to doctors and itored.10 Yet, the head of the VIGOR board incidence and it is mechanism cardiovascular risk other health professionals. was awarded a two year consulting contract based as we worried it was. associated with The New England Journal of two weeks before the trial ended and as the [Merck employees/consultants] rofecoxib Medicine reported problems trial was concluding disclosed family owner- were right about the metabo- with the study in an “expres- ship interest in Merck shares worth $70 000 lite meanings, ie, urine [prostacyclin] data.”13 sion of concern” published in 2006,w10 and the (£37 000; €55 000).11 12 Although it is not pos- This indicates that, at the least, there were editor in chief has said that the authors “with- sible to tell whether this financial relationship grounds for suspicion within Merck before held critical data on the cardiovascular toxi­ made any difference, the conflict of interest the VIGOR study was published that Vioxx city of Merck’s drug Vioxx.”14 Nevertheless, was not a matter of public record at the time was associated with cardiovascular risk. none of the authors has publicly conceded the trial was conducted or published and of error or taken responsibility for the biased itself calls into question the independence of Obscuring the risk presentation of the study results. In fact, two the safety board. Despite the concern articulated by Dr VIGOR authors and the head of the VIGOR The VIGOR study had enormous financial ­Scolnick, the published VIGOR study board continue to collaborate on high profile implications for Merck. If it showed rofecoxib obscured the cardiovascular risk associated research with Merck.15 to have better gastrointestinal safety than with rofecoxib in several ways. The report Except for a 2001 study published in

BMJ | 20 jANUARY 2007 | Volume 334 121 the rofecoxib case

JAMA that raised questions about the safety of concern”w9 w10 and a correctionw14 and yet, there is hardly a sense of outrage in the of rofecoxib and the validity of the naproxen publishing a methodological paperw15 and profession about the events that transpired. hypothesis,w12 few academic researchers other related comments and editorials.w16-w24 Defenders of Merck may say that we do ­publicly questioned the company before But other academic medical journals also not know how rofecoxib’s cardiovascular its voluntary withdrawal of the drug. More­ played important parts. In 2001, Circulation risk compares with that of other COX 2 over, Merck selectively targeted doctors ­published a pooled analysis of 23 phase IIb- inhibitors or traditional non-steroidal anti- who raised questions about the drug, going V studies examining the association between inflammatory drugs. But the proper place of so far as pressurising some of them through rofecoxib and cardiovascular risk. The paper these drugs in the medical armamentarium department chairs.16 had no editorial commentary or critique,w25 is beside the point. With billions of dollars at even though the study was coordinated inter- stake, Merck conducted the trials, stored and Short and long term use nally at Merck, the results highly favoured analysed the data internally, paid academic For several years, Merck continued to inves- the safety of rofecoxib, and five of the seven researchers as consultants to the investiga- tigate other indications for rofecoxib and authors were Merck employees (the two tive teams and the safety monitoring boards, conducted additional trials. The increased academic authors acknowledged being paid and maintained heavy involvement in the cardiovascular risk compared with pla- consultants to Merck). Moreover, in internal writing and presentation of findings. The cebo was reported in a 2004 analysis of the emails made public through litigation, even journals published the studies, and the aca- adenomatous polyp prevention on Vioxx an executive scientist at Merck criticised the demic community accepted the findings (APPROVe) study,w13 which led to the drug’s analysis, stating: “The data appears to have without expressing much concern. Nearly withdrawal. The financial implications were been interpreted to support a preconceived 107 million prescriptions for rofecoxib immense not only because of loss of rev- hypothesis rather than critically reviewing were dispensed in the US between 1999 enue but also because of expected litigation. the data to generate hypotheses.”17 and September 2004,21 when the drug was The key question was when the risk became The Annals of Internal Medicine published withdrawn from the market, and none of the manifest. If short term use was not associ- the assessment of differences between Vioxx people picking up those prescriptions had ated with increased cardiovascular risk, Mer- and naproxen to ascertain gastrointestinal the opportunity to consider the true balance ck’s liability would potentially be drastically tolerability and effectiveness (ADVAN- of its risks and benefits. reduced. TAGE) study.w26 It later learnt that article was What should we do going forward? The APPROVe authors, five of whom written by Merck without accreditation,w27 ­Academic medicine, industry, medical were Merck employees and the remain- w28 contained errors in the presentation of journals, and government agencies need to der of whom received con- cardiovascular events with come together to define a set of principles by sulting fees from Merck, Merck selectively rofecoxib (minimising car- which we can restore faith in collaborations asserted that the increased targeted doctors diovascular risk), and was on new treatments that can improve patient risk became apparent only who raised questions conducted for marketing care. We might consider adopting some new after 18 months of use.w13 about the drug, going purposes, a so called seeding approaches. Academics engaged in industry This conclusion was based so far as pressurising trial. The journal was quick designed and sponsored studies should insist w29 on an analysis that was not some of them through to condemn ghostwriting that the data are stored on an academic site, prespecified and a flawed and a full correction of the analysed by non-company investigators, and department chairs methodological approach. errors was published recent- eventually made accessible to the public for Merck subsequently admit- lyw30 after Merck scientists scrutiny. Several early, large clinical trials ted that it had incorrectly described the provided an initial, but incorrect explana- of rofecoxib were not published in the aca- statistical approach, and the New England tion.w31 Many other journals have published demic literature for years after Merck made Journal of Medicine issued a correction indi- articles with results favourable to rofecoxib them available to the FDA,22 preventing cating that statements regarding an increase that court documents have shown to be independent investigators from accurately in risk after 18 months should be removed ghostwritten by scientific writing companies determining its cardiovascular risk using from the article.w14 Again, mistakes that hired by Merck.w32-w36 meta-analysis. In addition, independent favoured the company, with colossal eco- audits should be conducted to ensure nomic implications, made it through the Promoting constructive collaboration that companies follow a standard- journal peer review process to the profes- The rofecoxib case is bad news for industry, ised, prespecified protocol. sion and the public. academics, journals, and the public. Merck Independent data and safety was once one of the US’s most publicly monitoring boards should be Medical journals admired companies,w37 and its behaviour mandated and their govern- The New England Journal of Medicine has had a may not be different from that of others in ance should not be under the prominent role in the story. It published the the pharmaceutical or biotechnology indus- control of the company. VIGOR and APPROVe studies, respond- try. Journalists have questioned the ethics of Industry should not be ing to their inaccuracies with “an expression industry and academic researchers.18-20 And allowed to select who

122 BMJ | 20 jANUARY 2007 | Volume 334 the rofecoxib case

serves on these boards or allowed to com- 208088, New Haven, CT 06520-8088, USA 7 Waxman HA. Merck documents show aggressive Joseph S Ross is instructor, Department of Geriatrics and Adult marketing of Vioxx after studies indicated risk. US House pensate members after their service. of Representatives Committee on Government Reform, Development, Mount Sinai School of Medicine, New York, USA In considering articles for publications, 2005. www.democrats.reform.house.gov/story.asp?ID journals should understand Amos H Presler is research associate, =848&Issue=Prescription+Drugs. Never Again Consulting, Attleboro, MA, 8 Waxman HA. Memo re: the marketing of Vioxx that studies with immense In considering articles USA to physicians. Washington, DC: US House of financial implications require Representatives, 2005. for publications, David S Egilman is clinical associate 9 Weinblatt M. Memo to Drs. Bjorkman, Neaton, Shapiro, a higher level of scrutiny journals should professor, Department of Bio Med Silman, and Sturrock re: Interim non-endpoint safety than others, especially when understand that Community Health, Brown University, analysis of VIGOR—unblinded minutes. November the study is conducted by the Providence, RI, USA 18, 1999. Found in sNDA S-007: P088C: Appendix studies with immense 3.9.2 at p.2939-2946 (Bates No. MRK-00420015464- company with the financial Correspondence to: H M Krumholz MRK-00420015471. www.vioxxdocuments.com/ stake. Journals should be pre- financial implications [email protected] Documents/Krumholz_Vioxx/VigorDSMB.pdf. 10 Merck. Medical affairs procedures and policies. pared to go beyond the usual require a higher level Contributors and sources: HMK’s research is focused on determining Procedure 23: collaborative research efforts and of scrutiny than others megatrials. Appendix 2: Merck guidelines for data high quality review, paying optimal clinical strategies and particular attention to the pos- and safety monitoring boards. 29 Feb, 1999. Bates identifying opportunities for Nos MRK-AFK0047772 to AFK0047791. www. sibility of bias. Articles should be accompa- improvement in the prevention, treatment, and outcome of vioxxdocuments.com/Documents/Krumholz_Vioxx/ nied by editorials by people without financial cardiovascular disease with emphasis on older populations. Merck1999guidelines.pdf. He was responsible for the concept and writing of this article 11 Reicin AS. Letter re: financial disclosure for Merck conflicts of interest. Moreover, ghostwriting and is its guarantor. JSR has studied and reported on conflict and Co, Inc sponsored protocol entitled: “A double- constitutes a false statement of authorship or a of interest in medicine. He participated in the concept and blind, randomized, stratified, parallel-group study design of this article and revised it for critical content. AHP to assess the incidence of PUBs during chronic false attribution of authorship, and academic treatment with MK-0966 or naproxen in patients with researchers who sign off or “edit” original has research interests in the history of public health and rheumatoid arthritis (VIGOR).” 4 Feb, 2000. Merck. law, including pharmaceutical research and marketing. publications or reviews written by industry Bates Nos MRK-MEW00012 to MRK-MEW00014. www. He consults on this topic at the request of plaintiffs in the vioxxdocuments.com/Documents/Krumholz_Vioxx/ should be penalised unless there is full disclo- Vioxx litigation. He contributed document research and ReicinWeinblatt2000.pdf. sure of the authorship, such as: “Representa- interpretation to this article and participated in its revision. 12 Merck and Co. Multidisciplinary strategic advisory tives from XYZ drafted the manuscript; the DSE has studied and reported corporate corruption of board for cox-2 inhibitors, consulting agreement. science. He contributed to the development of this article and February 29, 2000. Bates Nos MRK-STI0037747 to authors were responsible for critical revisions revised the work for critical content. This article arose from STI0037751. www.vioxxdocuments.com/Documents/ of the manuscript for important intellectual Krumholz_Vioxx/WeinblattContract.pdf. access to Merck documents as a result of tort litigation. 13 Scolnick EM. Email communication to Deborah Shapiro, content.” Competing interests: HMK has research contracts with the Alise Reicin, and Alan Nies re: Vigor. 9 Mar, 2000. Bates Even the best oversight cannot always American College of Cardiology and the Colorado Foundation No MRK-ABH0016219. www.vioxxdocuments.com/ detect mistakes. When journals discover that for Medical Care; serves on the advisory boards of Amgen, Documents/Krumholz_Vioxx/Scolnick2000.pdf. Alere, and UnitedHealthcare; is a subject expert for VHA; and 14 Drazen JM. Hidden data counfounds medical journal editors. Wall Street Journal 2006 May 19:A11. information has been withheld or that results is editor in chief of Journal Watch Cardiology. All authors have are incorrect, they need to rapidly dissemi- 15 Cannon CP, Curtis SP, FitzGerald GA, Krum H, Kaur A, been consultants at the request of plaintiffs for recent suits Bolognese J, et al. Cardiovascular outcomes with nate that information and ensure that any web against Merck related to rofecoxib. etoricoxib and diclofenac in patients with osteoarthritis search that identifies the errant manuscript and rheumatoid arthritis in the multinational etoricoxib and diclofenac arthritis long-term (MEDAL) programme: also identifies the correction. Authors should 1 United States Securities and Exchange Commission. Form 10-Q: quarterly report pursuant to section 13 or a randomised comparison. Lancet 2006;368:1771-81. sign agreements that they will notify journals 15(d) of the Securities Exchange Act of 1934 (Note 7 16 Fries JF. Letter to Raymond Gilmartin re: physician if such information becomes available or face to consolidated financial statements). Merck, 2006; intimidation. 9 Jan, 2001. Merck. Bates No http://phx.corporate-ir.net/phoenix.zhtml?c=73184& MRK-ABH0002204 to MRK-ABH0002207. www. being blacklisted by the journal. p=irolSECText&TEXT=aHR0cDovL2NjYm4uMTBrd2l6YXJ vioxxdocuments.com/Documents/Krumholz_Vioxx/ Our system depends on putting patients’ kLmNvbS94bWwvZmlsaW5nLnht Fries2001.pdf. 17 Morrison BW. Email communication to Rhonda interests first. Collaborations between aca- bD9yZXBvPXRlbmsmaXBhZ2U9NDMxNDgxMiZkb2 M9MSZudW09MTI=. Sperling, Alise Reicin, Deborah Shapiro, et al. re: fw: demics, practising doctors, industry, and 2 Merck. Vioxx (rofecoxib): frequently asked questions. for review [peer]: 2001-ms-2470 (full paper) - due journals are essential in advancing knowl- www.vioxx.com/rofecoxib/vioxx/consumer/faq.jsp. date Monday, 27 August 2001. 17 Aug, 2001. Bates 3 Morrison BW. Memo to Alan Nies, Barry Gertz, Beth Nos MRK-ACF0005697 to MRK-ACF0005699. www. vioxxdocuments.com/Documents/Krumholz_Vioxx/ edge and improving the care of patients. Seidenberg. Letter to Ken Lasseter; manuscript draft Morrison2001.pdf. Trust is a necessary element of this part- for protocol 023. 18 Feb, 1998. Merck. Bates Nos MRK- 18 Meier B, Kolata G, Pollack A. Medicine fueled by NJ0017794 to MRK-NJ0017822. www.vioxxdocuments. nership, but the recent events have made marketing intensified trouble for pain pills. New York com/Documents/Krumholz_Vioxx/Morrison1998.pdf Times 2004 Dec 19:A1. it necessary to institute proper systems that 4 Food and Drug Administration, Division of Anti- 19 Meier B, Saul S. Marketing of Vioxx: how Merck played Inflammatory, Analgesic, and Ophthalmic Drug protect the interests of patients. A renewed game of catch-up. New York Times 2005 Feb 11:A1. Products. HFD-550, medical officer review. Vioxx 20 Berenson A, Harris G, Meier B, Pollack A. Dangerous commitment by all those involved and the (rofecoxib), NDA 21-042/052. Washington, DC: FDA, data—retracing a medical trail. New York Times 2004 institution of these systems are the only way 1999. Nov 14:A1. 5 Food and Drug Administration. FDA advisory committee to extract something positive from this unfor- 21 IMS Health. National prescription audit plus time period briefing document NDA 21-042, s007: VIOXX tunate affair. 1999 to September 2004, extracted 2004. Plymouth gastrointestinal safety. Washington, DC: 2001. Meeting, PA: IMS Health, 2004. 6 Merck. Vioxx. In response to your questions: 22 Targum SL. Review of cardiovascular safety database, Harlan M Krumholz is Harold H Hines Jr professor of medicine cardiovascular system, clinical profile in osteoarthritis rofecoxib. NDA 21-042, S-007. Washington, DC: FDA, and epidemiology and public health, Department of Medicine, studies. Whitehouse Station NJ: Merck, 2000. www. 2001. www.fda.gov/ohrms/dockets/ac/01/briefing/ vioxxdocuments.com/Documents/Krumholz_Vioxx/ Yale University School of Medicine, 333 Cedar Street, PO Box 3677b2_06_cardio.pdf. Merck2000CVcard.pdf. References w1-w37 are on bmj.com

BMJ | 20 jANUARY 2007 | Volume 334 123 OBSERVATIONS

“This was made to be nurses were hired so that operating time in sensational television theatre was increased. and to find villains—the He finally called for a “million dollar consultants. I think Gerry man” to manage the whole system and Robinson was naive, but I for management consultants to be sacked. think also he eventually got “Management isn’t an amazing art. It’s day to day practical stuff.” the point” What doctors are saying On the whole, medics did not take kindly to chauffeur driven Robinson, clad in designer black and whiffing of the private sector. Few could deny that it made riveting TV, but at whose expense? One contributor to doctors.net noted the “collective despair” and “low morale” throughout doctors’ forums in the United Kingdom. Why, said another, did such a “cracking X E R /

I district general hospital”—it is a foundation N I C I trust with some of the lowest waiting lists in D I U G the country—come across on television as O C

S such a rudderless ship? E C N

A Another doctor posting on doctors.net R F said: “This was made to be sensational tel- evision and to find villains—the consultants. I THE WEEK IN MEDICINE think Gerry Robinson was naive, but I think also he eventually got the point. This only surfaces in the last 15 minutes of the third No quick fix for the NHS programme and in his subsequent interview on Newsnight.” When an industry hotshot was let loose on the NHS, The “point” was that even if the hospital doctors saw red became more efficient, primary care trusts will not fund the additional work, so why bother? Rotherham surgeons had eventu- What’s the story? “is just half-arsed.” He said, “In the NHS ally agreed to carry out Friday afternoon The internet talkboards were humming last everything is talk. No one thinks they are lists, but the local health economy could week—mainly with indignant doctors—after actually going to follow anything through.” not afford it. the BBC broadcast a series of films about the His tactic was to get staff members talking to An audience granted with the health secre- business guru Gerry Robinson’s attempts to each other, to devise initiatives on the shop tary, Patricia Hewitt, to discuss this issue was “fix” the NHS—or, more exactly, to cut the floor, rather than relying on senior manage- a missed opportunity, most doctors agreed. waiting lists of a local hospital. The former ment. But he remained frustrated by the NHS Blog Doctor (nhsblogdoc.blogspot. Granada boss was let loose on Rotherham painfully slow progress. In episode one, evi- com) quoted Mrs Hewitt’s line: “It is not a General Hospital with the task of getting dence mounted against the consultants, who case of more funds for the PCT [primary more patients through the system at no extra seemed to have a reflex reaction against any care trust], Gerry, for, you see, we have set cost. management led change. “Consultants are them free.” In Can Gerry Robinson Fix the NHS? we saw idiosyncratic. It’s part of the culture,” said The Blog Doctor goes on: “For one won- lingering shots of theatres lying empty on a the chief executive. derful moment we all thought he might Friday afternoon and an incredulous Robin- One surgeon further undermined his pro- strangle her. Sadly it was not to be. He com- son questioning this wisdom when waiting fession’s image by memorably complaining promised by calling her ‘slightly disingenu- lists mounted up. to Robinson, “Of the managers here, well, ous.’” “I thought they’d be packed,” said Rob- I can’t think of a single one with an MBA. On Newsnight Robinson was more forth- inson. “But it really wasn’t like that. The They’ve got three O levels or something, coming, saying that the politics should be theatres really weren’t being managed in and they’re trying to manage staff with five “pulled out” of the NHS, so that manag- any way that I would recognise as being or six degrees.” ers could get on with managing the system appropriate for an important and expensive Robinson succeeded in cutting paediat- rather than dealing with new initiatives from resource.” ric waiting lists from eight to two weeks, on high. He also dismissed Jeremy Paxman’s He stalked the corridors, talking to staff by cajoling each consultant to see an extra glib suggestion that consultants should be and to the uncomfortable looking chief patient per clinic. He adopted an idea from put “back in their box.” executive, Brian James. The exercise quickly a nurse consultant to pool patients and do No, said Robinson: “I honestly don’t think led him to the view that NHS management endoscopies on a taxi rank basis. More they are off playing golf. What I came across

124 BMJ | 20 JANUARY 2007 | VOLUME 334 OBSERVATIONS

was people wanting to do the right thing. I’ve never seen that sense of not being able to do WHAT’S ON BMJ.COM anything before, in any organisation.” The Trouble with Medical Journals On doctors.net one member was angry by Richard Smith: an alternative view at the programme’s failure to make the key Pritpal S Tamber point until the bitter end. “Robinson hinted at the problem but never dared to point it I am amazed that the BMJ has published including the pros and cons of peer review. out as the root of the problem (since it would such a negative review of a book that should This is usually not through any fault of their be pointing a finger at a central government be shaking medicine’s foundations (BMJ own, but the result of an institutional failing that is rather fond of him). So instead he 2007;334:45). Medical publishing has within medical publishing. Editors are many failings; the fact that this book is not often appointed without any clear process, spent time showing his shiny local initiatives compulsory reading for everyone may well be there is no formal training, only recently (which frankly is futile if reducing waiting yet another. has a code of conduct been devised (but lists creates a deficit). Dishonest TV.” Although starting positively, reviewer few journals have adopted it),4 and there is The Blog Doctor was sceptical about Rob- Stuart Derbyshire describes Smith’s no way to measure whether an appointee inson’s tactics for bringing down paediatric concerns as not “obviously supportable.” has done a “good” or “bad” job. I cannot waiting lists by adding extra patients to con- Smith, together with colleagues at JAMA, think of any other respected profession that sultants’ weekly clinic lists: “Not rocket sci- has probably done more than anyone to would allow such a flawed recruitment and ence, that one, Gerry, and, in the absence search for “evidence” for the true value in training process, especially when there is so the cornerstone of medical publishing: peer much at stake. of data, fraudulent. How busy were they review. Much of the research conducted Derbyshire then says that the “status of a before? If they were slacking, fine. They at the BMJ during his tenure showed that journal is a matter for the members of the might already have been working at full there is little or no objective value to the discipline that the journal supports,” but capacity. If they were, soon they will be process,1 yet journals and their editors those of us within the publishing industry overworked and stressed.” persist with—and advocate—peer review; will know that often the only entity that There were supportive comments: “Given their only defence is that “there’s nothing really matters is the owner, as is evidenced 5 6 most people’s perception of how consultants better,” even though few have tried to find by the constant firing of respected editors. behave in a hospital, the consultants were an alternative (to my mind there is a notable He also seems to be appalled at the idea exception, the system used by Biology of having “consumerist views” at the heart let off lightly,” said one. “I’ve been at sev- Direct2). of medical research—yet it would seem to eral meetings over the past few days aimed Derbyshire also talks of the “essentially me that being patient focused is key to the at changing the direction of ‘reform,’ all of positive fundamentals of the profession.” business of health care (and let us not be which benefited from [participants] having Although this may be true, one has to fooled into thinking it is anything other than seen his programme,” wrote another. remember that so much can rest on a a business). On the Guardian talkboard the former publication—a job, a promotion, a grant Finally, I am deeply saddened to see the editor of the BMJ, Richard Smith, found the approval, the licensing or the withdrawal BMJ waste this opportunity to raise such important issues. All publications cover programme to be another variant of real- of an intervention. It would be naive to believe that all medical publishing is their backs by saying that the opinions ity television, on a par with wrestling with informed by these positive fundamentals. expressed are those of the author and not crocodiles. “It was hard not to empathise Indeed, more and more articles describe necessarily of the journal, but in essence with Robinson, one time chief executive of how unscrupulous authors have “played the BMJ has chosen not to give these issues Granada; the cheery Irishman, a devotee of the system” and corrupted the scientific more of an airing. I’m afraid I see that as an ‘can do,’ seemed almost in danger of going record, bringing potential harm to patients awful failing. In her Editor’s Choice, Godlee under.” worldwide.3 Many editors have anecdotes confidently concludes: “Medicine and about the behaviour of authors, reviewers, communication are changing, and so too is Did the press like the programme? and readers whose “moral character” the BMJ.” Change is often good, but before has not been enhanced by their chosen any industry can devise where it wants Simon Calkin, in the Observer, was unim- profession. to go, it must understand where it is. This pressed with Robinson’s remedy for the To my mind, the book is brimming with book should be an important signpost in NHS. “It came suspiciously close to busi- important ideas, but Derbyshire describes medical communication’s journey, but it’s ness reality TV, and Robinson’s idea that these as “neither necessary nor helpful.” unlikely to be that if one of the world’s most one million-dollar supermanager can kick The one that he dismisses in detail is influential medical journals cannot see the the service into shape is dangerous and putting patients on to editorial boards: his wood for the trees. deluded—as with all calls for heroic leader- rationale is that this would put “the journal Pritpal S Tamber is managing director, into the hands of novices who ship, the counsel of someone who has run Medicine Reports Ltd have no stake in the out of substantive ideas.” Competing interests: PT is secretary of the World intellectual integrity Association of Medical Editors (WAME), a council Carol Midgley, in the Times, said that Rob- of the journal.” He member of the Committee on Publication Ethics inson did a good job but that he would run implies that editors (COPE), a friend of Richard Smith, a member of the a mile if he were asked to do it full time. “I are not novices, but BMA, and has been employed by the BMJ on two kept thinking of those Tory MPs who say it’s this is untrue. They occasions (the first of which was reporting to Richard may be experts in their Smith) and by Biology Direct, the publisher of the perfectly easy to live well on benefits and do “notable alternative” to peer review. so for a week, but so what? Try doing it in scientific or clinical fields, but few editors This article was posted as a rapid response to Stuart the long term with accumulated despair and have even a basic sickness of the soul.” Derbyshire’s review on 10 January 2007. The full understanding of response, with references, is at www.bmj.com/cgi/ Rebecca Coombes, journalist, London editorial or publishing matters, eletters/334/7583/45#153576 [email protected]

BMJ | 20 JANUARY 2007 | VOLUME 334 125 OBSERVATIONS

YANKEE DOODLING Douglas Kamerow Today’s doctor’s dilemma The secret to a longer life is nothing we can do anything about in health care

To George Bernard Shaw, the dilemma something about it are two different would the delivery of today’s statins doctors faced was choosing between things. and aspirin to all those who could helping patients and helping Ten years later, America is still benefit from them. themselves to lucrative fees. Today’s doing a rotten job of delivering Much of this is not rocket science. If doctor’s dilemma is that what we do clinical preventive services. The US there is a budget for it we can improve doesn’t make much difference. Agency for Healthcare Quality and healthcare quality, including the As a young man disillusioned by Research has just published its annual quality of preventive care, with known the politics of the late 1960s, I sought National Healthcare Quality Report. tools: evidence based guidelines, refuge in medicine. The government Their conclusion? “The use of proven reminder systems, computerised and politics were making things worse, prevention strategies lags significantly patient order entry systems, and and I thought that the only hope was behind other gains in health care.” electronic medical records. It would helping one patient at a time. I would Only half of adults over 50 get be nice if future years’ national quality work as a general practitioner, my screened for colorectal cancer. A third reports found great leaps of progress patients would feel better, and thus I of smokers don’t receive advice to Most authorities rather than what they admit are now would make the world a better place. quit. Less than 60% of elderly people “are now convinced only modest and uneven gains. This The answer was medical. have ever had pneumococcal vaccine. that education— must be the answer. I was wrong. After working in a Twenty per cent of children under three years in school— But it’s not. What is the single factor health centre in an economically have not received all recommended has the most direct that best predicts longevity? It is not deprived neighborhood only a short immunisations. causal effect smoking, diet, or receipt of appropriate time, even I figured out that I couldn’t One solution to this problem is to preventive care. It is not race or wealth. “fix” much of what was wrong with my pay doctors to do preventive care. on how long Iona Heath argued elegantly in this patients. Sure I could prescribe pills The current experience in the NHS is people live space two weeks ago that we should and insulin for my diabetic patients, instructive here. Shaw would have focus on treating the sick rather than but many of them had neither the loved learning that at least some of treating risk factors that we have means to buy healthy foods nor today’s British GPs are getting rich not turned into diseases, but that would access to well stocked shops from by doing unneeded surgeries but by have little or no effect on lifespan. which to purchase them. I could refer doing thousands of Pap smears and ” In fact, the secret to a longer life is patients to dietitians to learn about flu shots. The US Medicare program nothing we can do anything about in healthy eating, but many couldn’t says doctors will soon be able to health care. read the leaflets they were given. I increase their reimbursement if they That is today’s doctor’s dilemma. could immunise children against an deliver better care to the elderly, some It is not the conflict between what is increasing number of illnesses, but of which will no doubt be defined as good for our patients and what is good many fell victim to epidemics we didn’t more preventive care. for our pocketbook. It is not choosing have jabs to prevent: drugs, tobacco, But we have a long way to go. In between sickness care and violence. The answer was societal, not America we are much more prevention, or between medical care medical. interested in finding the next new and public health services. Most Twenty years later, J M McGinnis and blockbuster (or even “me-too”) drug authorities are now convinced that W H Foege published data showing than in what Steven Woolf, professor education—years in school—has that the leading causes of death in of family medicine at Virginia the most direct causal effect on the United States were not heart Commonwealth University, calls how long people live. We can work disease, cancer, and stroke. Medically, fidelity of existing interventions. We around the margins with our statins we knew what to do about them. spend billions of dollars inventing and and nicotine replacement patches Instead, tobacco, bad diet, physical testing new drugs that only marginally and mammograms. We can relieve inactivity, and alcohol were actually extend the benefits of those they suffering and tend to the sick. But for what was killing people. This was an replace, rather than putting resources every extra year spent in school, life unconventional conclusion, even into better delivery of existing effective expectancy is extended 18 months. among public health types. I certainly services. Woolf and his colleague Even bleeding heart liberals like me hadn’t spent much time learning about Robert Johnson have shown, for don’t think that it’s a doctor’s job to get any of these problems in medical example, that heroic searches for kids to stay in school. school or residency. The answer must better cholesterol-lowering and anti- Douglas Kamerow is former US assistant be preventive medicine. But knowing platelet drugs cost more and result surgeon general and the BMJ’s US editor what the problem is and doing in less population health gain than [email protected]

126 BMJ | 20 JANUARY 2007 | VOLUME 334 ANALYSIS

A graphical method for depicting randomised trials of complex interventions Making the what, when, and who of non-drug treatments easier to understand would benefit researchers and readers, say Rafael Perera and colleagues

Complex interventions consist of several separate Rafael Perera written description, will clarify the basic structure of the components combined to produce a desired outcome.1 senior research fellow in experimental intervention, and elucidate the differences Evaluation of such interventions in randomised trials statistics, Carl Heneghan, between treatment arms, as the examples below show. will generally lead to complex comparisons between deputy director, Centre for trial groups.2 Moreover, text descriptions in journal Evidence Based Medicine, Time Angina Educational session articles may obscure aspects of the interventions in the Department of Primary Health line plan (control) trial and hinder comparison between them. To counter Care, University of Oxford, Randomisation these problems we have produced a single image that Oxford OX3 7LF , reader in presents the components of all interventions in the trial Patricia Yudkin a f g medical statistics, Department and compares different treatment arms. The aim is to Baseline (time 0) clarify the structure of the contrasted interventions and of Primary Health Care, b c d thus aid interpretation of the trial results. University of Oxford Correspondence to: P Yudkin 1 week e The need for clear comparisons [email protected] We studied 169 randomised trials of non-drug inter- 4 weeks e ventions in primary care published between 1999 and 2003. We searched Medline, PSIQInfo, Bioabstracts, 8 weeks e and Embase using the free text search terms “ran- domised controlled trials” and “primary care” and their 12 weeks e synonyms, and excluding the term “placebo” appearing in the title or abstract; we also hand searched reference 6 months Measurement of outcomes lists of retrieved papers. In many of these papers the interventions were incompletely described. We identi- Questionnaire completed by patient to elicit whether he a or she has any of the common misconceptions about fied three principal problems: identifying the different angina components of the intervention, establishing the time 30-40 minute structured interview between patient (and at which components were delivered, and defining the partner if possible) and nurse. Nurse has extensive b experience of running primary care secondary prevention differences between intervention arms. clinics. Misconceptions discussed and corrected. To clarify these aspects we suggest that it would be Patient’s risk factors for coronary heart disease elicited and methods suggested to reduce them by introducing helpful to depict the experimental and control interven- lifestyle change tions graphically. The proposed graph is similar to a flowchart, with each treatment arm represented in a spe- The angina plan, a 70 page work book handed to patient c during interview. The plan uses cognitive behavioural cific column, and with all the intervention components techniques aimed at changing maladaptive coping practices and reversing disability (further details given) presented within that column. The time scale of the trial runs from top to bottom on the left hand side, with Audiotaped relaxation programme handed to patient d during interview. Patient asked to practice relaxation the times of randomisation and outcome measurement using the tape each day for 20 minutes (or measurements) clearly marked. Each component Phone call from nurse to patient (5-10 minutes). Success of an intervention is depicted separately. Components e with goals rewarded with praise and encouragement; delivered concurrently are displayed side by side, while patient invited to extend goals those delivered consecutively are shown one beneath Patient educational session with the same nurse (length the other. unspecified). Nurse identifies patient’s risk factors from research clinic measurements and personal history. We regard components either as objects or activi- f Nurse discusses with patient how risk factors could be ties. Objects are represented by squares (to reflect their reduced. Nurse responds to questions about each risk factor and about heart disease in general. Patient fixed nature) and activities by circles (to reflect their encouraged to discuss how it has affected his or her life

flexibility). Different components are labelled with Package of written information about coronary heart different letters. Below the diagram, a legend gives a g disease (from British Heart Foundation and other authoritative sources) handed to patient during the brief description of each component, including its form, educational session content, functions, and details of who delivers it. If nec- essary, additional material can be given in the text. For the full versions of these Fig 1 | Graphical depiction of interventions in a trial of self This approach will convey as much information as a articles see bmj.com management in patients with newly diagnosed angina

BMJ | 20 JANUARY 2007 | VOLUME 334 127 ANALYSIS

Time Community Usual general Time Manual Manual + Manual + Booklets line based practitioner line telephone nurse (control) specialist nurse care (control) Randomisation a a a a a Randomisation Baseline (time 0) b c b c d b c e f Baseline (time 0) b c d a a a During 2 years e f f 3 months from baseline b b d b e 2 years Measurement of outcomes

a a a 6 months a Course for nurses on meeting needs of people with Parkinson’s disease and their carers (referenced) b bd be

b Leased car given to specialist nurse 12 months Measurement of outcomes

c Mobile phone given to specialist nurse a Questionnaire to determine stage of change, processes of change, self efficacy and decisional balance of participant using the transtheoretical model

Description of areas of responsibility given to specialist “Expert system” personalised (6-8 sided) letter sent, based on participant’s responses to each nurse. These are counselling and educating patients and b questionnaire. This gave feedback on the stage of change, decisional balance, self efficacy, and carers about Parkinson’s disease; providing information the process of change, with second and third letters giving progress since the previous letter on drugs; monitoring clinical wellbeing and response to d treatment, and reporting to doctors where appropriate; Pro-change programme for a healthier lifestyle instigating respite or day hospital care where , a 64 page colour booklet enabling participants appropriate, seeing patients in hospital and liaising with c to stage themselves. Anglicised version of booklet used in American trials. Contains exercises hospital staff on patient discharge; assessing to help participants move from their current stage. Sent on return of questionnaire entitlement to state benefit; liaising with local primary care teams for ongoing care and treatment when d Telephone call made by trained lay person (three hours training) on return of each questionnaire; based on script and non-interactive Nurse works under the guidance of a nurse manager, assuming the stated areas of responsibility. Clinical e Appointment with nurse made on return of each questionnaire to discuss personalised letter, position of nurse is as adviser to general practitioner, e participant’s progress with the pro-change programme, and to encourage its implementation rather than clinically autonomous

Four standard items of self help material: Stopping made easier, a 24 page manual; The quit f Usual care from general practitioner (details not given) f guide to stopping smoking, a 12 page booklet; and 2 credit card sized reminder cards (Benefits of smoking cessation, Tips for staying quit) sent on return of questionnaire

Fig 2 | Graphical depiction of interventions in a trial of Fig 3 | Graphical depiction of interventions in a trial of an expert system and self help manual to community based nurses specialising in Parkinson’s disease aid smoking cessation

Intervention intensity and repeated components The experimental intervention consists of training Lewin et al reported a parallel arm trial of self man- nurses to specialise in Parkinson’s disease and, after agement in patients with newly diagnosed angina.3 clearly specifying their areas of responsibility, request- The experimental intervention introduces the “angina ing them to support patients for the two year trial. plan,” the objective of which is to allow patients to The diagram (fig 2) shows the resources needed for manage their condition using cognitive behavioural the community nurse intervention and also highlights techniques. The intervention is enhanced by nurse the possible variation in the interventions in both support in the form of interviews and telephone calls. experimental and control arms.In the experimental The original paper describes the intervention in about arm, the nurses are trained before randomisation; 590 words. after randomisation (baseline) the nurses are given a With the aid of the diagram (fig 1) we can easily car, a mobile phone, and a clear description of their recognise that the experimental arm has a much more areas of responsibility. Moreover, the timing of the intensive intervention than the control, with repeated intervention is not static; in both the experimental nurse telephone contacts at 1, 4, 8, and 12 weeks after and control arms patient care is given at any point the last contact in the control group. (An alternative (and potentially at several times) in the two years that way of depicting these repeated telephone contacts the trial lasts. would be to use a single circle, and to label the time line with the four times of delivery.) The diagram also Multiple arms for multiple comparisons clearly shows that no component is common to both Aveyard et al examined the effect on smoking cessa- interventions. tion of the pro-change course.5 The trial tested three experimental interventions using the pro-change Flexible interventions course with increasing levels of contact (none, tele- Jarman et al reported a parallel arm trial of commu- phone call from lay person, and appointment with nity based nurses specialising in Parkinson’s disease.4 nurse). The control group received standard support

128 BMJ | 20 JANUARY 2007 | VOLUME 334 ANALYSIS

material. The written description of the interventions By focusing attention SUMMARY POINTS had 715 words. Our diagram (fig 3) shows imme- on the components Complex interventions often require long explanations that diately the cumulative nature of the experimental of the intervention, it are difficult to follow interventions. prompts researchers Although each intervention is complex, the com- Graphical representation could clarify descriptions to think parison between successive interventions is relatively The graph would prompt researchers to focus on the simple, each differing from the last by a single compo- structure and timing and ensure appropriate comparisons nent. The year long interval between the control inter- Readers would be able to see the differences between vention and trial outcome also stands out, in contrast comparison groups immediately to the six month interval in the experimental arms.

Advantages of using graphs bers of participants at different stages of a study. Graphical depiction of an entire intervention allows We suggest that our proposed graphical method its structure to be quickly understood. With the would similarly increase the clarity of reporting of experimental and control interventions placed side by complex intervention trials. side on the diagram, differences between them—such as in the time elapsing between their delivery and the We thank Paul Glasziou for helpful comments. trial outcome—become obvious. Contributors and sources: RP and PY had the idea of a graphical depiction. RP created the graphical display, built the trial database, and We believe that the discipline of constructing a dia- drafted the article. CH helped in the creation of the graphical display gram will help at the design stage of a trial. By focus- and contributed to the writing. PY originated the investigation into ing attention on the components of the intervention, complex interventions, helped in the creation of the graphical display, and it prompts researchers to think through the structure, contributed to the writing. timing, and contents of each component in detail and Funding: CH is funded by a Department of Health Research Development Award. to describe the components adequately. Competing interests: None declared. The exercise should help to ensure that the con- trol intervention has been adequately considered 1 Medical Research Council. A framework for development and and described and that the difference between the evaluation of RCTs for complex interventions to improve health. London: MRC, 2000. experimental arm and the control arm is appropriate 2 Paterson C, Dieppe P. Characteristic and incidental (placebo) effects in for measuring the effect of the intervention. complex interventions such as acupuncture. BMJ 2005;330:1202-5. 3 Lewin RJ, Furze G, Robinson J, Griffith K, Wiseman S, Pye M, et al. A For the reader of the trial a graph will allow the randomised controlled trial of a self-management plan for patients details of an intervention to be quickly and easily with newly diagnosed angina. Br J Gen Pract 2002;52:194-6, 199-201. grasped. Aspects that may be missed in a long ver- 4 Jarman B, Hurwitz B, Cook A, Bajekal M, Lee A. Effects of community based nurses specialising in Parkinson’s disease on health outcome bal description stand out clearly, thus the differences and costs: randomised controlled trial. BMJ 2002;324:1072-5. between experimental and control interventions 5 Aveyard P, Griffin C, Lawrence T, Cheng KK. A controlled trial of an expert system and self-help manual intervention based on the stages become obvious. of change versus standard self-help materials in smoking cessation. The CONSORT trial flowchart has improved Addiction 2003;98:345-54. transparency and accurate reporting of the num- Accepted: 15 November 2006

The in-between world of knowledge brokering For research findings to effectively influence health services’ delivery of care needs an intermediary, says Jonathan Lomas

The ultimate aim of people engaged in health research Jonathan Lomas Disconnection between research and health services is to get the health service’s workforce, its employ- chief executive officer, Canadian worlds ers, and its suppliers to have knowledge of facts (as Health Services Research The old adage “form follows function” is poorly represented by research results) and to use these Foundation reflected in the production and use of health research. facts in their practices, policies, and products. How [email protected] The research world favours grant acquisition and well organised is research to achieve this aim? And academic publication over knowledge synthesis and how receptive and oriented are health services to engagement with the health service.2 Researcher this aim? The answers seem to be “not well organ- to researcher communication about the next study ised” and “not very receptive.” The interpersonal (“more research is needed”) is well organised and all connections needed to bridge this know-do gap are too common3 4; researcher to practitioner dialogue not yet in place.1 An emerging role therefore exists about implementing findings (“actionable messages”) 5 for knowledge brokers, supported by knowledge For the full versions of these is poorly organised and all too rare. brokering resources and agencies, to fill the gap. articles see bmj.com Structures and incentives in the health system do not

BMJ | 20 JANUARY 2007 | VOLUME 334 129 ANALYSIS

Box 1 | In the nirvana that is a research based health The mere knowledge over an event or manufacturing a product. In the case service. . . of a fact is pale; but of decision making, multiple interacting processes are Universities would reward when you come to used to build consensus around a course of action. • Inclusion of decision makers in research processes realize a fact, it takes In the case of research, accumulating sequential processes reveal the “facts of the matter” through often • Creation of centres that connect researchers directly with on color. It is all the health service clinicians, managers, and policy makers haphazard cycles of discovery and validation.6 When difference of hearing • Synthesis of interdisciplinary bodies of knowledge into these are disconnected processes, the facts tend to key actionable messages relevant to pressing service of a man being play second fiddle to the values that underpin con- questions stabbed to the heart, sensus. If they can be connected, however, the facts • Dissemination of brief, plain language research and seeing it done can actually help to create the consensus.8 Hence one summaries through face to face exchanges between the path to more research informed decision making is Mark Twain, A Connecticut doers and users of research to focus on better linkage and exchange between the The health service would reward Yankee in King Arthur’s Court, 1889 processes that create the facts (research) and the ones that • Active involvement of its clinicians, managers, and policy incorporate the values (decision making) (box 1).9 makers in research relevant to pressing health service Innvaer and colleagues’ systematic review of efforts questions to link research and policy better arrived at this same • Support for operational research and development on its own activities conclusion: “personal two-way communication • Change management driven by research based evidence between researchers and decision-makers should be • Inclusion of researchers in decision making processes used to facilitate the use of research. This can reduce mutual mistrust and promote a better understand- ing of policy-making by researchers and research by policy-makers.”10 fare much better. The governance, organisation, and delivery of services reward consensus more than use Research to action: knowledge brokering as a social of research; coordination with stakeholders generally solution trumps collaboration with researchers; and strategic This linkage and exchange model of connecting positioning triumphs over decision making informed research to action moves us away from the predomi- by research.6 Indeed, research is often seen as the nant view of evidence informed decision making as a opposite of action, not the antidote for ignorance. technical exercise that places products into events—the Exceptions to these generalisations exist—the rise implicit premise of, for instance, the clinical guidelines of research based guidance organisations such as the or performance indicators industries. Rather, it char- National Institute for Health and Clinical Excellence acterises the task of better informing decisions with in the United Kingdom, the development of data research as being as much social as technical. driven practice organisations such as the Veterans Gabbay and le May recently illustrated this in Administration Health System in the United States, their ethnographic “mindlines” study of how clini- or the creation of centralised knowledge transfer and cal guidelines were translated into practice through brokerage for the Scottish Executive’s health depart- social interaction and interpersonal networks in two ment.7 But the general picture is one of poorly con- general practice groups in England.11 Similarly, in nected worlds lacking knowledge of (and often respect their extensive systematic review of the innovation for) each other. The inner workings, implicit rules, diffusion literatures, Greenhalgh and colleagues con- cultures, and realities that dominate the day to day cluded that “knowledge depends for its circulation on lives of people working in the health system and those interpersonal networks, and will only diffuse if these doing research on that system remain, for the most social features are taken into account and barriers part, mysteries to people on the other side. overcome.”12 This is a lesson learnt long ago by the pharmaceutical industry, with its use of local opinion Research and decision making as processes, not leaders to influence patterns of drug prescribing. products and events This social focus points to human interaction as the Fundamental to this disconnection is a misapprehen- engine that drives research into practice. It implies sion by each side of what the other is doing. Research- the need for both human intermediaries between the ers tend to see decision making as an event—they deliver their edicts to the impenetrable cardinals’ Box 2 | Attributes and skills of a knowledge broker7 13 retreat and await the puff of smoke that signals “deci- sion,” while grumbling about irrationality within the • Entrepreneurial (networking, problem solving, innovating) conclave. Decision makers—the patients, the care pro- • Trusted and credible viders, the managers, and the policy makers—tend to • Clear communicator see research as a product they can purchase from the • Understands the cultures of both the research and decision making environments local knowledge store, but too often it is the wrong • Able to find and assess relevant research in a variety of size, needs some assembly, is on back order, and formats comes from last year’s fashion line. • Facilitates, mediates, and negotiates Neither side seems to recognise that the other is • Understands the principles of adult learning managing a complex process rather than presiding

130 BMJ | 20 JANUARY 2007 | VOLUME 334 ANALYSIS

Box 3 | Illustrative activities of a knowledge brokering Foundation has adopted a role as a knowledge brok- agency: the Canadian Health Services Research ering agency for the past 10 years. We have defined Foundation knowledge brokering as “all the activity that links deci- Setting the research agenda sion makers with researchers, facilitating their interac- • Triennial consultations with the health service on priority tion so that they are able to better understand each themes: Listening for Direction17 other’s goals and professional cultures, influence each • Research funding restricted to priority themes other’s work, forge new partnerships, and promote the • Each research project required to have 50% funding from use of research-based evidence in decision-making.”13 an organisation in the health service Box 3 lists some of the approaches we have used to Facilitating applied research link the people leading research processes (mostly in • Masters, PhD, and postdoctoral researcher training universities and granting councils) and those lead- awards all require a placement in the health service ing decision processes (health service managers and • Each research project required to include decision makers policy makers). in the health service as co-investigators Adopting a knowledge brokering role has both • Decision support research syntheses are co-produced by philosophical and practical dimensions. The philo- researchers and people who can implement the results sophy leads us to build into all our activities and Disseminating research programmes the expectation of ongoing linkage and • Production of plain language research summaries exchange between the researchers and their decision on pressing service questions: Evidence Boost and Mythbusters making counterparts. Capacity development—for • Support of virtual knowledge networks of researchers and researchers to be able to do applied research and decision makers in the priority theme areas decision makers to be able to use it—is part of the • Organising regular face to face exchanges on questions/ philosophy. The practice of knowledge brokering in problems and research results in priority theme areas itself leads us to support knowledge brokers, both with Getting research used employment and with tools and resources such as syn- • Funding and evaluating selected knowledge brokers; thesis of research, plain language research summa- providing support and resources to other brokers ries, networks and exchange events bringing together • Providing regional workshops to the health service on researchers and managers, self assessment checklists tools and techniques for research use for organisational capacity to use research, and other • Elite fellowships for decision makers in research use: “knowledge transfer and exchange” mechanisms. executive training for research application (EXTRA) We recently surveyed the network of more than 400 (See www.chsrf.ca for further details) Canadian health system knowledge brokers we have supported since 2003, only a few of whom have full time designation for this role.13 They report, as have worlds of research and action (knowledge brokers; others,18 that the supporting resources and tools are box 2) and supporting infrastructure (knowledge central to their role as brokers. They spend about 30% brokering agencies and resources). of their time on knowledge transformation (reading Knowledge brokering is not a new concept. For and disseminating research) and 20% on intermedia- instance, in the late 1800s the German dominance of tion (actually linking researchers and decision makers). the synthetic dye industry was explained by “an infor- The remaining time, spent doing management duties mal network of ties that connected players in industry or teaching, reflects the fact that this is often a part and academia . . . the academic-industrial knowledge time role. About 30% of knowledge brokers are based network.”14 In 1906 the University of Wisconsin cre- in universities, about 10% in foundations or research ated its extension division to support agricultural funding agencies, and the remaining 60% in differ- liaison officers linking local farmers and university ent levels of the health system (Gold I et al, National researchers, as they still do today.15 symposium on knowledge transfer and exchange, More than 20 years ago technology transfer officers Toronto, 2006). were created in universities to speed research discov- eries into patents and production, and organisational behaviourists were calling for “the development of SUMMARY POINTS hybrid researcher-practitioner roles (rather than the Neither universities nor the health service provide much reliance on external ‘scientists’) . . . [and] mechanisms incentive for ongoing connections between researchers and to promote active boundary spanning, dialogue and clinicians, managers, or policy makers 16 joint learning.” Thus were born “clinical epidemi- More formal recognition is needed for the interpersonal ologists,” clinicians who both see patients and do role of knowledge brokering in connecting the research and research, although their hybrid counterparts in the decision making processes governance or management of the health service are Knowledge brokering uses a portfolio of resources to yet to evolve. make health services research and decision making more accessible to each other Knowledge brokering in Canada Initial experience in Canada suggests that adopting a With a budget of approximately $C16m (£7m; €11m, knowledge brokering approach improves the culture for evidence informed decision making $14m) a year, the Canadian Health Services Research

BMJ | 20 JANUARY 2007 | VOLUME 334 131 ANALYSIS

ADDITIONAL RESOURCES The general picture With apologies to M G Vassanji (and Vikram Lall) for the title of this paper. Lomas J. Improving research dissemination and uptake is one of poorly Contributors: The concepts behind this article are based on JL’s 25 years’ in the health sector: beyond the sound of one hand experience as a knowledge broker: initially as an academic leading an connected worlds applied health services research unit at McMaster University in Hamilton, clapping. 1997 (www.chsrf.ca/knowledge_transfer/pdf/ lacking knowledge ON, Canada (15 years), and then as the inaugural chief executive officer of handclapping_e.pdf) a national knowledge brokering agency in Canada (10 years). JL developed Oldham G, McLean R. Approaches to knowledge-brokering. of (and respect for) the specific elements of the article for an invited presentation at the World 1997. (www.iisd.org/pdf/2001/networks_knowledge_ each other Ministerial Summit on Health Research in Mexico City on 18 November brokering.pdf) 2004. He further refined these elements on the basis of collegial feedback Max Lock Centre, School of the Built Environment, after subsequent presentations and finalised them for this version at the request of the BMJ’s editorial staff. Westminster University, London. Intermediaries in knowledge transfer and exchange. 1998. (www.wmin. Competing interests: JL is employed by the organisation featured in ac.uk/builtenv/maxlock/KTweb/KT_Guide4.htm) the final section of this paper, the Canadian Health Services Research Foundation. Scottish Executive, Office of the Chief Researcher. New directions for knowledge transfer and knowledge brokerage 1 World Health Organization. World report on knowledge for better in Scotland. 2005. (www.chsrf.ca/brokering/pdf/scotland- health. Geneva: WHO, 2004 (www.who.int/rpc/wr2004). KT-KB-keyfindings_e.pdf) 2 Tomlinson S. The research assessment exercise and medical Australian Government Land and Water Australia. research. BMJ 2000;320:636-9. Knowledge for regional NRM: connecting researchers 3 Haynes RB. Loose connections between peer-reviewed clinical journals and clinical practice. Ann Intern Med 1990;113:724-8. & practitioners. 2006. (www.lwa.gov.au/downloads/ 4 David AS. Wanted—more answers than questions. BMJ publications_pdf/PB051023.pdf) 2001;323:1462-3. Canadian Health Services Research Foundation. Knowledge 5 Lavis JN, Robertson D, Woodside JM, McLeod CB, Abelson J, brokering. 2006. (www.chsrf.ca/brokering/index_e.php) Knowledge Transfer Study Group. How can research organizations more effectively transfer research knowledge to decision makers? Milbank Q 2003;81:221-48. 6 Walshe K, Rundall TG. Evidence-based management: from theory to The effect of our capacity development and practice in health care. Milbank Q 2001;79:429-57. New directions for knowledge transfer and knowledge resources for knowledge brokering are demonstra- 7 Clark G, Kelly E. brokerage in Scotland. Edinburgh: Scottish Executive, Office of ble, although more cultural than instrumental in their the Chief Researcher, 2005 (www.scotland.gov.uk/Resource/ impact. Brokering research priorities with people Doc/69582/0018002.pdf). 8 Hutchings A, Raine R, Sanderson S, Black N. A comparison of formal working in the health system attracts the attention, consensus methods used for developing clinical guidelines. J Health resources, and engagement of these decision makers Serv Res Policy 2006;11:218-24. to the resulting research agenda.17 Research funded 9 Lomas J. Using ‘linkage and exchange’ to move research into policy at a Canadian foundation. Health Aff 2000;19:236-40. under the model is four times more likely than that 10 Innvaer S, Vist G, Trommald M, Oxman A. Health policy-makers’ funded by traditional means to be subject to active perceptions of their use of evidence: a systematic review. J Health Serv Res Policy 2002;7:239-44. efforts at dissemination and implementation (Graham 11 Gabbay J, le May A. Evidence based guidelines or collectively ID et al, Translating research into practice: advancing constructed “mindlines?” Ethnographic study of knowledge excellence from discovery to delivery, Washington, management in primary care. BMJ 2004;329:1013-7. 12 Greenhalgh T, Robert G, Macfarlane F, Bate P, Kyriakidou O. Diffusion 2004). Graduates from our researcher training pro- of innovations in service organizations: systematic review and grammes are just as likely to take up research careers recommendations. Milbank Q 2004;82:581-629. in the health service as in a university. Our brief, 13 Canadian Health Services Research Foundation. The theory and practice of knowledge brokering in Canada’s health system. Ottawa: plain language summaries of research—Mythbusters Canadian Health Services Research Foundation, 2003 (www.chsrf. or Evidence Boost—or our decision support syntheses ca/brokering/pdf/Theory_and_Practice_e.pdf). 14 Murmann JP. Knowledge and competitive advantage. Cambridge: of research are routinely used by brokers in govern- Cambridge University Press, 2003. ments, health authorities, and the health professions 15 Ward D. Serving the state: the Wisconsin idea revisited. Educational to generate dialogue and debate. Record 1992;73(Spring):12-6. 16 Ferlie E, Fitzgerald L, Wood M. Getting evidence into clinical practice: Developing capacity on the use of research for an organizational behaviour perspective. J Health Serv Res Policy those working in the health system also has results. 2000;5:96-102. For instance, only 21% of the health system managers 17 Lomas J, Fulop N, Gagnon D, Allen P. On being a good listener: setting priorities for applied health services research. Milbank Q entering our executive training for research applica- 2003;81:363-88. tion programme report using research in their day to 18 Van Kammen J, de Savigny D, Sewankambo N. Using knowledge brokering to promote evidence-based policy-making: the need for day work “most or all of the time;” two years later, at support structures. Bull World Health Organ 2006;84:608-12. graduation, this proportion has more than doubled to Accepted: 9 November 2006 50%. Sixty five per cent of these graduates also report an excellent or very good ability to create a more evidence based working environment in their home organisations; for those entering the programme the figure is only 8%. Endpiece Knowledge brokering is not a universal panacea. A historical view of the future However, the interpersonal linkages it creates are cer- tainly very promising as one of the “in-between” miss- There is virtually no limit to the amount of health care an individual is capable of absorbing. ing pieces that can bridge the know-do gap for health J Enoch Powell, British minister of health 1960-1963 services. Perhaps for the new year every health serv- Submitted by Ruth Reed, senior house officer in psychiatry, ices researcher should adopt a health services decision Enfield, Middlesex maker, and vice versa.

132 BMJ | 20 JANUARY 2007 | VOLUME 334 Cite this article as: BMJ, doi:10.1136/bmj.39014.503692.55 (published 21 November 2006)

Research BMJ

Impact of a theoretically based sex education programme (SHARE) delivered by teachers on NHS registered conceptions and terminations: final results of cluster randomised trial M Henderson, D Wight, G M Raab, C Abraham, A Parkes, S Scott, G Hart

Abstract potential for presentational bias.12 We are aware of only one pub- lished randomised trial of school sex education that uses Objective To assess the impact of a theoretically based sex anything other than self reported data to evaluate effectiveness. education programme (SHARE) delivered by teachers In a programme recommending abstinence at a girls’ high compared with conventional education in terms of conceptions school in Chile, researchers used clinical data on pregnancies to and terminations registered by the NHS. show that the intervention reduced conceptions over four years.13 Design Follow-up of cluster randomised trial 4.5 years after While these results are encouraging, they assume that any preg- intervention. nancies that were terminated (which was illegal) were balanced Setting NHS records of women who had attended 25 across the groups in the trial. secondary schools in east Scotland. Between 1993 and 1996 a sex education programme Participants 4196 women (99.5% of those eligible). delivered by teachers (SHARE) was developed for 13-15 year Intervention SHARE programme (intervention group) v olds in Scotland; this was evaluated between 1996 and 1999 in a existing sex education (control group). cluster randomised trial. Interim outcomes at six months after Main outcome measure NHS recorded conceptions and the intervention (average age 16 years 1 month) showed that, terminations for the achieved sample linked at age 20. compared with those receiving conventional sex education, Results In an “intention to treat” analysis there were no SHARE improved knowledge and the quality of sexual relation- significant differences between the groups in registered ships but had no impact on reported sexual or contraceptive conceptions per 1000 pupils (300 SHARE v 274 control; behaviour.14 15 At this age, however, only a third of the sample − difference 26, 95% confidence interval 33 to 86) and reported having sexual intercourse and follow-up data were − terminations per 1000 pupils (127 v 112; difference 15, 13 to obtained from only 70% of the original sample. 42) between ages 16 and 20. We report on the impact of the intervention on conceptions Conclusions This specially designed sex education programme and terminations by age 20 (4.5 years after the intervention) as did not reduce conceptions or terminations by age 20 recorded by the NHS. By linking to NHS data for the whole compared with conventional provision. The lack of effect was cohort of young women we have outcomes that are not subject not due to quality of delivery. Enhancing teacher led school sex to reporting biases and much less affected by sample attrition education beyond conventional provision in eastern Scotland is than outcomes from self reported data. unlikely to reduce terminations in teenagers. Trial registration ISRCTN48719575. Methods The intervention programme Introduction The SHARE (sexual health and relationships) programme was Worldwide, young people are at risk of unwanted pregnancies,12 developed and piloted in Scotland over three years in consulta- sexually transmitted diseases, and unsatisfactory3 or coerced4 tion with teachers, sex education specialists, and education and early sexual relationships. Sexual health services help to improve health promotion departments.16 It is a five day training sexual health in young people,56 but school sex education is programme for teachers plus a 20 session pack: 10 sessions in regarded as the most effective way of targeting this group.78 the third year of secondary school (at ages 13-14) and 10 in the Several overviews have concluded that sex education can fourth year (at ages 14-15).17 It is intended to reduce unwanted have beneficial effects on sexual behaviour,6910although the ear- pregnancies, reduce unsafe sex, and improve the quality of lier reviews relied almost entirely on quasi-experimental studies sexual relationships. The total cost of training each teacher was rather than randomised trials. Generally, the more rigorous about £900 (€1343, $1684), including a copy of the package. evaluations of school sex education have been less likely to find The sociopsychological and sociological theoretical basis of positive outcomes.11 The only review restricted to randomised the programme has been set out previously.18 The programme trials concluded that primary prevention strategies did not delay combines active learning (such as small group work and games), sexual intercourse, improve use of contraceptives, or reduce information leaflets on sexual health, and the development of pregnancies.12 Furthermore, experimental evaluations often have considerable attrition from the intended target group or Further details on the final outcomes of the SHARE programme can be rely on self reported behavioural outcomes, or both, which has found bmj.com.

BMJ Online First bmj.com page1of5

Copyright 2006 BMJ Publishing Group Ltd

Research skills to negotiate sexual encounters, handle condoms, and We calculated the sample size to have 80% power to detect a access services, primarily through the use of interactive video but 33% decrease in the cumulative termination rate (identified as also role playing.7 It has all 10 characteristics Kirby identified as the main outcome) from after the delivery of the programme in necessary for effective programmes.9 The exercises were intervention schools to age 20 with a two sided t test. A design developed specifically for the programme or modified from effect of 1.5 was assumed. other packs.17 Follow-up data and statistical analysis Conventional education The information services division (ISD) of the NHS in Scotland In the 12 control schools sex education for third and fourth years holds data on all births, stillbirths, miscarriages, and terminations varied from seven to 12 lessons in total, primarily devoted to in Scotland.2 We submitted records of surname, forename, date provision of information and discussion. Only two control of birth, and postcode(s) for the women in the trial (excluding schools routinely demonstrated how to handle condoms, and withdrawals) for linkage to the NHS data. The division provided none systematically developed negotiation skills for sexual the results of the linkage to researchers in aggregate form, as per encounters. The cost of conventional education varied, with indi- the requirements of their privacy advisory committee. The data vidual packages starting from about £20. Few teachers had more were aggregated by school, cohort, status as early or late school than one day’s training, which would have cost about £180 a day, leaver (leaving school at the minimum age legally allowed (16 and some had received none or only a few hours’ training. years) or staying on later), and parents’ socioeconomic status Recruitment and randomisation of schools (manual v non-manual workers) only for those who stayed on at We invited all non-Roman Catholic state schools within 15 miles school. Confidentiality constraints did not allow further of the main cities in Tayside and Lothian regions to participate disaggregation. These selections were informed by our previous (figure). Roman Catholic schools were not included as the feasi- work, which had shown them to be powerful predictors of self bility study showed that they would not agree to deliver the pro- reported sexual behaviour.20 The division also provided gramme as it takes a harm reduction approach, encouraging aggregated data on the ages at conception for all linked events. those who have sex to use condoms. We submitted records for 4196 women (2109 from the first Incentives offered were the full cost of the teacher training, cohort and 2087 from the second) to the division, and 922 (22%) including supply cover, or, for schools allocated to the control linked to one or more conceptions with an estimated date arm, the equivalent (£2000-£2500) to spend on personal and between the pupils’ 15th and 20th birthdays. By their 15th birth- social education (PSE) but not on sex education. A balanced ran- day, most pupils had received the first year of the programme domisation took into account socioeconomic characteristics of and at least a proportion had received the second year. If we had the school populations, the proportion of pupils staying at chosen age 16 as the start date (with complete delivery of the school beyond the age of 16, school size, and local sexual health programme) then some of the conceptions and terminations services, among other factors.19 One of these was the quality of after the programme would have been missed. The ideal cut-off school sex education before the trial, a measure that included age would have been 15~5 years but that was not possible levels of teacher training. because of the linkage requirements. The results were the same

Invited to participate (47 schools)

Excluded: Refused to participate (22 schools), mainly beacause of envisaged practical difficulties in implementing programme

Randomised (25 schools)

Allocated to intervention (13 schools) Allocated to control (12 schools) Received intervention: Acted as controls: 13 schools 12 schools 2080 female participants 2135 female participants

Lost to follow-up: Lost to follow-up: 0 schools 0 schools 9 participants withdrawn by parents 10 participants withdrawn by parents

Clusters: Clusters: Analysed (13 schools) Analysed (12 schools) Participants: Participants: 2071 (99.6%) participants analysed 2125 (99.5%) participants analysed

Flow of clusters and female participants through the trial

page2of5 BMJ Online First bmj.com Research

Table 1 Characteristics of schools and female pupils according to allocation Table 2 Rates of terminations and conceptions in teenagers per 1000 pupils to teacher delivered sex education (SHARE) or existing sex education and rates of any termination and conception per 1000 pupils* according to (control). Figures are numbers (percentages) of pupils unless stated allocation to teacher delivered sex education (SHARE) or existing sex otherwise education (control).† Figures are differences between SHARE minus control with 95% confidence intervals and P values‡ SHARE Control No of schools 13 12 Any Any Mean (range) socioeconomic factor 0.1 (−2.9-4.8) −0.2 (−2.8-2.8) Termination Conception termination/ conception/1000 rate/1000 rate/1000 1000 women women Mean (range) No of women/school 163.5 (57-259) 172.6 (112-259) SHARE schools 126.6 300.2 108.9 222.6 No of pupils 2125 2071 (n=2125) No in cohort 1 1069 (50) 1018 (49) Control schools 112.0 273.8 104.3 216.8 No in cohort 2 1056 (50) 1053 (51) (n=2071) School leaver status and social class: Unadjusted 14.6 26.4 4~5 5.7 Early leavers 531 (25) 546 (26) (−13.1 to 42.2), (−33.3 to 86.2), (−18.0 to 27.0), (−34.2 to 45.0), P=0.32 P=0.40 P=0.71 P=0.87 Late leavers, parents with manual 855 (40) 776 (38) occupations Adjusted for 15.3 30.4 5~1 15.3 school (−11.5 to 42.2), (−12.8 to 73.6), (−17.1 to 27.3), (−20.5 to 37.2), Late leavers, parents with non-manual 739 (36) 749 (36) socioeconomic P=0.28 P=0.18 P=0.67 P=0.57 occupations measure Adjusted for 15.7 31.9 5~6 9.7 school (−10.7 to 42.1), (−16.1 to 79.9), (−16.0 to 27.2), (−21.8 to 41.2), when we used an age of 16 as the start date (data not shown). A socioeconomic P=0.26 P=0.22 P=0.67 P=0.54 few events may have been missed when conceptions or termina- measure and leaver/social tions were not registered in Scotland, though this should affect class measure both arms of the trial equally. *Includes live births, stillbirths, therapeutic terminations, and miscarriages. To be able to assess the impact of deprivation at school level †Some pupils have ≥1 termination/conception. “Termination rate per 1000” and “conception and to adjust for it in the analysis, we developed a school level rate per 1000” show rates for terminations and conceptions even when some of these events belong to same woman. “Any termination per 1000 women” and “any conceptions per 1000 socioeconomic measure based on several of the school level woman” show rates for “any” women experiencing these events. indicators used to balance the randomisation at baseline. We ‡Restricted randomisation test, two sided. used factor analysis to weight the Carstairs’ deprivation category21 and the unemployment rate for the schools’ related to socioeconomic factors, whereas the relation between catchment areas, staying on rates to S5 (at age 15-16) and S6 (at socioeconomic factors and terminations was somewhat weaker. age 16-17), percentage of school leavers unemployed or on gov- Table 2 shows that SHARE pupils had slightly higher rates of ernment training schemes, percentage of pupils claiming free conceptions and terminations than the control pupils, but differ- school meals, and rates of school attendance. Each indicator ences were not significant. The model based analyses for the contributed in approximately equal proportions to the final binary outcomes confirmed the lack of evidence of differences measure, which was scaled to a mean of zero and a standard by treatment arm. Approximate intracluster correlations23 were deviation of one, with high scores indicating a more deprived 0.04 for any conception and 0.005 for any abortion, correspond- school. ing to design effects of 8 and 2, respectively. After adjustment for Rates of conceptions and terminations were calculated from covariates the correlations were reduced to 0.003 and 0.002, the events identified at linkage. We used a restricted randomisa- both yielding design effects under 1.5. tion test22 that allows for the balance imposed by the design to compare the rates between the programme and control arms of Discussion the trial. Analyses were adjusted for the individual factors (school leaver status and social class) and the socioeconomic factor at This rigorous evaluation of a sex education programme school level. Further model based analyses of the two binary out- delivered by teachers did not find any benefit on rates for comes (any conception and any abortion) were carried out with conception or termination compared with normal sex the SAS GLIMMIX procedure to compute design effects and education. A balanced randomisation ensured optimal matching intracluster correlations. of control and intervention arms, and linkage to NHS data on conceptions and terminations ensured no reporting bias and Results only minimal attrition from the original eligible sample. This is in contrast to the 60% attrition in the sample when followed up We have previously shown that the schools were well balanced by postal questionnaire to age 20. In the current study, there may on characteristics of individual pupils at baseline.14 At age 14, the have been a small level of attrition across both arms because of sample was representative of young people in Scotland in terms women attending private health care (less than 2% of of socioeconomic status and family structure according to the terminations), moving from Scotland during the study period 1991 census.20 Table 1 shows characteristics of schools and pupils (1% average annual migration out of Scotland), or having their for women only. terminations in England or Wales (2.7% of all the terminations The overall observed rates of 287 per 1000 for conceptions performed on Scottish residents).24–26 On balance, the compari- and 119 per 1000 for terminations compare with figures for all son between this study and national rates suggests that the link- Scotland of 319 per 1000 for conceptions and 127 per 1000 for age was broadly effective. terminations for the same age group over the same time span. We followed up all the girls in the sample to the age of 20 (4.5 There were large differences in rates of conceptions and ter- years after the intervention) and found no significant differences minations between the schools. The termination rate did not fol- between the two groups in levels of conceptions and low the same pattern as the conception rate: the proportion of terminations. Though by the age of 20 some of the conceptions conceptions terminated was higher for those schools with low may have been planned, to have adequate statistical power we overall conception rates. The conception rate was strongly had to include conceptions until each girl’s 20th birthday.

BMJ Online First bmj.com page3of5 Research

Furthermore, there is no reason to assume that planned NHS Scotland for providing data and support; Douglas Kirby for advice; pregnancies would not be balanced across both arms of the trial. our advisory committee and colleagues in the MRC Social and Public Health Sciences Unit and Applied Statistics Group, Napier University, for Further analysis (not presented) indicated that these results can- advice and encouragement; and Sally Macintyre for helping with the study not be explained by differential quality of delivery of the design and support. programme. The observed lack of impact on the primary sexual Contributors: DW (principal investigator) and SS originated the study. DW, outcomes is in keeping with results of several other rigorous GMR, SS, CA, and GH designed the original study, while MH, GMR and evaluations of school sex education12 but not all.27 There were DW collaborated with the NHS Scotland to access the data, which was ana- large differences in rates of conceptions and terminations lysed by GMR and MH and commented on by DW, CA, and AP. MH, GMR, and DW drafted the paper and MH revised subsequent drafts based on between schools. The strong relation between conceptions and coauthors’ comments. GMR, CA, AP, GH, DW, and SS commented on sub- social deprivation, and the inverse relation between rate of con- sequent drafts of the paper and agreed on the final version. MH, GMR, and ceptions and proportion of those conceptions that were DW are guarantors. terminated, mirror findings for the United Kingdom as a whole.2 Funding: UK Medical Research Council and the Health Education Board Overall, the findings suggest that an enhanced programme, for Scotland. which included the 10 characteristics previously suggested to be Competing interests: None declared. important,9 was no more effective than conventional provision in Ethical approval: Glasgow University’s ethical committee for non-clinical research involving human subjects. NHS Scotland’s privacy advisory overcoming the strong socioeconomic and cultural influences committee. that shape rates of conception and termination. As with most tri- als of school sex education in developed countries, we compared 1 Singh S, Darroch JE. Adolescent pregnancy and childbearing: levels and trends in this programme with conventional sex education rather than industrialized countries. Fam Plann Perspect 2000;32:14-23. 2 Social Exclusion Unit. Teenage pregnancy. London: Stationery Office, 1999 (4342) nothing at all, and conventional sex education may have an www.socialexclusion.gov.uk/downloaddoc.asp?id = 69. important impact on limiting unwanted conceptions. 3 Wight D, Henderson M, Raab G, Abraham C, Buston K, Scott S, et al. Extent of regret- ted sexual intercourse among young teenagers in Scotland: a cross sectional survey. We have previously published interim outcomes from this BMJ 2000;320:1243-4. study, based on self reported data from both boys and girls at 4 Watts C, Zimmermann C. Violence against women: global scope and magnitude. Lan- median age 16 years 1 month (six months after the intervention), cet 2002;359:1232-7. 5 Diamond I, Clements S, Stone N, Ingham R. Spatial variation in teenage conceptions when only a third of the sample had reported having had sexual in south and west England. JRStatSocSerAStatSoc1999;162:273-89. intercourse.14 At this age, pupils and teachers preferred the 6 NHS Centre for Reviews and Dissemination. Preventing and reducing the adverse effects of unintended teenage pregnancies. York: NHS Centre for Reviews and Dissemination, 1997. SHARE programme compared with conventional sex education. 7 Abraham C, Wight D, Scott S. Developing the SHARE sex education programme: from It also increased pupils’ practical knowledge of sexual health, theory to classroom implementation. In: Rutter D, Quine L, eds. Changing health behav- iour:research and practice with social cognition models. Maidenhead: Open University Press, reduced regret of first sexual intercourse with most recent part- 2001. ner (all significant), and had small but significant beneficial 8 Gallant M, Maticka-Tyndale E. School-based HIV prevention programmes for African youth. Soc Sci Med 2004;58:1337-51. effects on beliefs about alternatives to sexual intercourse and 9 Kirby D. Effective approaches to reducing adolescent unprotected sex, pregnancy and intentions to resist unwanted sexual activities and to discuss con- childbearing. JSexRes2002;39:51-7. 14 15 10 Robin L, Dittus P, Whitaker D, Crosby R, Ethier K, Mezoff J, et al. Behavioral interven- doms with partners. These results contributed to the Scottish tions to reduce incidence of HIV, STD, and pregnancy among adolescents: a decade in Executive’s decision to continue to use the SHARE programme review. J Adolesc Health 2004;34:3-26. 28 11 Guyatt GH, DiCenso A, Farewell V, Willan A, Griffith L. Randomized trials versus in Scottish schools. The programme did not, however, affect observational studies in adolescent pregnancy prevention. J Clin Epidemiol reported sexual experience or use of contraception.14 2000;53:167-74. 12 DiCenso A, Guyatt G, Willan A, Griffith L. Interventions to reduce unintended A cluster randomised trial of pupil led sex education pregnancies among adolescents: systematic review of randomised controlled trials. BMJ compared with conventional teacher led sex education in 2002;324:1426-30. England (RIPPLE study) was conducted over a similar time 13 Cabezon C, Vigil P, Rojas L, Leiva ME, Riquelme R, Aranda W, et al. Adolescent preg- nancy prevention: an abstinence-centered randomized controlled intervention in a period to our study and collected similar, but only self reported, Chilean public high school. J Adolesc Health 2005;36:64-9. outcomes. The RIPPLE study found that by the age of 16, signifi- 14 Wight D, Raab G, Henderson M, Abraham C, Buston K, Hart G, et al. The limits of teacher-delivered sex education: interim behavioural outcomes from a randomised cantly fewer girls in the peer led group reported having trial. BMJ 2002;324:1430-3. intercourse compared with the conventional education group (35% v 41%), but proportions were similar for boys (33% v 31%). The proportions of pupils reporting unprotected first sexual What is already known on this topic 29 intercourse did not differ for girls or boys. While the results of Overviews have suggested that sex education delivered by the RIPPLE study were broadly positive, they are all based on self teachers can delay age at first sexual intercourse and reduce reports and the size of the effects on behaviour were modest. unsafe or unprotected intercourse The potential for whole class sex education delivered by teachers to influence young people’s behaviour might have More rigorous evaluations, however, generally have less already been reached by conventional provision. To have a positive outcomes and, to date, few evaluations have used stronger impact on the sexual health outcomes for young objective measures of sexual risk taking people, complementary interventions should be considered. The social patterning of conceptions and terminations, evident from What this study adds the variations between schools in this study, suggests that Enhanced sex education (SHARE) improved knowledge effective programmes have to address fundamental socioeco- and reduced regret but did not reduce conceptions or nomic divisions in society, while the influence of parenting terminations compared with conventional sex education factors on sexual experience30 31 points to strategies involving parents. To date, the most promising programmes have greater High quality sex education should be continued, but to scope and duration than school sex education and aim to change reduce unwanted pregnancies complementary, longer term 932 future life opportunities for young people. interventions that address socioeconomic inequalities and the influence of parents should be developed and We thank the young people and teachers involved for their cooperation and support, without which the study would not have been possible; Rod Muir, rigorously evaluated Alan Finlayson, and David Clark of the information and statistics division of page4of5 BMJ Online First bmj.com Research

15 Abraham C, Henderson M, Der G. Cognitive impact of a research-based school sex 30 Wight D, Williamson L, Henderson M. Parental influences on young people’s sexual education programme. Psychol Health 2004;19;689-703. behaviour: a longitudinal analysis. J Adolesc 2006;29:473-94. 16 Wight D, Abraham C. From psycho-social theory to sustainable classroom practice: 31 Santelli JS, Lowry R, Brener ND, Robin L. The association of sexual behaviors with developing a research-based teacher-delivered sex education programme. Health Educ socioeconomic status, family structure, and race/ethnicity among US adolescents. Am J Res 2000;15:25-38. Public Health 2000;90:1582-8. 17 Wight D, Dixon H. SHARE: the rationale, principles and content of a research-based 32 Philliber S, Kaye J, Herrling S, West E. Preventing pregnancy and improving health care teacher-led sex education programme. Educ Health 2004;22:3-7. access among teenagers: an evaluation of the Children’s Aid Society-Carrera 18 Wight D, Abraham C, Scott S. Towards a psycho-social theoretical framework for sexual programme. Perspec Sex Reprod Health 2002;34:244-51. health promotion. Health Education Research 1998;13:317-30. (Accepted 6 October 2006) 19 Raab G, Butcher I. Balance in cluster randomized trials. Stat Med 2001;20:351-65. 20 Henderson M, Wight D, Raab G, Abraham C, Buston K, Hart G, et al. Heterosexual risk doi 10.1136/bmj.39014.503692.55 behaviour among young teenagers in Scotland. J Adolesc 2002;25:483-94. 21 McLoone P. Carstairs scores for Scottish postcode sectors from 2001 census.Glasgow:MRC Social and Public Health Sciences Unit, 2004. www.msoc-mrc.gla.ac.uk/Publications/ Medical Research Council Social and Public Health Sciences Unit, Glasgow G12 pub/PDFs/PHRU/Carstairs.pdf. 8RZ 22 Raab GM, Butcher I. Randomization inference for balanced cluster-randomized trials. M Henderson research scientist Clin Trials 2005;2:130-40. D Wight programme leader 23 Goldstein H, Browne WJ, Rasbash J. Partitioning variation in multilevel models. Under- standing Statistics 2002;1:223-32. A Parkes research associate 24 ISD Scotland. Scottish health statistics: (terminations of pregnancy) abortions performed in School of Nursing Midwifery and Social Care, Napier University, Edinburgh EH4 Scotland. www.isdscotland.org/isd/1916.html. 2LD 25 Information and Statistics Division. Abortion statistics 1998— ISD Scotland Health G M Raab professor Briefing. Edinburgh: ISD, 1999. 26 General Register Office for Scotland. Scotland’s census 2001—statistics on migration. School of Social Sciences, University of Sussex, Brighton BN1 9SN www.gro-scotland.gov.uk/statistics/library/occpapers/scotlands-census-2001- C Abraham professor statistics-on-migration/index.html. 27 Coyle K, Basen-Engquist K, Kirby D, Parcel G, Banspach S, Collins J, et al. Safer choices: Faculty of Humanities and Social Sciences, University of Keele, Keele ST5 5BG Reducing teen pregnancy, HIV, and STDs. Public Health Rep 2001;116:82-93. S Scott professor 28 Scottish Executive. Respect and responsibility: strategy and action plan for improving sexual Centre for Sexual Health and HIV Research, University College London, London health. Edinburgh: Scottish Executive, 2005. WC1E 6AU 29 Stephenson JM, Strange V, Forrest S, Oakley PA, Copas A, Allen E, et al. Pupil-led sex education in England (RIPPLE study): cluster-randomised intervention trial. Lancet GHartdirector 2004;364:338-46. Correspondence to: M Henderson [email protected]

BMJ Online First bmj.com page5of5 Cite this article as: BMJ, doi:10.1136/bmj.39000.486400.55 (published 3 November 2006)

Research BMJ

Effect of isoniazid prophylaxis on mortality and incidence of tuberculosis in children with HIV: randomised controlled trial Heather J Zar, Mark F Cotton, Stanzi Strauss, Janine Karpakis, Gregory Hussey, H Simon Schaaf, Helena Rabie, Carl J Lombard

Abstract Prevention of tuberculosis in children with HIV through prophylaxis with isoniazid may be effective in reducing mortality Objectives To investigate the impact of isoniazid prophylaxis in areas with a high prevalence of tuberculosis. In studies of on mortality and incidence of tuberculosis in children with HIV. adults with HIV, prophylaxis with isoniazid significantly reduced Design Two centre prospective double blind placebo controlled the incidence of tuberculosis and produced a favourable trend in trial. mortality in those with a positive result on a tuberculin skin ≥ Participants Children aged 8 weeks with HIV. test.9–11 The effect of such prophylaxis in children, however, is Interventions Isoniazid or placebo given with co-trimoxazole unknown. either daily or three times a week. We investigated the effect of isoniazid prophylaxis on mortal- Setting Two tertiary healthcare centres in South Africa. ity in children with HIV living in an area with high tuberculosis Main outcome measures Mortality, incidence of tuberculosis, prevalence. We also looked at the incidence of tuberculosis, the and adverse events. susceptibility of M tuberculosis isolates, the occurrence of toxicity, Results Data on 263 children (median age 24.7 months) were and the impact of two different prophylactic regimens. available when the data safety monitoring board recommended discontinuing the placebo arm; 132 (50%) were taking isoniazid. Median follow-up was 5.7 (interquartile range 2.0-9.7) months. Methods Mortality was lower in the isoniazid group than in the placebo group (11 (8%) v 21 (16%), hazard ratio 0.46, 95% confidence We carried out a prospective double blind placebo controlled interval 0.22 to 0.95, P = 0.015) by intention to treat analysis. trial of isoniazid versus placebo given with co-trimoxazole (CTX) The benefit applied across Centers for Disease Control clinical either daily or three times a week in children with HIV in two categories and in all ages. The reduction in mortality was centres in Cape Town, South Africa. The study started in January similar in children on three times a week or daily isoniazid. The 2003; the placebo arm of the study was ended on 17 May 2004 incidence of tuberculosis was lower in the isoniazid group (5 on the recommendation of the data safety monitoring board cases, 3.8%) than in the placebo group (13 cases, 9.9%) (hazard (DSMB) on the basis of the results of interim analyses. ratio 0.28, 0.10 to 0.78, P = 0.005). All cases of tuberculosis confirmed by culture were in children in the placebo group. Participants Conclusions Prophylaxis with isoniazid has an early survival Participants were children aged ≥ 8 weeks with HIV who were benefit and reduces incidence of tuberculosis in children with attending the Red Cross Children’s Hospital, University of Cape HIV. Prophylaxis may offer an effective public health Town, or Tygerberg Children’s Hospital, Stellenbosch University. intervention to reduce mortality in such children in settings Additional inclusion criteria were weight ≥ 2.5 kg, access to with a high prevalence of tuberculosis. transport, and informed consent from a parent or legal guardian. Trial registration. Clinical Trials NCT00330304 Exclusion criteria were chronic diarrhoea, current use or need for prophylaxis with isoniazid, previous hypersensitivity reaction Introduction to isoniazid or to sulphur drugs, severe anaemia (haemoglobin < 70 g/l), neutropenia (absolute neutrophil count < 400 Tuberculosis and HIV are dual pandemics in children in cells/
l), thrombocytopenia (platelet count < 50 000×109/l), or sub-Saharan Africa. Tuberculosis accelerates the course of HIV, non-reversible renal failure. Children who were receiving highly increasing morbidity, mortality, and the frequency of opportun- active antiretroviral therapy (HAART) were eligible for istic infections.1–4 It is an important cause of acute and chronic enrolment if they had been stable on treatment for two to three pneumonia in African children with HIV5–7 and is responsible for months. Children were enrolled during working hours from a major proportion of mortality.8 Infection with Mycobacterium Monday to Friday. Written informed consent was obtained from tuberculosis confirmed by culture has been found in about 8% of a parent or legal guardian. children with HIV admitted to hospital for pneumonia in areas Researchers took each child’s history, carried out a physical with high prevalence of tuberculosis and HIV.5–7 In a postmortem examination, and collected sociodemographic, clinical, and labo- study of Zambian children dying from respiratory disease M ratory data at enrolment. Children were seen by the study team tuberculosis was found in 18% of children with HIV; in children every four weeks for the first six months then every six weeks for older than 12 months with HIV, tuberculosis was second to pyo- the next six months and then every two to three months, genic pneumonia as a cause of death.8 depending on the medical and social circumstances.

BMJ Online First bmj.com page1of7

Copyright 2006 BMJ Publishing Group Ltd Research

Assignment Diagnosis of tuberculosis Pharmacists in each site labelled the trial drugs with sequential Children were screened for tuberculosis on enrolment as study numbers according to variable blocked randomisation lists described above. Any child who developed clinical signs of a prepared by the trial statistician. At enrolment children were also lower respiratory tract infection underwent a tuberculin skin test allocated a sequential study number by the research nurse and and chest radiography. When chest radiography yielded were then randomly assigned to one of the treatment combina- abnormal results, three sequential gastric aspirates and sputum tions. Study investigators were blinded to the assignment. The induction with nebulised hypertonic saline12 were performed for research pharmacist dispensed the drugs. acid fast staining and M tuberculosis culture. Additional specimens were sent for M tuberculosis staining and culture as Prophylaxis clinically indicated. Children were classified as having confirmed Children were randomised to receive prophylaxis with tuberculosis if they were culture positive for M tuberculosis. Prob- co-trimoxazole either daily or three times a week on Monday, able pulmonary tuberculosis was diagnosed when chest Wednesday, and Friday. As recommended by current South Afri- radiography suggested tuberculosis (lymphadenopathy, miliary can guidelines, prophylaxis (5 mg/kg/dose of the trimethoprim pattern, pleural effusion, bronchial compression, or parenchy- component) was given until the age of 12 months after which it mal infiltrate) and the child had at least one of: a positive tuber- was continued in those with clinical category B or C disease culin skin test result, a history of a close contact with tuberculosis, (according to the Centers for Disease Control), in those with loss of weight or failure to gain weight within the previous three severe immunological impairment (CD4 count of < 15% total months, or a positive smear result for acid fast bacilli. The diag- lymphocyte count), or in those who had previous episode of nosis of probable tuberculosis was subject to independent review Pneumocystis jirovecii pneumonia (PCP). The dose of isoniazid (100 mg tablets, Be-Tabs Pharmaceuticals, Johannesburg, South by a blinded investigator. Africa) was 10 mg/kg/day with a variability of 8-12 mg/kg Children with confirmed or probable tuberculosis were ran- depending on whether half or quarter tablets were required. Pla- domised at enrolment but isoniazid or placebo was started only cebo was manufactured to have an identical appearance to isoni- after they had finished standard tuberculosis treatment. Children azid tablets. Children received isoniazid or placebo according to who developed confirmed or probable tuberculosis during the the frequency of the co-trimoxazole schedule. The duration of study were unblinded; those in the placebo group were given prophylaxis was planned for two years but was subject to review standard tuberculosis treatment while those in the isoniazid depending on interim results. group were treated with four drugs (isoniazid, rifampicin, pyrazi- namide, and ethambutol or ethionamide). Treatment was modi- Other medication fied according to the antimicrobial susceptibility of cultured Multivitamin supplementation and immunisations were given isolates. according to a standard protocol. HAART was not widely avail- able but was obtained for some children through participation in Toxicity pharmaceutical trials or charitable donations. Standard guide- Clinical or laboratory events were graded 1 to 4 according to the lines for HAART in children in the Western Cape were toxicity criteria of the National Institutes of Health’s division of developed during the study period; although these were not yet AIDS (DAIDS). A grade 3 or 4 reaction was considered an widely implemented, some children were able to access HAART important adverse event and managed according to a standard according to medical and social criteria. protocol.

Investigations Sample size and statistical analysis HIV status was assessed at enrolment by two enzyme linked The primary outcome measure was mortality. Using a survival immunosorbent assays (Abbott AxSYM HIV antibody/antigen analysis approach, we estimated that a sample size in each group ELISA) in those aged > 15 months and by polymerase chain of 196 with a 0.050 level one sided log rank test for equality of reaction (Amplicor HIV-1, Roche Diagnostic Systems) in survival curves would provide 80% power to detect the difference younger children. The CD4 cell count and percentage was meas- between an isoniazid-placebo mortality proportion at time t of ured at study entry and every six months. Full blood count, renal 0.100 and an isoniazid mortality proportion at time t of 0.050 (a function (urea, creatinine), and alanine transaminase (ALT) were constant hazard ratio of 0.769), assuming no dropouts before measured at baseline and every six months. A screening tubercu- time t. Therefore, with an assumed 10% dropout rate, the lin skin test (PPD, 2 TU RT23, Staten Serum Institut, estimated required sample size was 216 children per arm or a Copenhagen, Denmark) was done on enrolment and repeated total sample size of 432. every six months if previous results were negative. A positive All analyses were by intention to treat. We used the result was regarded as ≥ 5 mm transverse induration. Children underwent screening chest radiography at enrolment and there- Kaplan-Meier method to analyse the time to event outcomes, after every six months. Blood tests, tuberculin skin tests, and made comparisons with a one sided log rank test, and used Cox chest radiography were performed more frequently if clinically proportional hazards regression to estimate hazard ratios after indicated. In addition, for children receiving HAART, full blood confirming the validity of the proportional hazards assumption. count and alanine transaminase were measured one and three This assumption was tested with the Grambsch and Therneau 13 months after randomisation and thereafter every six months or test. We did subgroups comparisons for severity of disease, more frequently if clinically indicated. If children were admitted dose, age, and study site to assess the consistency of the interven- to hospital or were ill between study visits, we took a detailed his- tion effect. Tests for heterogeneity were done with the Cox tory and carried out clinical examination and laboratory tests as regression model, except for tuberculin positivity as there were clinically indicated. The hospital or clinic records were obtained too few events in this subgroup. Anthropometric measurements whenever possible for children who died; in the absence of these were standardised with reference to standards from the National a verbal autopsy was performed when feasible. Center for Health Statistics. page2of7 BMJ Online First bmj.com Research

Data safety monitoring board Table 1 Comparison of children in isoniazid prophylaxis and placebo groups A safety monitoring board comprising international and South at randomisation. Figures are medians (interquartile range) for continuous African experts reviewed and monitored the study at regular variables or percentage (number) of children as shown intervals. Interim data analyses were undertaken every three to six months to ensure the safety of the study and review progress. Isoniazid (n=132) Placebo (n=131) Total (n=263) Age (months): Median 29.6 (11.0-55.3) 22.1 (8.9-45.4) 24.7 (9.4-51.6) <12 months 26% (35) 32% (42) 29% (77) Results 12-24 months 19% (25) 20% (26) 19% (51) >24 months 55% (72) 48% (63) 51% (135) At the first meeting of the monitoring board, data up to 30 Sep- Boys 55% (72) 57% (74) 56% (146) tember 2003 were analysed. Of the 129 children enrolled 13/61 Co-trimoxazole prophylaxis: died in the placebo group and 4/68 died in the isoniazid group Daily 49% (64) 46% (60) 47% (124) (P = 0.009). The second meeting of the board in April 2004 con- Three times a week 52% (68) 54% (71) 53% (139) sidered data up to 30 December 2003. At that time, of the 148 Weight for age (z score) −1.55 −1.62 (−2.61-−0.44) −1.56 children enrolled, 16 died in the placebo group and five in the (−2.47-−0.44) (−2.49-−0.43) isoniazid group (P = 0.002 by intention to treat and P < 0.001 for Weight for height (z score) −0.07 −0.19 (−1.14-0.92) −0.15 (−1.07-0.88) (−1.08-0.88) on treatment analysis). Both of these analyses met the Centers for Disease Control classification : O’Brien-Fleming rule for stopping a study, which requires N 1% (1) 2% (2) 1% (3) 14 P<0.01. As soon as the board recommended it, we terminated A 10% (13) 12% (16) 11% (29) the placebo arm of the study. At this time, 277 children were B 66% (87) 66% (86 66% (173) enrolled. We excluded 14 children from the analysis (10 tested C 24% (31) 21% (27) 22% (58) negative for HIV, four were lost to follow-up within a month after Tuberculosis: randomisation) and included 263 children (146 (56%) boys) in Prior TB treatment 17% (23) 14% (18) 16% (41) the analysis (fig 1). Of these, 132 were assigned to isoniazid (three Positive tuberculin skin 12% (15/128) 5% (7/129) 9% (22/257) times a week in 68 and daily in 64). Median follow-up time was test Immune classification: 5.7 months (interquartile range 2.0-9.7 months). 1 30% (40) 23% (30/129) 27% (70/261) Table 1 provides the baseline characteristics of the children. 2 40% (53) 41% (53/129) 41% (106/261) About half were younger than 24 months. Most children (231, 3 30% (39) 36% (46/129) 33% (85/261) 88%) were symptomatic, either Centers for Disease Control Laboratory tests: clinical category B or C. The median CD4 percentage was 20%; Alanine transaminase 23 (7-30) (n=113) 22 (6-31) (n=111) 30 (22-43) the proportion of moderate or severely immunosuppressed chil- (U/l) dren was similar in both groups. Overall, children were malnour- White cell count (109/l) 9.5 (7.0-13.4) 10.7 (7.7-13.8) 10.3 (7.2-13.4) ished with the median weight for age z score equal to − 1.6 (n=120) (n=122) − − CD4 (% lymphocytes) 21 (14-29) 19 (14-27) (n=118) 20 (14-28) (interquartile range 2.5- 0.4) and the median weight for (n=114) − − height z score equal to 0.2 ( 1.1-0.9). Forty one (16%) children Receiving HAART at 10% (13) 8% (10) 9% (23) had a history of tuberculosis, with a similar number in both enrolment groups. Tuberculin skin test results were positive in 22 (9%); these HAART=highly active antiretroviral therapy. children had previously received either prophylaxis or treatment for tuberculosis. At enrolment, 23 (9%) were receiving HAART, Effect on mortality while 58 (22%) started HAART during the trial. The number of Mortality (32 deaths in 263 children, 12%) was lower in the isoni- children who received HAART during the trial was similar both azid group than in the placebo group (11/132 (8%) v 21/131 groups (41 in isoniazid group and 40 in placebo group). (16%), hazard ratio 0.46, 95% confidence interval 0.22 to 0.95,

Eligible at screening (n=326) Excluded (n=49): Refused consent (n=11) Enrolled in another trial (n=17) Lost to follow-up (n=17) Children randomised (n=277) Died (n=4)

Tygerberg Hospital (n=155) Red Cross Hospital (n=122)

Allocated to isoniazid Allocated to placebo Allocated to isoniazid Allocated to placebo (n=77): (n=78): (n=62): (n=60): Three times a week Three times a week Three times a week Three times a week (n=38) (n=41) (n=31) (n=31) Daily (n=39) Daily (n=37) Daily (n=31) Daily (n=29)

Did not have HIV (n=5) Did not have HIV (n=3) Did not have HIV (n=0) Did not have HIV (n=2) Lost to follow-up (n=1) Lost to follow-up (n=2) Lost to follow-up (n=1) Lost to follow-up (n=0)

Analysed (n=71) Analysed (n=73) Analysed (n=61) Analysed (n=58)

Fig 1 Allocation of participants and flow through trial

BMJ Online First bmj.com page3of7 Research

1.00 Table 3 Incidence of tuberculosis in children allocated to isoniazid INH prophylaxis or placebo 0.95 Placebo Isoniazid Placebo n=131 Total n=263 HR (95% CI) 0.90 n=132 (%) (%) (%) Intention to treat 5/132 (4) 13/131 (10) 18/263 (7) 0.28 (0.10 to 0.78)

Survival probability 0.85 Frequency of dose: Three times a 3/68 (4) 5/71 (7) 8/139 (6) 0.45 (0.11 to 1.90) 0.80 week Daily 2/64(3) 8/60(13) 10/124 (8) 0.16 (0.03 to 0.76) 0.75 Centers for Disease Control classification: 0.70 A+N 0/14 (0) 1/18 (6) 1/32 (3) No estimate 0 100 200 300 400 500 B 4/87 (5) 11/86 (13) 15/173 (9) 0.22 (0.07 to 0.70) Time since randomisation (days) C 1/31 (3) 1/27 (4) 2/58 (3) 0.86 (0.05 to 13.8) Age group (months): Fig 2 Survival in children on isoniazid (INH) or placebo <12 0/35 (0) 0/42 (0) 0/77 (0) No estimate 12-24 2/25 (8) 5/26 (19) 7/51 (14) 0.50 (0.10 to 2.60) >24 3/72 (4) 8/63 (13) 11/135 (8) 0.26 (0.07 to 0.98) P = 0.015 for the one sided log rank test and P = 0.226 for the Tuberculin skin test result (n=257): proportional hazards assumption; fig 2). The benefit applied to Positive 0/15 (0) 1/7 (14) 1/22 (5) No estimate children across all categories of severity of clinical disease (test Negative 5/113 (4) 12/122 (10) 17/235 (7) 0.32 (0.11 to 0.90) for heterogeneity P = 0.933) and in all ages (test for heterogene- Receiving HAART at enrolment: ity P = 0.678), table 2. The reduction in mortality was similar in Yes 0/13 (0) 1/10 (10) 1/23 (4) No estimate children assigned to isoniazid three times a week compared with No 5/119 (4) 12/121 (10) 17/240 (7) 0.31 (0.11 to 0.87) every day (test of heterogeneity P = 0.943). HAART=highly active antiretroviral therapy. There were no deaths among children with positive results on tuberculin skin testing. The estimated hazard ratio for annual non-HAART mortality of 43.6 per 100 in the placebo children negative for tuberculin was 0.51 (0.24 to 1.07). For chil- group and 14 per 100 in the isoniazid group. dren who received HAART at anytime during the study, three of In most children (27, 84%) the cause of death could be 41 in the isoniazid group died with 195 months of exposure (the reliably determined. Clinical sepsis was the cause in 14 (44%); 10 three children who died had a combined HAART exposure of (31%) had bacteraemia confirmed by culture. Most cultures were 1.8 months) compared with none of the 40 children in the Gram negative bacteria (Klebsiella pneumoniae in four, acineto- placebo group with 170 months of HAART exposure. The over- bacter in two, citrobacter in two, K xanii in one, enterobacter spe- all time without HAART was 577 months in the placebo group cies in one, Escherichia coli in one), and Staphylococcus aureus was and 685 months in the isoniazid group. This converts to an isolated in two children. Three children had polymicrobial sepsis, and one child had concomitant cryptosporidial diarrhoea. Other causes of death included pneumonia (7, 22%), gastroenteritis (3, Table 2 Mortality and hazard ratios (HR) in children allocated to isoniazid 9%), and wasting syndrome, HIV encephalopathy with prophylaxis or placebo respiratory depression, and Burkitt’s lymphoma in a single case Isoniazid (%) Placebo (%) Total (%) HR (95% CI) each. In five (16%) children the cause of death could not be Intention to treat 11/132 (8) 21/131 (16) 32/263 (12) 0.46 ascertained: three died at home and two died at a local day health (0.22 to 0.95) facility. Frequency: Three times a 5/68 (7) 9/71 (13) 14/139 (10) 0.44 Incidence of tuberculosis week (0.17 to 1.18) The incidence of confirmed or probable tuberculosis cases by Daily 6/64 (9) 12/60 (20) 18/124 (15) 0.49 intention to treat analysis (18 cases in 263 children, 7%) was (0.17 to 1.47) lower in the isoniazid group than in the placebo group (5/132 Centers for Disease Control classification: (4%) v 13/131 (10%), hazard ratio 0.28, 0.10 to 0.78, P = 0.005 for A+N 1/14 (7) 2/18 (11) 3/32 (9) 0.82 (0.07 to 9.22) the one sided log rank test and P = 0.919 for the proportional B 6/87 (7) 11/86 (13) 17/173 (10) 0.45 hazards assumption, table 3). The total child months of intent to (0.16 to 1.21) treat follow-up in the two groups for the incidence of tuberculo- C 4/31 (13) 8/27 (30) 12/58 (21) 0.41 sis (total time from randomisation until a positive tuberculosis (0.12 to 1.35) event or censoring event for each child) was 667 months in the Age group (months): <12 7/35 (20) 13/42 (31) 20/77 (26) 0.43 placebo group and 839 months in the isoniazid group. This (0.17 to 1.09) translated into 7.2 cases of tuberculosis annually per 100 12-24 3/25 (12) 5/26 (19) 8/51 (16) 0.75 children in the isoniazid group compared with 23.4 cases in the (0.18 to 3.13) placebo group and an incidence rate ratio of 0.31 (0.09 to 0.91). >24 1/72 (1) 3/63 (5) 4/135 (3) 0.26 The protective effect of isoniazid on incidence of tuberculosis (0.03 to 2.49) Tuberculin skin test result (n=257): occurred in all categories of severity of clinical disease in Positive 0/15 (0) 0/7 (0) 0/22 (0) No estimate children aged > 1 year and in both dose regimens (table 3). All Negative 11/113 (10) 20/122 (16) 31/235 (13) 0.51 five cases of tuberculosis confirmed by culture occurred in the (0.24 to 1.07) placebo group. All M tuberculosis isolates were sensitive to Receiving HAART at enrolment: anti-tuberculosis drugs including isoniazid. Yes 0/13 (0) 0/10 (0) 0/23 (0) No estimate No 11/119 (9) 21/121 (17) 32/240 (13) 0.46 Toxicity (0.22 to 0.95) The incidence of grade 3 or 4 toxicity was low with five (4%) in HAART=highly active antiretroviral therapy. the isoniazid group and eight (6.1%) in the placebo group. Of page4of7 BMJ Online First bmj.com Research these, two were increases in alanine transaminase activity, both in HIV,312raising the possibility that children with early or subclini- the placebo group, while 11 were haematological events includ- cal M tuberculosis infection were not detected. ing neutropenia, thrombocytopenia, or anaemia. Alternative Possible mechanisms of isoniazid efficacy causes of haematological events included infections, other drugs, The effect of isoniazid prophylaxis on incidence of tuberculosis and HIV infection. No child required permanent discontinua- may therefore have been because of treatment of early, subclini- tion of trial drug. No cutaneous or neurological toxicity was cal, or latent M tuberculosis infection. In addition, ongoing isoni- observed. No grade 3 or 4 toxicity occurred among children azid treatment may have provided primary or secondary receiving HAART. prophylaxis against infection. Recently it has been reported that treatment for latent tuberculosis with isoniazid (but not Discussion rifampicin) alters the immune response, resulting in an increase in the number of interferon  producing T cells within a month Efficacy of isoniazid prophylaxis 19 of therapy. Therefore isoniazid prophylaxis, by enhancing the Isoniazid prophylaxis significantly reduced mortality in children host immune response, may provide longstanding protection. with HIV who were living in an area with a high prevalence of This mechanism may explain the long term protection (in excess tuberculosis. The impact on mortality was evident in all of 19 years) provided by nine months of isoniazid prophylaxis in categories of clinical disease, across age groups, and for varying 19 20 Alaskan people without HIV. The ability of isoniazid to degrees of immune suppression. The effect on survival occurred stimulate such immunity in people with HIV is not known. Alter- within six months of the initiation of prophylaxis and was in natively, isoniazid may have effectively provided primary addition to that provided by co-trimoxazole. Furthermore, isoni- prophylaxis against M tuberculosis in an area where children are azid prophylaxis reduced the incidence of tuberculosis by about 16 continuously at high risk of infection. 70%. The impact on survival and incidence of tuberculosis was The mechanism whereby isoniazid prophylaxis improves similar for isoniazid three times a week or once a day. Few survival in children with HIV is unclear but could occur in children were taking HAART at randomisation, reflecting the several ways. Co-infection with M tuberculosis and HIV results in poor access and unaffordability of antiretroviral therapy for most more rapid deterioration of immune dysfunction, viral children in sub-Saharan Africa, so we could not evaluate the 21 replication, and progression of HIV. Such immune decline has impact of isoniazid prophylaxis on mortality in this subgroup. been reported to result in more frequent, severe infections such The three deaths that occurred in children taking isoniazid and 22 as bacterial sepsis, which was the cause of death in most of our HAART occurred soon after HAART initiation, suggesting that children. Tuberculosis, however, may also be a direct cause of it was started too late in these children with advanced HIV mortality, as has been reported in a postmortem study of disease. Zambian children in which tuberculosis was the second The Western Cape area of South Africa has one of the high- commonest cause of death in children aged > 1 year with HIV est incidences of tuberculosis in the world, with reported rates of 8 who died from respiratory disease. As confirmation of M tuber- 988/100 000 population in 200415 and an estimated annual risk culosis by culture is difficult and positive in about a third of chil- of infection of 3.8%.16 Children aged < 15 years contribute about dren with clinically suspected tuberculosis, and, as bacterial 20% of the case load.17 In addition, the incidence of HIV co-infection may occur, it is possible that tuberculosis was under- infection, as reflected by the prevalence in pregnant women, has 12 diagnosed in our study. A substantial number of children (22%) increased exponentially from 8.6% in 2001 to 15.4% in 2004.18 In died from clinical pneumonia, some of whom may have had contrast with our findings, a Cochrane review of prophylaxis in tuberculosis. A limitation of our study is that we were unable to adults with HIV did not find a significant reduction in mortality, perform postmortem studies to investigate this. A further although a favourable trend on survival was reported for adults mechanism for the efficacy of isoniazid may be activity against with positive results on tuberculin skin test.11 Furthermore, our organisms other than M tuberculosis. For example, isoniazid in observed reduction in incidence of tuberculosis in children combination with rifampicin has been reported to be effective taking isoniazid prophylaxis was greater than that reported for 23 11 against leishmaniasis, and isoniazid has been shown to inhibit adults with HIV and also occurred in children with negative 24 development of the malaria parasite. Current understanding of tuberculin results. In contrast, chemoprophylaxis in adults with the molecular mechanism of isoniazid activity is incomplete, HIV has been found to be significantly effective only in those though the drug has been shown to interfere with fatty acid with positive results on tuberculin skin test, reducing the risk of 25 metabolism. Although isoniazid has not been shown to have active tuberculosis by about 60%.11 activity against other bacteria, this pathway has been postulated In our study, only a few children had positive results on to affect microbes such as E coli and other gram negative bacte- tuberculin skin test. The impact of isoniazid on mortality and 26 ria. Furthermore, isoniazid blocks the lethality of endotoxin in incidence of tuberculosis could therefore be reliably assessed 27 28 mice and is an efficient scavenger of free radicals. Finally, only in children with a negative result on a tuberculin skin test, in co-trimoxazole prophylaxis has been reported to improve whom we found a consistently protective effect of isoniazid. The 29 survival in African children with HIV ; isoniazid may have an high number of children with negative results on tuberculin skin additive antimicrobial effect. tests may reflect anergy as a result of HIV mediated immunosuppression (most children had moderate or severe Safety and tolerability immunosuppression by CD4 counts), depressed cell mediated The safety and tolerability of isoniazid prophylaxis was excellent, immunity because of malnutrition (most children were also mal- even in the subgroup of children who were taking HAART. Data nourished), or early or lack of infection with M tuberculosis. from adults have confirmed the safety of isoniazid, with reported Tuberculosis in young children with HIV differs from adult dis- hepatotoxicity in only 0.3% of people treated for latent tubercu- ease in that it usually reflects primary infection and may often losis.30 The optimal duration of prophylaxis, however, is not develop into severe or disseminated disease, which may cause known and long term studies are needed. Limited studies in death.3 Although all children were screened carefully for adults with HIV suggest that prolonged use may be associated tuberculosis, diagnosis is notoriously difficult in those with with longer protection.31 32 Careful screening of children for

BMJ Online First bmj.com page5of7 Research tuberculosis before prophylaxis is started is important. Reassur- 1 Palme IB, Gudetta B, Bruchfeld J, Muhe L, Giesecke J. Impact of human immunodefi- ciency virus 1 infection on clinical presentation, treatment outcome and survival in a ingly, the incidence of resistant M tuberculosis infection did not cohort of Ethiopian children with tuberculosis. Pediatr Infect Dis J 2002;21:1053-61. increase in children on prophylaxis, suggesting that it did not 2 Mukadi YD, Wiktor SZ, Coulibaly IM, Coulibaly D, Mbengue A, Folquet AM, et al. Impact of HIV infection on the development, clinical presentation, and outcome of promote the development of resistance. Ongoing monitoring of tuberculosis among children in Abidjan, Cote d’Ivoire. AIDS 1997;11:1151-8. this is needed. 3 Chintu C, Mwaba P. Tuberculosis in children with human immunodeficiency virus infection. Int J Tuberc Lung Dis 2005;9:477-84. Most children with HIV currently live in sub-Saharan Africa, 4 Hesseling AC, Schaaf SH, Westra AE, Werschkull H, Donald PR, Beyers N, et al. where tuberculosis is highly prevalent and where there is limited Outcome of HIV-infected children with culture-confirmed tuberculosis. Arch Dis Child access to antiretroviral therapy. Mortality among these children 2005;24:1171-4. 5 Jeena PM, Pillay P, Pillay T, Coovadia HM. Impact of HIV-1 co-infection on presentation is much higher than that in children with HIV in the developed and hospital-related mortality in children with culture proven pulmonary tuberculosis world, with infant mortality and mortality in those aged < 5 in Durban, South Africa. Int J Tuberc Lung Dis 2002;6:672-7. 33 6 Zar HJ, Hanslo D, Tannenbaum E, Klein M, Argent A, Eley B, et al. Aetiology and out- around 25% and 65%, respectively. Prophylaxis with isoniazid come of pneumonia in human immunodeficiency virus-infected children hospitalized offers an available, well tolerated, and effective means for in South Africa. Acta Paediatr 2001;90:119-25. 7 Madhi SA, Petersen K, Madhi A, Khoosal M, Klugman KP. Increased disease burden improving survival in these children in addition to that provided and antibiotic resistance of bacteria causing severe community-acquired lower respira- by co-trimoxazole. Tuberculosis prophylaxis is cost effective, tory tract infections in human immunodeficiency type 1-infected children. Clin Infect Dis 2000;31:170-6. extends life expectancy, reduces the incidence of tuberculosis, 8 Chintu C, Mudenda V, Lucas S, Nunn A, Lishimpi K, Maswahu D, et al. Lung disease at and provides savings in medical and social costs in adults with necropsy in African children dying from respiratory illnesses: a descriptive necropsy 34 study. Lancet 2002;360:985-90. HIV who are tuberculin skin test positive. Therefore, isoniazid 9 Wilkinson D, Squire SB, Garner P. Effect of preventive treatment for tuberculosis in prophylaxis may be an important public health intervention for adults infected with HIV: systematic review of randomised placebo controlled trials. BMJ 1998;317:625-9. children with HIV living in areas with high prevalence of tuber- 10 Bucher HC, Griffith LE, Guyatt GH, Sundre P, Naef M, Sendi P, et al. Isoniazid prophy- culosis, particularly when antiretroviral therapy is not available. laxis for tuberculosis in HIV infection: a meta-analysis of randomized controlled trials. AIDS 1999;13:501-7. Our results support the routine use of isoniazid prophylaxis in 11 Woldehanna S, Volmink J. Treatment of latent tuberculosis infection in HIV infected such children who cannot access HAART. Further studies in persons. Cochrane Database Syst Rev 2006;(3):CD000171. children of the cost efficacy of this intervention, the long term 12 Zar HJ, Hanslo D, Apolles P, Swingler G, Hussey G. Comparison of induced sputum with gastric lavage for microbiologic confirmation of pulmonary tuberculosis in infants durability of protection, the efficacy in areas with low prevalence and young children—a prospective study. Lancet 2005;365:130-4. of tuberculosis, and the applicability of our findings to those 13 Grambsch PM, Therneau TM. Proportional hazards test and diagnostics based on weighted residuals. Biometrika 1994;81:515-26. receiving HAART is needed. 14 Pocock SJ. When to stop a clinical trial. BMJ 1992;305:235-40. 15 Health Systems Trust Health Statistics. www.hst.org.za/healthstats/16/data (accessed We thank the children and their caregivers for participating. The study team 19 Oct 2006). comprised P Apolles, N Dlaku, E Dobbels, T Fakir, C Ford, D Gray, M 16 Beyers N, Michaelis I, Gie RP, Schaaf HS, Richardson M, Warren R. Transmission of George, L Holt, T Jennings, A Joachim, B Leibbrandt, A Loggie, G Lotter- tuberculosis to children in a high incidence area. Int J Tuberc Lung Dis 2001;5(suppl 1):S185. ing, M Louw, I Mong, P Mtiya, D Nchuna, F Ngcokovana, V Nkondlala, K 17 Soeters M, de Vries AM, Kimpen JLL, Donald PR, Schaaf HS. Clinical features and Orpen, H Smit, R Streicher, E Swanepoel, E Walters, and M A Wolff. The outcome in children admitted to a TB hospital in the Western Cape—the influence of data and safety monitoring committee comprised J Kaplan (chair), W El HIV and drug resistance. SAfrMedJ2005;95:602-6. Sadr, P Donald, and N Beyers. 18 National Department of Health. National HIV and syphilis seroprevalence survey in South Africa 2004. Pretoria: National Department of Health, 2005. Contributors: HJZ and MFC conceived the study, wrote the protocol and 19 Wilkinson KA, Kon OM, Newton SM, Meintjes G, Davidson RN, Pasvol G, et al. Effect grant application, and supervised the study. SS and JK were trial physicians. of treatment of latent tuberculosis infection on the T cell response to Mycobacterium HSS and GH assisted with design, supervision, and coordination of the tuberculosis antigens. J Infect Dis 2006;193:354-9. study. HSS also reviewed clinical diagnoses. HR supervised trial 20 Comstock GW, Baum C, Snider DE. Isoniazid prophylaxis among Alaskan Eskimos: a final report of the bethel isoniazid studies. Am Rev Respir Dis 1979;119:827-30. management. CJL contributed to study design and was responsible for 21 Toossi Z. Virological and immunological impact of tuberculosis on human immunode- development of the database and statistical analysis. HZ drafted the paper ficiency virus type I disease. J Infect Dis 2003;188:1146-55. and is guarantor, and all authors contributed to the final manuscript. 22 Whalen C, Horsburgh CR, Hom D, Lahart C, Simberkoff M, Ellner J. Accelerated Funding: Rockefeller Foundation, USA. course of human immunodeficiency virus infection after tuberculosis. Am J Resp Crit Care Med 1995;151:129-35. Competing interests: None declared. 23 Peters W, Lainson R, Shaw JJ, Robinson BL, Leao AF. Potentiating action of rifampicin Ethical approval: Research and ethics committees of the Universities of and isoniazid against Leishmania mexicana amazonensis. Lancet 1981;i:1122-4. 24 Arai M, Alavi YI, Mendoza J, Billker O, Sinden RE. Isonicotinic acid hydrazide: an anti- Cape Town and Stellenbosch, South Africa. tuberculosis drug inhibits malarial transmission in the mosquito gut. Exp Parasitol 2004;106:30-6. 25 Ducasse-Cabanot S, Cohen-Gonsaud M, Marrakchi H, Nguyen M, Zerbib D, Bernadou J, et al. In vitro inhibition of the Mycobacterium tuberculosis beta-ketoacyl-acyl carrier What is already known on this topic protein reductase MabA by isoniazid. Antimicrob Agents Chemother 2004;48:242-9. 26 Baldock C, de Boer GJ, Rafferty JB, Stuitje AR, Rice DW.Mechanism of action of diaza- Prophylaxis with isoniazid significantly reduces the borines. Biochem Pharmacol 1998;55:1541-9. incidence of tuberculosis in adults with HIV and a positive 27 Urbaschek R, Männel DN, Urbanczik R. Isoniazid protects mice against endotoxin lethality without influencing tumor necrosis factor synthesis and release. Antimicrob tuberculin skin test result Agents Chemother 1991;35:1666-8. 28 Choi D, Leininger-Muller B, Kim YC, Leroy P, Siest G, Wellman M. Differential role of There are no published data on the impact on mortality or CYP2E1 binders and isoniazid on CYP2E1 protein modification in NADPH- dependent microsomal oxidative reactions: free radical scavenging ability of isoniazid. incidence of tuberculosis in children with HIV Free Radic Res 2002;36:893-903. 29 Chintu C, Bhat GJ, Walker AS, Mulenga V, Sinyinza F, Lishimpi K, et al. Co-trimoxazole What this study adds as prophylaxis against opportunistic infections in HIV-infected Zambian children (CHAP): a double-blind randomised placebo-controlled trial. Lancet 2004;364:1865- 71. Prophylaxis with isoniazid in children significantly reduced 30 LoBue PA, Moser KS. Use of isonaizid for latent tuberculosis infection in a public mortality by about 50% and incidence of tuberculosis by health clinic. Am J Resp Crit Care Med 2003;168:443-7. 31 Casado JL, Moreno S, Fortun A, Antela A, Quereda C, Navas E, et al. Risk factors for about 70% development of tuberculosis after isoniazid chemoprophylaxis in human immunodefi- ciency virus-infected patients. Clin Infect Dis J 2002;34:386-9. The reduction in mortality occurred in all categories of 32 Fitzgerald DW, Desvarieux M, Severe P, Joseph P, Johnson WD Jr, Pape JW. Effect of post-treatment isoniazid on prevention of recurrent tuberculosis in HIV-1-infected clinical disease, in children in all age groups, and for individuals: a randomised trial. Lancet 2000;356:1470-4. varying degrees of immune suppression 33 Spira R, Lepage P, Msellati P, van de Perra P, Leroy V, Simonon A, et al. Natural history of human immunodeficiency virus type 1 infection in children: a five-year prospective study in Rwanda. Pediatrics 1999;104:e56. Such prophylaxis may offer an effective public health 34 Bell JC, Rose DN, Sacks HS. Tuberculosis preventive therapy for HIV-infected people intervention to reduce mortality in children with HIV in sub-Saharan Africa is cost-effective. AIDS 1999;13:1549-56. (Accepted 29 September 2006) page6of7 BMJ Online First bmj.com Research

doi 10.1136/bmj.39000.486400.55 Mark F Cotton associate professor Janine Karpakis research medical officer School of Child and Adolescent Health, Red Cross Children’s Hospital, University of Cape Town, South Africa H Simon Schaaf associate professor Heather J Zar associate professor Helena Rabie clinical researcher Stanzi Strauss research medical officer Biostatistics Unit, Medical Research Council, South Africa Gregory Hussey professor Carl J Lombard senior statistician Department of Paediatrics and Child Health, Tygerberg Children’s Hospital, Correspondence to: H Zar [email protected] Stellenbosch University, South Africa

BMJ Online First bmj.com page7of7 Cite this article as: BMJ, doi:10.1136/bmj.39035.482396.55 (published 28 November 2006)

Research BMJ

Effects of fortified milk on morbidity in young children in north India: community based, randomised, double masked placebo controlled trial Sunil Sazawal, Usha Dhingra, Girish Hiremath, Jitendra Kumar, Pratibha Dhingra, Archana Sarkar, Venugopal P Menon, Robert E Black

Abstract cycle between deficiency and morbidity and mortality from infections.7 Objective To evaluate the efficacy of milk fortified with specific Reducing the associated higher risk of infectious disease by multiple micronutrients on morbidity in children compared alleviating deficiencies in micronutrients among young children with the same milk without fortification. in developing countries is now important in research and imple- Design Community based, double masked, individually mentation in public health. The World Health Organization and randomised trial. the International Nutritional Anemia Consultative Grouphave Setting Peri-urban settlement in north India. recommended iron supplementation in preschool children.8 Participants Children (n = 633) aged 1-3 randomly allocated to Clinical trials have shown that supplementation with zinc has a receive fortified milk (n = 316) or control milk (n = 317). promising effect.9 Because of the inherent technical and practical Intervention One year of fortified milk providing additional limitations of a supplementation strategy,9–11 the limited success
12 7.8 mg zinc, 9.6 mg iron, 4.2 g selenium, 0.27 mg copper, 156 of nutritional education programmes, and the difficulty in pro-
13 g vitamin A, 40.2 mg vitamin C, 7.5 mg vitamin E per day moting animal based foods, delivery of zinc and iron through (three feeds). fortification of commonly consumed foods remains the only Main outcome measures Days with severe illnesses, incidence practical and sustainable option. National fortification pro- and prevalence of diarrhoea, and acute lower respiratory illness. grammes and fortification studies have shown that fortification Results Study groups were comparable at baseline; compliance of common home based food can overcome the limitations in the groups was similar. Mean number of episodes of inherent to supplementation at a lower cost.14–16 Milk is a popular diarrhoea per child was 4.46 (SD 3.8) in the intervention food for fortification because it is readily available, widely (fortified milk) groupand 5.36 (SD 4.1) in the control group. accepted, and frequently fed to young children. Mean number of episodes of acute lower respiratory illness was To evaluate acceptability and efficacy of delivering specific 0.62 (SD 1.1) and 0.83 (SD 1.4), respectively. The fortified milk micronutrients, including zinc and iron, through fortified milk reduced the odds for days with severe illnesses by 15% (95% we conducted a community based, double masked, randomised confidence interval 5% to 24%), the incidence of diarrhoea by controlled trial comparing fortified milk with the same milk 18% (7% to 27%), and the incidence of acute lower respiratory without fortification for prevention of common childhood mor- illness by 26% (3% to 43%). Consistently greater beneficial bidities. effects were observed in children aged ≤ 24 months than in older children. Conclusion Milk is well accepted as a means of delivery of Methods micronutrients. Consumption of milk fortified with specific Participants micronutrients can significantly reduce the burden of common The trial was carried out in Sangam Vihar, a peri-urban popula- morbidities among preschool children, especially in the first tion in New Delhi, India, from April 2002 to April 2004. Details two years of life. of the population have been previously published.17 18 From a Trial registration NCT00255385. regularly updated demographic database we invited all permanently resident families with children aged 1-3 years to Introduction participate in the study and sought their consent. Irrespective of groupallocation all children who had severe anaemia at baseline The widespread prevalence of deficiencies in multiple micronu- were given a therapeutic dose of iron for three months in addi- trients among preschool children is an important determinant tion to their milk supplement. of child health.12 Because they affect the immune system and have antioxidant properties some micronutrients have a crucial Baseline assessment, blood samples, and laboratory role in generation, maintenance, and amplification of immune procedures response.34 Animal models have shown that deficiency in After we obtained consent, children were enrolled and scheduled specific micronutrients facilitates mutations in infectious agents, to visit the clinic. A detailed baseline assessment included exami- enhancing their infectivity and virulence.56 Infectious diseases nation by a physician and the collection of socioeconomic and can also lead to deficiency of micronutrients, creating a vicious demographic information. Two independent observers meas-

BMJ Online First bmj.com page1of5

Copyright 2006 BMJ Publishing Group Ltd

Research ured weight to the nearest 10 g with an electronic scale (SECA Table 1 Composition of milk preparation as nutritive value in three serving Corporation, Columbia, MD) and length or height to the nearest 0.1 cm with length boards (Shorr Productions, Olney, MD). Fortified milk Control milk Researchers collected a venous blood sample at baseline and at Energy (MJ) 1.89 1.89 the end of the study in trace element-free syringes. The blood Protein (g) 20.1 20.1 Taurine (mg) 48 48 was transferred into vials with EDTA, and vials with zinc-free Carbohydrate (g) 48.9 48.9 heparin. Plasma was separated within 15 minutes of blood Fat (g) 18.9 18.9 collection and aliquots transferred into trace element-free Vitamin A* (
g) 330 174 Eppendorf plastic tubes for storage at − 20°C. The EDTA blood Vitamin D3 (
g) 3.6 3.6 was analysed on the same day with a Coulter automated haema- Vitamin E† (mg) 8.1 0.6 tology analyser for a detailed haemogram. One dropof blood Vitamin C (mg) 48.0 7.8 was used to estimate zinc protoporphyrin with a fluorometer Thiamin (mg) 0.6 0.6 (Aviv Biomedical, Lakewood, NJ). Plasma concentration of zinc Riboflavin (mg) 1.8 1.8 was estimated with standard methods with an atomic absorption Niacin (mg) 4.5 4.5 19 spectrometer (AAS 800-Perkin Elmer). Serum ferritin and Vitamin B6 (mg) 0.6 0.6 serum transferrin concentrations were estimated in plasma sam- Pantothenic acid (mg) 2.7 2.7 ples with a commercial enzyme linked immunoabsorbent assay Folate‡ (
g) 114 114 Vitamin B ( g) 2.7 2.7 (ELISA; Ramco Laboratories, Houston, TX). 12
Biotin (
g) 24.9 24.9 Randomisation and masking Choline (mg) 114 114 This study was implemented concurrently with another Calcium (mg) 720 720 two-armed clinical trial to evaluate the efficacy of a different milk Phosphorus (mg) 600 600 preparation fortified with probiotic (compared with that Magnesium (mg) 84 84 Iron (mg) 9.6 0 preparation without fortification) in a non-factorial design with Zinc (mg) 9.6 1.8 joint randomisation. Each treatment group(across the two trials) Iodine (
g) 36 36 was identified by a letter code A, B, C, or D. A random allocation Selenium (
g) 6.6 2.4 sequence of groupcodes was generated with our own software, Copper (mg) 0.3 0.03 with permuted blocks of length 16. Sodium (mg) 360 360 We generated two separate randomisation lists—one for chil- Potassium (mg) 1260 1260 dren with baseline Hb > 70 g/l and another for children with Chloride (mg) 900 900 ≤ baseline Hb 70 g/l. This resulted in two serially numbered lists *Retinol activity equivalents. with allocated treatment codes before we enrolled any children. † tocopherol equivalents. When we entered the baseline information for each child ‡Dietary folate equivalents. (including haemoglobin concentration), the child was automati- cally allocated the next available serial number and the letter sachets given at the parents’ request to children going away from code in the haemoglobin group. The supplementation sachets home. To prevent sharing of supplement with other siblings, were identical in colour, size (weight 32 g), and taste and were mothers could be given more sachets. The intervention (fortified labelled with a letter code. The product corresponding with the milk in three servings a day) was designed to deliver zinc 9.6 mg letter code was known only to the manufacturing supervisor in (7.8 mg more than control group) and iron 9.6 mg. The fortified New Zealand; it was not known to investigators or anyone in the milk also included extra vitamin C to improve iron absorption, field until the study was finished and the data analysed. In the copper to counteract possible effects of zinc and iron on copper field, the letter code of the supplementation box was stripped off absorption, selenium, vitamin A, and vitamin E to help and labelled with the child’s identification information. We antioxidant and immune effects of zinc. report here on children allocated to two of the four letter codes.

Sample size and power Household surveillance during follow-up We calculated the sample size using actual data from one of our During weekly visits to households assistants distributed the milk recently concluded studies. We used a rate of 0.019 (6.8 sachets and collected unconsumed sachets and information on episodes/child year) for diarrhoea and 0.003 (1.05 episodes/ compliance. Other assistants visited homes twice a week to child) for pneumonia. Assuming a Poisson distribution for collect morbidity information since the last visit. Both teams incidence, we aimed at a 15% reduction in diarrhoea and 25% gathered information on compliance and were retrained for two reduction in pneumonia, at an  of 0.05 and 90% power. We to three months on the specific instruments used. If the child or inflated the estimated sample size by 10% to account for possible the parent was not available on a scheduled visit day, the team clustering of events and by another 10% to account for potential revisited the household the next day. There were two levels of attrition of children participating in the trial and therefore supervision and random checking. About 10% of the houses enrolled 325 children in each group. were randomly visited to verify the overall information gathered. At each home visit, supervisors recorded information for each of Intervention the previous three to four days since the last visit, including We used 32 g single serving sachets of fortified milk powder number of diarrhoeal stools, consistency of stools, blood in (Fonterra Brands, New Zealand). At enrolment, mothers were stools, pneumonia, fever, vomiting, and history of feeding. They shown how to reconstitute the powdered milk. Table 1 shows the also cross checked compliance with consumption of milk. Moth- composition of milk in the two groups. Assistants delivered 21 ers were advised to contact study physicians at the clinic if they sachets each week to each home and advised that the child thought that the child was seriously ill between visits. All visits should consume up to three sachets a day. Supplementation was either to the study physicians or to private physicians were continued for one year. We recorded the use of additional recorded. Treatment of diarrhoea, dysentery and pneumonia as page2of5 BMJ Online First bmj.com Research per WHO guidelines was provided free to the participating chil- Assessed for eligibility (age 1-3 years, permanent residents, dren throughout the study. no severe illness, no severe malnourishment) (n=660) Primary outcomes and definition of clinical outcomes Our primary outcomes were episodes of diarrhoea and acute Consent obtained lower respiratory tract infections or pneumonia and days with severe illness. Diarrhoea was defined as three or more loose or Declined to participate (n=27) watery stools in 24 hours, and children were considered to have Children enrolled (n=633) recovered after three days without diarrhoea. Acute lower respi- ratory tract infection was diagnosed if the child had reported dif- ≥ Randomisation: ficulty in breathing and rapid breathing ( 40/min). Severe (stratified by severity of anaemia: Hb ≤70 g/l or Hb >70 g/l) illness consisted of temperature ≥ 38.4°C or admission to hospi- tal or respiratory rate ≥ 50/min or chest indrawing. Dysentery was defined as diarrhoea with visible blood in stools, severe acute Intervention group Control group lower respiratory tract infections (worsening of existing infection (fortified milk) (n=316) (milk not fortified) (n=317) Hb ≤70 g/l (n=32) Hb ≤70 g/l (n=33) or new onset of cough or difficulty in breathing with high respi- Hb >70 g/l (n=284) Hb >70 g/l (n=284) ration rate ( ≥ 50/min) or chest indrawing), fever (axillary ≥ temperature 37.2°C), and high fever (axillary temperature Premature termination (n=27) Premature termination (n=36) ≥ 38.4°C). Withdrawn consent (n=16) Withdrawn consent (n=16) Left area (n=9) Left area (n=18) Data management and statistical methods Died (n=2) Died (n=2) We used Visual Basic 6.0 and Oracle 8i to manage our data, with stringent range, consistency, and logical checks. Real time data Children included for analysis Children included for analysis entry ensured data quality and accuracy. We used double data (n=316) (n=317) entry and manual checking of frequencies during data cleaning before the code was broken. For primary analysis we used alpha- Flow of the participants through study betical codes for groups still blinded to real group identity. We performed intent to treat analysis (all children were included in lower respiratory tract infection was 26% (3% to 43%) lower in analyses irrespective of their adherence to the supplement) and the intervention groupthan in the control group.In children included all data gathered during the intervention period of one ≤ 24 months of age, the incidence was 47% (19% to 65%) lower. year. For children leaving the study area or withdrawing from the There was a significant reduction in prevalence of days with high study, we included data until the date of censorship. Person-time respiratory rate ( ≥ 40/min). Overall days with severe illness were analysis was performed with actual follow-up as denominator. 15% (5% to 24%) lower in the intervention groupthan in the For the effect on incidence of diarrhoea, acute lower respiratory control group. Although there was 88% (1% to 98%) reduction in tract infection, and measles, we estimated relative risk using Pois- risk for measles in the intervention group, there were few son regression; and for prevalence, we estimated odds ratio using episodes (1 v 8). There was 4% (1% to 8%) lower use of antibiot- GLM for binomial outcomes (maximum likelihood logit estima- ics in the intervention groupcomparedwith the control group. tion for grouped data). In both estimations, we used robust clus- Table 3 summarises the results. Adherence to the milk feeds was tered standard error estimation, the clustering variable being the similar in the groups; 85.6% in the intervention group and 86.7% child.20 21 This modified sandwich estimator is unbiased for clus- in the control groupconsumed two or three servings on more ter correlated data regardless of the setting. All statistical analyses than 80% of days. This did not vary by the intervention period, were performed in STATA 9.2, (StataCorp, College Station, TX). and there were no adverse events observed.

Results Discussion Out of 660 eligible children contacted, we received consent for In this study children who received fortified milk compared with 633 children (316 in intervention, 317 in control) to participate those who received same quantity and quality milk without spe- in the study (figure) and collected data on 190 324 child days of cific fortification had 18% lower incidence of diarrhoea, 26% follow-up. The children in both the groups were comparable at lower incidence of pneumonia, 7% fewer days with high fever, baseline for sociodemographic variables, haematology, and and 15% fewer days sick with severe illnesses. The significantly plasma zinc status (table 2). The mean number of sachets lower rates of measles and use of antibiotics suggest that fortifi- consumed was 2.58 in the intervention groupand 2.54 in the cation resulted in a substantial prevention of morbidity. Together control group. On 77.4% and 77.3% child days, respectively, an these effects suggest improved immunity against common infec- intake of three sachets was recorded. Only 30 (9.5%) children in tions. The fact that all children in the study received about 1.89 the intervention groupand 19 (6.0%) in the control groupdid MJ energy, 20 g protein, 49 g carbohydrates, and 19 g fat, in addi- not have any episodes of diarrhoea. The mean number of tion to their complementary foods or breast milk, suggests that episodes of diarrhoea per child per year was 4.46 (SD 3.8) and these effects could be expected even in moderately well fed chil- 5.36 (SD 4.1), respectively, the figures for acute lower respiratory dren. tract infection being 0.62 (SD 1.1) and 0.83 (SD 1.4). Compared Our study was in keeping with the recently ratified WHO with non-fortified milk, consumption of fortified milk was global strategy on infant and young child feeding, which notes associated with an 18% (95% confidence interval 7% to 27%) that industrially processed complementary foods are an option lower incidence of diarrhoea and a similarly lower prevalence of for mothers who can afford them and have the knowledge and diarrhoea. The difference was significant, irrespective of age. The facilities to safely prepare and feed them.22 We used fortified milk subgroup analyses were performed more as explanatory to deliver zinc and iron. Addition of copper to foods fortified analyses ( ≤ 24 months v > 24 months). The incidence of acute with zinc and iron is recommended because of potential negative

BMJ Online First bmj.com page3of5 Research

There is a growing consensus that both zinc and iron should Table 2 Baseline sociodemographic characteristics of two groups according 24–26 to allocation to fortified (intervention) or non-fortified (control) milk. Figures be given to vulnerable groups, including preschool children. are numbers (percentages) of participants unless stated otherwise The recent WHO/UNICEF review of complementary foods in developing countries concluded that iron and zinc requirements Intervention (n=316) Control (n=317) may be difficult to meet from non-fortified complementary 27 Age (months): foods. In view of their interaction shown in isotope studies, the 28 29 Mean (SD) 22.4 (6.8) 23.0 (6.7) mode of provision needs to be considered carefully. Evidence 30 ≤24 191 (60) 187 (59) from supplementation trials is inconclusive. Interaction does >24 125 (40) 128 (41) not seem to occur when two elements are presented in food Boys 154 (49) 163 (51) products or in the presence of dietary ligands.31–33 Thus, it is likely Illiterate father 54 (17) 52 (16) that the preferred delivery of these two micronutrients will be as Illiterate mother 168 (53) 169 (53) fortification rather than supplementation. Mean (SD) SES* score 7.14 (2.7) 7.28 (2.5) The effects on febrile illness, measles, diarrhoea, and acute Water supply from tap 183 (58) 187 (59) lower respiratory infections are consistent with results of recently Haemoglobin (g/l): published clinical trials on the effect of routine zinc supplemen- Mean (SD) 89.2 (15.0) 91.0 (15.0) tation on incidence of diarrhoea31 33–36 and acute lower ≤70† 32 (10) 33 (10) respiratory tract infections.37 Given the design of this study we >70† 284 (90) 284 (90) Mean (SD) serum transferrin (
g/ml) 25.8 (11.3) 26.9 (11.0) cannot estimate interaction between these minerals, but our Mean (SD) serum ferritin (
g/l) 9.6 (12.9) 9.9 (9.4) results do suggest an overall benefit and thus a lack of any Mean (SD) plasma zinc (
mol/l) 9.3 (3.6) 9.6 (4.1) significant negative interaction between zinc and iron when pro- Nutritional status: vided in the fortified milk. Wasted and stunted 62 (21) 69 (23) In recent years mechanisms for nutritional modulation of the Normal 113 (38) 112 (37) immune response have been investigated and are now better Stunted 68 (23) 65 (21) understood. Lack of selected micronutrients—especially zinc, Wasted 55 (18) 57 (19) selenium, iron, and the antioxidant vitamins—can lead to *Socioeconomic status based on ownership and type of house and household items; clinically important immune deficiency and infections in maximum=26; higher scores=higher status. children. These nutrients act as antioxidants and cofactors at the level of cytokine regulation and can affect cytokine response and effects of zinc and iron supplementation on copper absorption immune cell trafficking.38 and status.23 We added vitamin A, selenium, and vitamin E to the supplement because they might improve the immune and mem- Conclusion brane stabilising effects and vitamin C to improve iron Given that the latest estimates of the percentage of gross domes- absorption. These were added to recommended levels for tic product lost to all forms of vitamin and mineral deficiency is fortified complementary foods. Though this strategy might opti- 1-2% in the developing world,39 there is an urgent need to mise the effect of the intervention, it limits our ability to developand implementstrategies to reduce the burden of these determine the contributions of individual micronutrients. deficiencies. We have shown that micronutrients, especially zinc and iron, at levels that have been traditionally delivered by Table 3 Effect on morbidity according to allocation to fortified (intervention) supplementation, can be delivered successfully through fortified or non-fortified (control) milk (robust estimates) milk. This mode of delivery is well accepted and retains the pre- vention of morbidity shown in zinc supplementation trials while Intervention Control Relative risk or P (n=316) (n=317) odds ratio (95% CI) value also being able to deliver iron and other specific micronutrients. Gastrointestinal morbidity Episodes of diarrhoea: All children (aged 1-3 years) 1408 1700 0.82 (0.73 to 0.93) 0.002 What is already known on this topic Age ≤24 months 529 555 0.84 (0.71 to 0.96) 0.03 Age >24 months 879 1145 0.80 (0.69 to 0.93) 0.004 Deficiency in specific micronutrients, especially iron and Days of diarrhoea 3277 4010 0.81* (0.77 to 0.85) 0.00 zinc, is prevalent in preschool children in developing Episodes of dysentery 121 133 0.91 (0.67 to 1.22) 0.52 countries and predisposes these children to common Respiratory morbidity childhood infections Episodes of acute lower respiratory illness: All children (aged 1-3 years) 195 262 0.74 (0.57 to 0.97) 0.03 Supplementation with zinc leads to significant reduction in Age ≤24 months 71 117 0.53 (0.35 to 0.81) 0.003 morbidity from diarrhoea or pneumonia Age >24 months 124 145 0.89 (0.67 to 1.19) 0.42 Days with respiration rate 279 368 0.75* (0.65 to 0.88) 0.00 Supplementation is difficult to implement and has not been >40/min successful in prevention of iron deficiency Severe episodes of acute lower 79 110 0.72 (0.49 to 1.05) 0.09 respiratory illness What this study adds Febrile illness and others Days with severe illness: Fortified milk is well accepted as an intervention to deliver All children (aged 1-3 years) 530 621 0.85* (0.76 to 0.95) 0.006 specific micronutrients with sustained compliance Age ≤24 months 186 254 0.64* (0.53 to 0.77) 0.00 Age >24 months 344 367 0.97* (0.84 to 1.13) 0.73 Consumption of milk fortified with zinc, iron, and specific Days with high fever 2899 3099 0.93* (0.88 to 0.98) 0.005 micronutrients is associated with decreased incidence of Measles 1 8 0.12 (0.02 to 0.99) 0.05 diarrhoea, acute lower respiratory tract infections, and days Antibiotics consumed (doses) 7166 7437 0.96* (0.92 to 0.99) 0.01 with severe illness *Odds ratios.

page4of5 BMJ Online First bmj.com Research

We gratefully acknowledge the contributions of parents of children enrolled 19 Hambidge KM, King JC, Kern DL, English-Westcott JL, Stall C. Pre-breakfast plasma in the study and the study team, including health workers, supervisors, phy- zinc concentrations: the effect of previous meals. J Trace Elem Electrolytes Health Dis 1990;4:229-231. sicians, data management, and other support staff. 20 Williams RL. A note on robust variance estimation for cluster-correlated data. Biomet- Contributors: SS was the principal investigator, designed the study, formu- rics 2000;56:645-6. lated the hypothesis, directed the study and the analyses, wrote the paper, 21 Rogers WH. Regression standard errors in clustered samples. Stata Tech Bull and is guarantor. UD helped to design and develop data collection forms, 1993;13:19-23. 22 World Health Organization. Global strategy for infant and young child feeding. Geneva: designed the database management system, analysed data, and edited the World Health Organization, 2003 (WS 120). paper. GH helped to interpret the results and write the paper. JK supervised 23 Lutter CK, Dewey KG. Proposed nutrient composition for fortified complementary the field procedures, implementation of the study, data collection, data foods. J Nutr 2003;133:3011-20S. entry, and quality control. PD and AS prepared the operation manual, were 24 Idjradinata P, Pollitt E. Reversal of developmental delays in iron-deficient anaemic infants treated with iron. Lancet 1993;341:1-4. involved in design and translation of data collection forms and data collec- 25 Fischer WC, Black RE. Zinc and the risk for infectious disease. Annu Rev Nutr tion, and prepared the manuscript. VPM helped in design, interpretation, 2004;24:255-75. and writing. REB contributed to design, interpretation, and writing. 26 Brown KH, Peerson JM, Rivera J, Allen LH. Effect of supplemental zinc on the growth Funding: Fonterra Brands, Auckland, New Zealand, funded the study and and serum zinc concentrations of prepubertal children: a meta-analysis of randomized controlled trials. Am J Clin Nutr 2002;75:1062-71. provided the milk powder used in the trial. 27 World Health Organization. Complementary feeding of young children in developing Competing interests: None declared. countries: a review of current scientific knowledge. Geneva: WHO, 1998 (WHO/NUT/98.1). 28 Solomons NW, Jacob RA. Studies on the bioavailability of zinc in humans: effects of Ethical approval: Human research and ethical review committee at the heme and nonheme iron on the absorption of zinc. Am J Clin Nutr 1981;34:475-82. Johns Hopkins Bloomberg School of Public Health, and the Annamalai 29 Rossander-Hulten L, Brune M, Sandstrom B, Lonnerdal B, Hallberg L. Competitive University, India. inhibition of iron absorption by manganese and zinc in humans. Am J Clin Nutr 1991;54:152-6. 30 Fischer WC, Kordas K, Stoltzfus RJ, Black RE. Interactive effects of iron and zinc on biochemical and functional outcomes in supplementation trials. Am J Clin Nutr 1 Gross R, Benade S, Lopez G. The international research on infant supplementation 2005;82:5-12. initiative. J Nutr 2005;135:628-30S. 31 Solomons NW, Jacob RA, Pineda O, Viteri FE. Studies on the bioavailability of zinc in 2 Ramakrishnan U. Prevalence of micronutrient malnutrition worldwide. Nutr Rev man. Effects of the Guatemalan rural diet and of the iron-fortifying agent, NaFeEDTA. 2002;60:S46-52. J Nutr 1979;109:1519-28. 3 Bhaskaram P. Micronutrient deficiencies in children—the problem and extent. Indian J 32 Sandstrom B, Davidsson L, Cederblad A, Lonnerdal B. Oral iron, dietary ligands and Pediatr 1995;62:145-56. zinc absorption. J Nutr 1985;115:411-4. 4 Calder PC, Kew S. The immune system: a target for functional foods? Br J Nutr 33 Davidsson L, Kastenmayer P, Hurrell RF. Sodium iron EDTA [NaFe(III)EDTA] as a 2002;88(suppl 2):165-77. food fortificant: the effect on the absorption and retention of zinc and calcium in 5 Beck MA, Handy J, Levander OA. Host nutritional status: the neglected virulence fac- women. Am J Clin Nutr 1994;60:231-7. tor. Trends Microbiol 2004;12:417-23. 34 Sazawal S, Black RE, Bhan MK, Jalla S, Sinha A, Bhandari N. Efficacy of zinc 6 Black RE, Sazawal S. Zinc and childhood infectious disease morbidity and mortality. Br supplementation in reducing the incidence and prevalence of acute diarrhea—a J Nutr 2001;85(suppl 2):125-9. community-based, double-blind, controlled trial. Am J Clin Nutr 1997;66:413-8. 7 Guerrant RL, Lima AA, Davidson F. Micronutrients and infection: interactions and 35 Rosado JL, Lopez P, Munoz E, Martinez H, Allen LH. Zinc supplementation reduced implications with enteric and other infections and future priorities. J Infect Dis morbidity, but neither zinc nor iron supplementation affected growth or body compo- 2000;182(suppl 1):134-8. sition of Mexican preschoolers. Am J Clin Nutr 1997;65:13-9. 8 International Nutritional Anemia Consultative Group. Consensus statement. Safety of iron 36 Penny ME, Marin RM, Duran A, Peerson JM, Lanata CF, Lonnerdal B, et al. supplementation programs in malaria-endemic regions. Washington DC: International Life Randomized controlled trial of the effect of daily supplementation with zinc or multi- Sciences Institute Press, 1999. ple micronutrients on the morbidity, growth, and micronutrient status of young Peru- 9 Bhutta ZA, Black RE, Brown KH, Gardner JM, Gore S, Hidayat A, et al. Prevention of vian children. Am J Clin Nutr 2004;79:457-65. diarrhea and pneumonia by zinc supplementation in children in developing countries: 37 Sazawal S, Black RE, Jalla S, Mazumdar S, Sinha A, Bhan MK. Zinc supplementation pooled analysis of randomized controlled trials. Zinc investigators’ collaborative group. reduces the incidence of acute lower respiratory infections in infants and preschool J Pediatr 1999;135:689-97. children: a double-blind, controlled trial. Pediatrics 1998;102:1-5. 10 Allen LH. Iron supplements: scientific issues concerning efficacy and implications for 38 Cunningham-Rundles S, McNeeley DF, Moon A. Mechanisms of nutrient modulation research and programs. J Nutr 2002;132(4 suppl):813-9. of the immune response. J Allergy Clin Immunol 2005;115:1119-28. 11 World Health Organization. Iron deficiency anaemia: assessment, prevention, and control. 39 Adamson P. Vitamins and mineral deficiency: a global progress report. Ottawa, Canada/New Geneva, Switzerland: World Health Organization, 2001 (WHO/NID/01.3). York: UNICEF, 2004. 12 Bhandari N, Mazumder S, Bahl R, Martines J, Black RE, Bhan MK. An educational (Accepted 3 November 2006) intervention to promote appropriate complementary feeding practices and physical growth in infants and young children in rural Haryana, India. J Nutr 2004;134:2342-8. doi 10.1136/bmj.39035.482396.55 13 Rivera JA, Hotz C, Gonzalez-Cossio T, Neufeld L, Garcia-Guerra A. The effect of micronutrient deficiencies on child growth: a review of results from community-based supplementation trials. J Nutr 2003;133(suppl 2):4010-20. Johns Hopkins Bloomberg School of Public Health, Baltimore, MD 21205, USA 14 Torrejon CS, Castillo-Duran C, Hertrampf ED, Ruz M. Zinc and iron nutrition in Chil- Sunil Sazawal associate professor ean children fed fortified milk provided by the complementary national food program. Usha Dhingra data analyst Nutrition 2004;20:177-80. 15 Salgueiro MJ, Zubillaga M, Lysionek A, Caro R, Weill R, Boccio J. Fortification strategies Girish Hiremath postdoctoral fellow to combat zinc and iron deficiency. Nutr Rev 2002;60:52-8. Robert E Black professor 16 Beinner MA, Lamounier JA. Recent experience with fortification of foods and Center for Micronutrient Research, Department of Biochemistry, Annamalai beverages with iron for the control of iron-deficiency anemia in Brazilian children. Food Nutr Bull 2003;24:268-74. University, Chidambaram, India 17 Sazawal S, Black RE, Menon VP, Dhinghra P, Caulfield LE, Dhingra U, et al. Zinc sup- Jitendra Kumar project manager plementation in infants born small for gestational age reduces mortality: a prospective, Pratibha Dhingra senior research scientist randomized, controlled trial. Pediatrics 2001;108:1280-6. Archana Sarkar senior research scientist 18 Black MM, Sazawal S, Black RE, Khosla S, Kumar J, Menon V. Cognitive and motor development among small-for-gestational-age infants: impact of zinc supplementation, Venugopal P Menon professor birth weight, and caregiving practices. Pediatrics 2004;113:1297-305. Correspondence to: S Sazawal [email protected]

BMJ Online First bmj.com page5of5 For the full versions of these articles see bmj.com CLINICAL REVIEW

Syphilis

Patrick French

University College London, Syphilis remains common worldwide, and since the SUMMARY POINTS Camden Primary Care Trust, late 1990s infectious early syphilis has re-emerged as London WC1E 6AUA Syphilis remains a common disease worldwide, and Patrick.French@camdenpct. an important disease in western Europe, including the infectious syphilis has re-emerged in western Europe 1 nhs.uk United Kingdom. The clinical presentation of both Syphilis causes considerable morbidity and facilitates HIV early and late syphilis is diverse, and patients may transmission BMJ 2007;334:143-7 doi: 10.1136/bmj.39085.518148.BE present to a wide range of services and clinicians, The clinical presentation of syphilis is diverse, with patients including general practitioners. This review will empha- presenting to a wide range of practitioners and services sise the clinical presentation of syphilis because once A high index of suspicion of syphilis and a low threshold for syphilis has been suspected diagnosis and curative treat- testing are essential ment are usually simple to achieve. Diagnosing and treating syphilis are usually straightforward

Sources and selection criteria This review is based on Pubmed and Medline searches for syphilis (key words: syphilis, English, human) for the past five years (2000 to February 2006). I supplemented this with the literature review for the UK National Early and Late Syphilis Guidelines (2002).w1-w3

Why is syphilis important? Syphilis, caused by Treponema pallidum (box 1, fig 1), is a common infection worldwide, with an estimated 10-12 million new infections each year.w4 Early syphilis causes significant morbidity, and a systematic review of HIV transmission studies confirms that it is an important facilitator of HIV transmission.3 Congenital syphilis remains a major cause of stillbirth, childhood morbid- ity, and mortality worldwide.4 w4 The broad range of manifestations of late syphilis means that this diagnosis should be considered in a wide range of settings. SCIENCE SOURCE/SPL Fig 1 | Treponema pallidum Who gets syphilis? Syphilis is a sexually transmitted infection, and the In the late 1990s syphilis re-emerged as an important more sexual partners that individuals (or other mem- infection in western Europe. Between 1984 and 1997 bers of their sexual network) have, the more likely they acquisition of syphilis in the UK was rare,1 but since the are to acquire syphilis. Mobility, social disruption, and late 1990s a sustained epidemic of syphilis has occurred a collapse of medical services have all been recognised in homosexual men. as factors that have contributed to syphilis epidemics: In parallel to the outbreak of syphilis in homosexual the UK during the second world war; the United States men, early syphilis among heterosexual men and women with the emergence of crack cocaine use in the late in the UK has also been increasingly recognised.5 Clus- 1980s; the countries of the former Soviet Union in the ters of cases have been noted in Cambridgeshire and mid-1990s. Walsall,w6 w7 and syphilis outbreaks in south and east London (particularly associated with female commercial w8 Box 1| Characteristics of Treponema pallidum sex workers) have recently been described. • Coiled, motile spirochaete bacterium How is syphilis transmitted and classified? • Humans are its only natural host It is estimated that 30-60% of sexual contacts of • Genome sequenced, very small, circular2 individuals with early syphilis will acquire syphilis • Obligate parasite (limited metabolic capabilities) themselves.w9 w10 Entry of T pallidum probably occurs • No in vitro culture through areas of “microtrauma,” usually in mucous

BMJ | 20 JANUARY 2007 | VOLUME 334 143 CLINICAL REVIEW

Box 2 | Stages of syphilis • Primary syphilis Incubation period 2-3 weeks (range 9-90 days) Local infection • Secondary syphilis Incubation period 6-12 weeks (range 1-6 months); generalised infection • Early latent syphilis Asymptomatic syphilis of <2 years’ duration • Late latent syphilis Asymptomatic syphilis of 2 years’ duration

• Late symptomatic syphilis (tertiary syphilis) DR M F REIN/CDC | Cardiovascular syphilis, neurosyphilis, gummatous Fig 3 Rash on palms accompanying secondary syphilis syphilis which is often widespread and may also involve the scalp, palms (fig 3), and soles. Occasionally this rash membranes, and most sexual transmission of syphilis is predominantly papular, and rarely these papules probably occurs from the genital and mucous mem- ulcerate. This can be associated with generalised brane lesions of primary and secondary syphilis. The lymphadenopathy and mucosal ulceration. w11 w13 These classification of syphilis has not changed for over 100 ulcers may coalesce on the bucal mucosa, forming years and is usually described in terms of disease “snail track” ulcers, and in the genital regions (where stages (box 2, and more detail on bmj.com). there are opposing membranes) they can cause wart- like lesions called condylomata lata. These features are What is the natural course of untreated syphilis? often accompanied by constitutional symptoms such as Primary syphilis fevers and malaise. The lesion of primary syphilis occurs at the site of initial The widespread vasculitis during secondary syphi- inoculation of T pallidum. It is usually single and pain- lis may lead to a broad range of syndromes such as less but can be multiple and painful. It tends to begin as hepatitis, iritis, nephritis, and neurological problems a macule that becomes a papule, which then ulcerates. (early meningovascular syphilis) with headache and A two to three week incubation period usually occurs involvement of the cranial nerves, particularly the between the inoculation of T pallidum and development V (auditory) nerve. These complications of second- of the lesion (the range of incubation period is reported ary syphilis are relatively uncommon, occurring in as being 9-90 days). Local, non-tender lymphaden- less than 10% of individuals.w12 opathy is often associated with this lesion. Figure 2 shows primary syphilis lesions on the penis. Relapsing secondary syphilis and latent syphilis If left untreated, a lesion heals spontaneously four Individuals with secondary syphilis who do not have or five weeks later (range of healing 3-10 weeks).w12 w13 treatment improve spontaneously over three to six Because the ulcers are usually painless and can occur weeks. About a quarter of patients have relapsing at sites where they are not visible (perianally or in episodes of secondary syphilis, with recurrence of rash, the anal canal, vagina, or cervix) or not recognised mucosal ulceration, and fevers. These relapses are rare (mouth ulceration), many individuals with primary after one year and almost never occur after two years.6 syphilis do not present to services or are not diag- The infection then becomes asymptomatic (latent). nosed at presentation. Late syphilis Secondary syphilis About 35% of individuals with late latent syphilis will Four to eight weeks after primary syphilis, T pallidum develop the late manifestations of syphilis (tertiary becomes a systemic infection with bacteraemia. This syphilis).6 The three main manifestations of late syphilis secondary stage of syphilis is characterised by a gener- are neurosyphilis, cardiovascular syphilis, and gumma- alised and usually symmetrical macular papular rash, tous syphilis, and all these complications are currently rare outside resource poor countries.7

Neurosyphilis As well as being a manifestation of secondary syphi- lis, meningovascular syphilis can also occur in terti- ary syphilis. The incubation period is usually 5-12 years, and its symptoms are similar to those of early meningovascular syphilis. Parenchymatous neurosyphilis is involvement of the spinal cord (predominantly dorsal columns) and

DR G HART, DR N J FIUMARA/CDC DR G HART, brain (and occasionally both) by syphilis. The incuba- Fig 2 | Primary syphilis lesion on penis tion period of this is usually 10-20 years. The spinal

144 BMJ | 20 JANUARY 2007 | VOLUME 334 CLINICAL REVIEW

cord syndrome is called tabes dorsalis, and the brain Box 4 | Treponemal tests v non-treponemal tests syndrome is called general paralysis of the insane (see • Treponemal tests* extra boxes on bmj.com). Both syndromes remain • T pallidum particle agglutination assay (TPPA): incubation important differential diagnoses for a wide range of period† 4-6 weeks neurological presentations, including dementia, psychi- • T pallidum haemagglutination assay (TPHA): incubation 7 atric disease, and mobility problems. period† 4-6 weeks • Enzyme immunosorbant assay (EIA) IgG/IgM: incubation Cardiovascular syphilis period 3 weeks Cardiovascular syphilis usually occurs 15-30 years after • Non-treponemal tests‡ primary syphilis and may occur in any large vessel. It is • Rapid plasma reagin (RPR): incubation period 4 weeks characterised, however, by an aortitis usually affecting • Venereal Disease Reference Laboratory (VDRL): the proximal aorta. It may cause aortic incompetence incubation period 4 weeks (which may be complicated by heart failure), coronary • Serological response to yaws ostial stenosis (presenting as angina), and aortic medial • The serological response to yaws (caused by the non- necrosis causing aortic aneurysm. sexually transmitted organism T pertenue, which rarely causes serious late disease) is identical to syphilis, so in Gummatous syphilis practice most patients with suspected yaws are managed These are granulomatous locally destructive lesions as though they have syphilis which usually occur three to 12 years after primary *Usually only positive if current or past syphilis; usually positive lifelong syphilis. They can occur in almost any tissue but most after treatment commonly present when they affect skin or bone. †Incubation period is the usual time after infection that the test becomes positive The pathophysiology (particularly reasons for the vari- ‡Give a titre that acts as measure of disease “activity” (titre reduced with ation of symptomatology between individuals) of second- treatment, raised with reinfection); biological false positives occur with ary and tertiary syphilis is not clearly understood. other acute and chronic infections/autoimmune disease

Congenital syphilis in Europe, individuals who have been diagnosed with Pregnant women with syphilis can transmit the infec- HIV are at particular risk of acquiring syphilis.4 w14 All tion to the fetus. Transmission is usually transplacental patients diagnosed with syphilis must therefore be tested and is particularly likely during the first two years of for HIV, and those having follow-up for HIV must have infection. About a third of babies born to mothers with regular screening for syphilis.w14 early syphilis are born without infection and a third The clinical presentation, serological tests, and treat- with congenital syphilis; a third of pregnancies will ment response among individuals with HIV infection result in miscarriage or stillbirth. Between half a million who also have syphilis are usually the same as among and a million cases of congenital syphilis occur each individuals without HIV infection who acquire syphi- year worldwide,4 and in some resource poor countries lis,10 11 but with some variation (box 3). up to a fifth of neonatal mortality is directly attributable Some specialists recommend that a possible differ- to syphilis.8 ence in the natural course and treatment response (par- Almost all cases of congenital syphilis are easily pre- ticularly the possibility that neurosyphilis is a greater vented by antenatal screening for syphilis and treat- risk among individuals with HIV infection16) justifies ment during pregnancy.9 Even in countries where this the use of higher doses of antibiotics and longer courses is an unusual condition (such as the UK), an increase for adequate treatment. But most evidence suggests that in cases has recently been reported,5 and continuing identical management of HIV positive and negative vigilance remains vital.w5 Congenital syphilis is classi- patients is reasonable, especially in early infection.17 fied as either early or late congenital syphilis depending on whether it presents before or after 2 years of age (see What questions should be asked? extra box on bmj.com). The prognosis is particularly The diagnosis of syphilis (and the interpretation of poor if symptoms of syphilis are present in the first few syphilis serology) is often thought to be complex, but weeks after birth. diagnosis is usually straightforward. The history is guided by presenting symptoms. HIV infection and syphilis A brief sexual history may be useful to identify those As syphilis is an ulcerative sexually transmitted disease, individuals most at risk of syphilis; this is particularly individuals with syphilis are at increased risk of acquiring important in asymptomatic patients. A history of nega- and transmitting HIV. In the current syphilis outbreak tive syphilis tests (such as at sexually transmitted infec- tion clinics or at blood donor sessions or antenatal Box 3 | How syphilis affects patients with HIV screening)—as well any previous diagnosis and treat- ment for syphilis—may also be useful in evaluating • Primary syphilis: larger, painful multiple ulcers12 13 patients and interpreting positive serology. • Secondary syphilis: genital ulcers more common and higher titres with rapid plasma reagin testing and Venereal Disease Reference Laboratory testing12 13 Tests for syphilis • Possibly more rapid progression to neurosyphilis14 15 As culture of T pallidum is not possible in vitro and culture in animal models is purely a research tool,

BMJ | 20 JANUARY 2007 | VOLUME 334 145 CLINICAL REVIEW

UK guidelines for syphilis treatment, 2005 of 85-98% compared with TPHA/TPPA testing and a w17 Stage First line treatment Second line (or specificity of 93-98%. These tests may increase the alternative) treatments coverage of syphilis screening programmes by allow- Primary, secondary, early Benzathine penicillin 2.4 megaunits (intramuscular, Doxycycline 100 mg twice ing testing in settings without laboratory facilities. All latent syphilis single dose), or procaine penicillin 600 000 units daily for 14 days serological tests may be negative in incubating syphilis (intramuscular, once daily for 10 days) and early primary infection. Late latent syphilis Benzathine penicillin 2.4 megaunits (intramuscular, Doxycycline 200 mg twice Screening for syphilis is usually done with an enzyme three injections over 2 weeks: days 0, 7, 14), or daily for 28 days immunosorbant assay test. Several syphilis testing algo- procaine penicillin 900 000 units (intramuscular, daily for 17 days) rithms are available to allow the rational use of these tests (see algorithm on bmj.com).20 Neurosyphilis Procaine penicillin 2.4 units once daily Doxycycline 200 mg twice (intramuscular, for 17 days) with oral probenecid daily for 28 days Patients with symptoms or signs of possible neuro- 500 mg four times a day syphilis should have a cerebrospinal fluid examination. Most patients with neurosyphilis will have positive non- diagnosis testing depends on direct identification of treponemal tests in the cerebrospinal fluid examination, the bacterium and serological tests. as well as a raised white cell count and protein.w18

Direct tests How is syphilis treated? Identification of T pallidum (seen as a motile spiro- Penicillin was established as a highly effective treat- chaete in a saline solution) by dark ground micro- ment for syphilis long before randomised clinical trials scopy from samples taken from the genital lesions of became the norm for determining treatment efficacy. primary and secondary syphilis allows the immediate Penicillin in a variety of doses and regimens was shown diagnosis of syphilis, with a sensitivity rate of up to to cure rapidly the lesions of early syphilis and to pre- 97% being reported in a study from 2004.18 But it is vent the clinical progression of early and latent syphilis rarely feasible to perform this test outside special- to later stages of the disease.21 ist services. DNA amplification (polymerase chain Standard antisyphilis therapy rarely fails to cure the reaction) may prove to be important in the diagnosis disease, and strains of T pallidum that are intrinsically of early syphilis—with a sensitivity of 94.7% and a resistant to penicillin have not been described. The specificity of 98.6% in primary syphilis (compared table shows the current UK syphilis treatment recom- with clinical diagnosis with serological confirmation) mendations, and box 5 shows online sources of the being reported in a recent study.19 major national and international syphilis guidelines.

Serology Newer treatments Serological tests for syphilis remain the mainstay of An effective single dose oral therapy for syphilis would diagnosis. There are two groups of tests: treponemal (or be a major advance in syphilis control, and a recent specific) and non-treponemal (or non-specific). The most large prospective randomised trial suggested that 2 g important of these tests and their different and comple- oral azithromycin is as effective in treating early syphi- mentary characteristics are summarised in box 4. lis as benzathine penicillin.22 This important study will In the past five years, enzyme immunosorbant assay probably lead to the increasing use of azithromycin (EIA) tests have become established as the screening for the treatment of early syphilis, but the study find- test of first choice in syphilis.w15 These tests can be ings have been treated with some caution as macrolide automated and are generally reliable. A recent Health treatment failure is well recognised and seems to be Protection Agency assessment of 10 such tests showed associated with intrinsic macrolide resistance in some the sensitivity of nine of these tests to be 100% (con- strains of T pallidum.23 fidence interval 98.5% to 100%) with a specificity of 100% in seven tests (97% to 100%).w16 Further management and follow-up Recently, several rapid simple dipstick treponemal All individuals with syphilis should be tested for tests have been developed. These tests have sensitivities other sexually transmitted infections, including HIV. The patient’s partner(s) should be notified, but Box 5 | National and international treatment guidelines for the role of partner notification is limited in syphilis syphilis outbreaks where many partners are not identifiable World Health Organization. Guidelines for the managment or contactable.1 w19 of sexually transmitted infections. http://whqlibdoc.who. Patients who acquire syphilis are at significant risk int/publications/2003/9241546263.pdf (last updated of reinfection, so recommending regular serological 2003) screening for syphilis and providing sexual health pro- Centers for Disease Control and Prevention (US guidelines). www.cdc.gov/std/treatment/ (last updated motion are essential parts of syphilis management. 2006) International Union Against STIs (European). www.iusti. Syphilis in the future org/guidelines.pdf (last updated 2001) Syphilis is likely to remain a common disease world- Clinical Effectiveness Group, British Association for Sexual wide, and some awareness of its prevention, presen- Health and HIV (UK guidelines). www.bashh.org (last tation, diagnosis, and treatment is important for all updated 2005) clinicians. Many of the tools for effective syphilis con-

146 BMJ | 20 JANUARY 2007 | VOLUME 334 CLINICAL REVIEW

Acta Derm Venereol KEY AREAS OF FUTURE RESEARCH upon a re-study of the Boeck-Bruusgard material. 1955;35(suppl 34):1. Developing a syphilis vaccine 7 Danielsen AG, Weismann K, Jorgensen BB, Heidenheim M, Fugleholm AM. Incidence, clinical presentation and treatment of neurosyphilis in Establishing effectiveness of single dose oral therapy Denmark 1980-1997. Acta Derm Venereol 2004;84:459-62. Developing cheap, bedside diagnostic tests 8 Watson-Jones D, Changalucha J, Gumodoka B, Weiss H, Rusizoka M, Ndeki L, et al. Syphilis in pregnancy in Tanzania. 1. Impact of maternal syphilis on outcome of pregnancy. J Infect Dis 2002;186:940-7. ADDITIONAL EDUCATIONAL RESOURCES 9 Watson-Jones D, Gumodoka B, Weiss H, Changalucha J, Todd ABC J, Mugeye K, et al. Syphilis in pregnancy in Tanzania. II. The Adler M, Cowan F, French P, Mitchell H, Richens J, eds. effectiveness of antenatal syphilis screening and single dose of sexually transmitted infections. 5th ed. London: BMA benzathine penicillin treatment for the prevention of adverse Publications, 2005. pregnancy outcomes. J Infect Dis 2002;186:948-57. 10 Goeman J, Kivuvu M, Nzila N, Behets F, Edidi B, Gnaore E, et al. Similar Holmes KK, Frederick Starling P, Mardh P-A, Lemon SM, serological response to conventional therapy for syphilis among HIV- Stamm WE, Piot P, et al, eds. Sexually transmitted positive and HIV-negative women. Genitourin Med 1995;71:275-9. diseases. New York: McGraw Hill, 1999. 11 Rolfs RT, Joesoef MR, Hendershot EF, Rompalo AM, Augenbraun Sex Transm Dis MH, Chiu M, et al. A randomised trial of enhanced therapy for early Goh BT. Syphilis in adults. 2005;81:448-52. syphilis in patients with and without human immunodeficiency Hayden D. Pox: genius, madness and the mysteries of virus infection. The Syphilis and HIV Study Group. N Engl J Med syphilis. New York: Basic Books, 2003. 1997;337:307-14. 12 Rompalo AM, Joesoef MR, O’Donnell JA, Augenbraun M, Brady W, Radolf JD, et al. Clinical manifestations of early syphilis by HIV status trol (such as antenatal screening to prevent congenital and gender: results of the syphilis and HIV study. Sex Transm Dis syphilis) are already well established but have not been 2001;28:158-65. 13 Rompalo AM, Lawlor J, Seaman P, Quinn TC, Zenilman JM, Hook fully implemented in many parts of the world. EW 3rd. Modification of syphilitic genital ulcer manifestations by The likely absence of a syphilis vaccine in the fore- coexistent HIV infection. Sex Transm Dis 2001;28:448-54. 14 Johns DR, Tierney M, Felsenstein D. Alteration in the natural seeable future means that syphilis control will depend history of neurosyphilis by concurrent infection with the human largely on reducing risk taking among individuals and immunodeficiency virus. N Engl J Med 1987;316:1587-72. communities affected by syphilis and on the diagnosis 15 Berry CD, Hooton TM, Collier AC, Lukehart SA. Neurologic relapse of syphilis after benzathine penicillin therapy for secondary syphilis in a and treatment of individuals with early syphilis. Com- patient with HIV infection. N Engl J Med 1987;316:1587-9. prehensive sexual health promotion programmes have 16 Marra CM, Maxwell CL, Smith SL, Lukehart SA, Rompalo AM, 24 Eaton M, et al. Cerebral spinal fluid abnormalities in patients with been shown to reduce syphilis prevalence, as have syphilis: association with clinic and laboratory features. J Infect Dis new treatment approaches such as syndromic manage- 2004;189:369-76. ment of genital ulcer disease.25 Primary prevention, 17 Parkes R, Renton A, Meheus A, Laukamm-Josten U. Review of current evidence and comparison of guidelines for effective treatment in together with provision of easily accessible syphilis Europe. Int J STD AIDS 2004;15:73-88. diagnostic and treatment services, will remain the 18 Wheeler HL, Agarwal S, Goh BT. Dark ground microscopy and treponemal tests in the diagnosis of early syphilis. Sex Transm Infect cornerstone of syphilis control. 2004;80:411-4. I thank Debbie Sumner and Tim Gerrard for reviewing this manuscript. 19 Palmer HM, Higgins SP, Herring AJ, Kingston MA. Use of PCR in the Contributors: PF is the sole contributor. diagnosis of early syphilis in the United Kingdom. Sex Transm Dis Competing interests: None declared. 2003;79:479-83. 20 Egglestone SI, Turner AJL, for the PHLS Syphilis Serology Working 1 Simms I, Fenton KA, Ashton M, Turner KM, Crawley-Boevey Group. Serological diagnosis of syphilis. Commun Dis Public Health EE, Gorton R, et al. The re-emergence of syphilis in the 2000;3:158-62. United Kingdom: the new epidemic phases. Sex Transm Dis 21 Hahn RD, Cutler JC, Curtis AC, Gammon A, Heymann E, Johnwick JH, 2005;32:220-6. et al. Penicillin treatment of asymptomatic central nervous system 2 Fraser CM, Norris SJ, Weinstock GM, White O, Sutton GG, Dodson R, et syphilis. 1. Probability of progression to symptomatic neurosyphilis. al. Complete genome sequence of Treponema pallidum, the syphilis Arch Dermatol 1956;74:355-66. spirochete. Science 1998;281:375. 22 Riedner G, Rusizoka M, Todd J, Maboko L, Hoelscher M, Mmbando D, 3 Rottingen JA, Cameron DW, Garnett GP. A systematic review of the et al. Single dose azithromycin versus penicillin G benzathine for the epidemiologic interactions between classic sexually transmitted treatment of early syphilis. N Engl J Med 2005;353:1236-44. diseases and HIV: how much is really known? Sex Transm Dis 23 Lukehart SA, Godornes C, Molini BJ, Sonnett P, Hopkins S, Mulcahy 2001;28:579-97. F, et al. Macrolide resistance in Treponema pallidum in the United 4 Saloojee H, Velaphi S, Goga Y, Afadapa N, Steen R, Lincetto O. The States and Ireland. N Engl J Med 2004;351:154-8. prevention and management of congenital syphilis: an overview 24 Celentano DD, Nelson KE, Lyles CM, Beyrer C, Eiumtrakul S, Go VF, et and recommendations. Bull World Health Organ 2004; al. Decreasing incidence of HIV and sexually transmitted diseases in 82:424-30. young Thai men: evidence for the success of the HIV/AIDS prevention 5 Health Protection Agency, UK. 2005 STI data. www.hpa.org. program. AIDS 1998;12(5):F29-36. uk/infections/topics_az/hiv_and_sti/epidemiology/ 25 Mayaud P, Mosha F, Todd J, Balira R, Mgara J, West B, et al. Improved datatables2005.htm treatment services significantly reduce the prevalence of sexually 6 Gjestland T. The Oslo study of untreated syphilis: an epidemiologic transmitted diseases in rural Tanzania: results of a randomised investigation of the natural course of the syphilitic infection based controlled trial. AIDS 1997;11:1873-80.

ENDPIECE What is research? There are two kinds of researchers: some who are just assistants, and others whose mission is to invent. Inventions should be made in all areas, even in the humblest search for facts or the simplest experiment. Science cannot begin to exist without personal and original effort. Henri Bergson (1859-1941) in his presidential address to the Society for Psychical Research in 1913 Submitted by Amar Bhat, senior house officer, Doncaster Royal Infirmary

BMJ | 20 JANUARY 2007 | VOLUME 334 147 PRACTICE For the full versions of these articles see bmj.com

QUALITY IMPROVEMENT REPORT Redesign and modernisation of an NHS cataract service (Fife 1997-2004): multifaceted approach

Adrian Tey1, Barbara Grant2, Dawn Harbison2, Shona Sutherland2, Patrick Kearns2, Roshini Sanders2

EDITORIAL by Wood Better design of space and staffing can often waiting for more than one year for cataract sur- gery.6 7 International studies suggest that not only do 1 result in more operations and training Department of Ophthalmology, patients lose further visual acuity while on a waiting list Ninewells Hospital and Medical opportunities and reduced waiting times School, Dundee DD1 9SY but that this wait has important effects on rehabilitation 8 2Cataract Unit, Queen Margaret and life expectancy. Hospital, Dunfermline KY12 0SU Context In 2000 the UK Department of Health produced a Correspondence to: A Tey The significant improvement in quality of life measures white paper titled “Action on cataracts,” giving guid- [email protected] in elderly people after cataract surgery is undisputed ance on the reorganisation of cataract services.9 This 1-3 BMJ 2007;334:149!53 worldwide. Cataract is a common cause of visual encouraged trust managers and ophthalmic consult- doi: 10.1136/bmj.39050.520069.BE impairment, and cataract surgery is the most common ants of the United Kingdom’s national health service elective surgical procedure carried out in the United to investigate methods of reorganising their service to Kingdom.4 Thirty per cent of those aged 65 in the achieve the targets set by the government. The key end United Kingdom have visually impairing cataract.5 point was that by 2003 patients with cataract were to be Visual impairment is nationally defined as a best visual treated within six months of referral. To achieve this, acuity of less than 6/12, below the criterion required the projected increase in productivity was 43%. Con- for driving.5 sultants had the continued responsibility of maintaining Improving the quantity and quality of care for cata- quality of care and protecting training opportunities ract surgery is currently a major focus for UK govern- while increasing throughput. ment policy for ophthalmic services. As life expectancy The difficulty with keeping up with demand within increases so will the demand for cataract surgery. The the NHS led to the opening of independent sector treat- extent of this unmet need was highlighted by an epide- ment centres in England, with cataract surgery being miological survey in 2000, which provided estimates of the most common operation carried out by them.10 the shortfall by using holistic modelling.6 At that time Although the centres contributed to reduced waiting the elderly population in the United Kingdom were time; concerns remained about the absence of data on

Abstract Problem A Scottish national health service ophthalmic facility was unable mix thus allowing trainees to continue to fulfil the number of operations to cope with increasing demand for cataract surgery. required to acquire higher surgical training standards. Design Multifaceted approach to redesign hospital space to accommodate Effects of change In the same three month period 237 cataract operations a cataract unit; to invest in cataract nursing staff to allow more operations were carried out in 1997 and 374 in 2004, representing an increase of under local anaesthesia and as day cases; and to enhance input by general productivity by 60%. The waiting time for surgery decreased from more practitioners and optometrists to streamline and reduce false positive than one year to three months. The redesign resulted in almost complete cataract referrals. A prospective audit for productivity was undertaken in preoperative and postoperative assessment by nursing staff, thus freeing 2004 (two years after the redesign) and compared against the national medical time and allowing for more operations. Optometrists’ referrals cataract surgery audit data for Fife from 1997. with reports increased significantly (P<0.0001). The number of operations Setting District general hospital serving a population of 400 000 in south carried out as day cases under local anaesthesia increased, with fewer east Scotland. intraoperative complications and postoperative visits (P<0.0001). The Key measures for improvement Increasing throughput of cataract surgery number of operations carried out by trainees more than doubled, from 43 while assessing quality of care provided against predefined evidence and to 100 cases, thus improving training opportunities. Royal College of Ophthalmologists’ guidelines, and evaluating training Lessons learnt Modest capital investment in rebuilding space and in standards for ophthalmic surgical trainees against higher surgical training staff for cataract services can improve the quality and volume of cataract requirements. surgery. Enhancing existing NHS services provides for future need while Strategies for change Cataract services were redesigned to increase maintaining training standards, thus potentially obviating the need for throughput and to reduce waiting times while preserving the quality of independent treatment centres. This model could be used throughout the patient care. A secondary end point was to maintain surgical case load United Kingdom.

148 BMJ | 20 JANUARY 2007 | VOLUME 334 PRACTICE

quality, continuity of care, and management of postop- their second or third consultation with either a doctor erative complications.11 Concerns were also expressed, or a nurse, and finally the general theatre suite. Patients that the centres have detracted from NHS investment could have between two and four hospital appoint- to develop long term services for patients.12 13 Finally ments, depending on the length of wait for cataract the centres may also result in dramatic reductions in surgery. training opportunities in certain districts.10 Redesign and strategies for change Background In 1997, 863 cataract operations were carried out at the The Queen Margaret Hospital in Fife serves a Queen Margaret Hospital. Population estimates sug- population of 400 000 and carries out all cataract surgery gested that 1280 cataract operations were required per for the region. The staff component in 1997 consisted annum in a population of 400 000 for visually impair- of four consultants, two specialist registrars, three sen- ing cataract (best visual acuity worse than 6/12).16 This ior house officers, and four cataract nurses. The four implied that the hospital had to increase its throughput consultants were responsible for specialist services that by at least 50%. To tackle this problem a multidiscipli- included glaucoma, diabetes, paediatrics, and oculoplas- nary team was set up, consisting of the unit’s consultant tics. In 1997 the hospital took part in a prospective three ophthalmologists, senior cataract nursing staff, and the month national survey of cataract surgery carried out trust’s business manager. This team was in continuous by the Royal College of Ophthalmologists.14 15 The indi- communication with lead members of the local general vidual results for the hospital were comparable to those practitioner and optometry committees. of the survey. In the survey 86% of cataract surgery was carried out under local anaesthesia compared with One stop cataract clinics 87% in the hospital. Seventy seven per cent of patients We decided to pilot the feasibility of one stop cata- in the survey had phacoemulsification cataract surgery ract clinics. A grant of £20 000 (€30 000; $39 000) was compared with 67% in the hospital. At final postopera- obtained from the Scottish Executive Office in 2000 tive outcome 86% of patients in the survey achieved a to enable the pilot study to be carried out. These clin- visual acuity of 6/12 or better compared with 89% in the ics were organised so that cataract nursing staff would hospital. Listing practices were comparable, with 69% initially assess patients and then the consultant would of patients in the survey having a preoperative vision make a decision on the same day as to the necessity worse than 6/12 compared with 65% in the hospital. and suitability of the patients for day care surgery. This We therefore knew that we were in line with national would reduce the number of preoperative assessments practice and not functioning suboptimally compared and waiting time for surgery. The initial experience with other units. We also had no reason to believe and study in 2000 of 100 patients showed the clinics to that our patient population was anything other than a be cost effective. The cost of a hospital appointment to standard one. the Queen Margaret Hospital in 2000 was estimated at £68; therefore the saving of even one extra hospital Outline of problem appointment per 100 patients resulted in a saving of The Queen Margaret Hospital was beginning to £6800.17 From these clinics 85% of patients were listed experience the national trend of greater demand for for cataract surgery. The waiting time for patients to be cataract surgery from an ageing population.16 As with seen at the clinic never exceeded 12 weeks. A patient other national centres the waiting time for cataract sur- satisfaction questionnaire showed that over 97% were gery was more than a year. As a result the quality of extremely satisfied with the clinic’s service. care delivered to patients was affected. The long wait resulted in over 20% of patients having vision of count- Optometry and general practitioner support ing fingers in the cataractous eye on the day of surgery. Patients who attended the one stop cataract clinic would The long wait also resulted in changes in the patient’s previously have been routed towards cataract surgery general health, thus resulting in more preoperative from general ophthalmic clinics. The waiting time for visits for repeated assessment. Although we made no these clinic appointments varied from 6-12 months. formal assessment, we were aware that patients had Some priority was given to patients with profoundly important problems with transport for multiple hospi- reduced vision, but this was not consistent practice. tal visits while dealing with the effects of major visual Patients were referred to these clinics by their general impairment during the wait for cataract surgery. practitioners, and inclusion of an optometry report Trust management acknowledged that the unit was was variable. An optometry report was invaluable as functioning at maximum capacity, given the limit- it informed the ophthalmologist of the best corrected ations of space and medical and nursing time as well visual acuity with updated refraction. This was crucial as the patients’ circuitous journey through various in the decision to operate. In addition the report gave departments of the hospital. At this time patients were important information on other ophthalmic disease that largely assessed before and after surgery by doctors. might have been contributing towards visual impair- The patients’ journey for cataract surgery was initially ment. This would allow the patient to be appointed to through general outpatients, where they had their first the appropriate specialist ophthalmic clinic. consultation with the ophthalmologist, then day case To reduce false positive referrals and streamline cata- assessment areas and biometry clinics, where they had ract referrals we decided on a strategy that only patients

BMJ | 20 JANUARY 2007 | VOLUME 334 149 PRACTICE

Results of three month audit of consecutive cases of cataract surgery carried out in 1997 at Queen payment of £250 000, which was matched by the Margaret Hospital, Fife, and in 2004 in the second year of the cataract unit. Values are numbers Fife NHS Trust. In December 2002 a rebuild at a (percentages) unless stated otherwise cost of £500 000 (capital) was completed in a vacated Variables 1997 2004 P value area of the Queen Margaret Hospital. The new cata- No of cataract operations 237 374 — ract unit had the feasibility for complete preopera- tive and postoperative assessment and surgery. Four Retrieval rate 222 (94) 374 (100) extra cataract nursing staff, at a cost of £100 000 per Input by optometrists 124 (56) 296 (79) <0.0001 annum, were also employed (recurrent trust annual Patients aged over 85 years 53 (24) 28 (8) <0.0001 expenditure). In addition a subsection of the gen- Vision at listing: — eral theatre suite nursing staff were appointed solely Counting fingers or worse 49 (22) 29 (8) <0.0001 for ophthalmic surgery. Patient appointments were prioritised on the basis of degree of visual impair- 6/18-6/60 96 (43) 200 (54) 0.016 ment. All patients with a vision of 6/60 or worse had 6/12 or better 69 (31) 145 (39) 0.059 surgery within 12 weeks of referral. Second eye listing 84 (38) 158 (42) 0.29 Day case admission 141 (64) 344 (92) <0.0001 Data collection and statistics Local anaesthesia 191 (87) 361 (98) <0.0001 The Queen Margaret Hospital took part in the data collected by the national cataract surgery survey in Phacoemulsification 146 (67) 361 (98) <0.0001 1997, which involved 100 UK hospitals.14 15 The sur- Intraoperative complications 36 (16) 22 (6) <0.0001 vey set predefined preoperative and postoperative Grade of surgeons: data, which were collected prospectively by expe- Senior house officer 7 (3) 38 (10) 0.043 rienced senior cataract nursing staff at the Queen Specialist registrar 36 (16) 62 (17) — Margaret Hospital. Identical data were collected Associate specialist 22 (10) 0 — prospectively in 2004 using the same methodology as the national audit. The data were entered and Consultant 154 (69) 270 (72) — analysed using Microsoft Office Excel and Micro- Cancelled on day 3 4 — soft Office Access. Statistical analysis was carried out Postoperative visits: using a two sample t test. ≥2 32 (15) 312 (84) <0.0001 Effects of change 1 13 (6) 91 (25) <0.0001 Process outcome Postoperative vision 6/12 or better 194 (89) 343 (93) 0.089 The table outlines the results from a prospective three month survey of consecutive cases of cataract with optometrists’ reports would be seen at the one surgery carried out in 1997 at the Queen Margaret stop cataract clinic. The local general practitioner and Hospital and again in 2004 in the second year of the optometry committees were canvassed for their sup- cataract unit. Overall, 863 cataract operations were port with oral presentations and written information carried out in 1997 compared with 1473 in 2004, on the new clinic’s service. Both parties were encour- representing an increase of 71%. Of the 15 Scottish aged to send an optometrist’s report with their cataract health boards, Fife achieved the highest increase in referrals. rates of cataract surgery between 1997 and 2004.18 In 2004 there was significantly more optometric input Specialist cataract nurses into cataract referrals (P<0.0001). The number of On the basis of our pilot data with the one stop clinic operations carried out by trainees more than doubled it was apparent that patients’ care and throughput between 1997 and 2004. were also enhanced by continuity of nursing assess- ment for patients before and after surgery. For this Clinical outcome reason we sought to deploy a fixed pool of nurses Significantly more patients were aged 85 or more in for this process. 1997 at the time of cataract surgery (P<0.001). Over 20% of patients had profoundly reduced vision at Theatre journey counting fingers before cataract surgery, and this Despite all the advantages of the one stop clinic we was significantly more than in 2004 (P<0.001). More recognised that the patient’s journey on the day of operations were carried out as day cases with local surgery was suboptimal. A general theatre suite was anaesthesia using the phacoemulsification technique being used and staffed by non-ophthalmic staff. This (P<0.001). Significantly fewer intraoperative compli- meant that the handing over and transfer procedure cations (posterior capsule rupture, vitreous loss, zonular was an important limitation in patient turnover time dehiscence, and iris prolapse) led to fewer postoperative between operations. complications and postoperative visits before discharge in 2004 (P<0.0001). More patients achieved a postop- Redesign build erative visual acuity of 6/12 or better in 2004 compared After the success of our one stop clinic the Scot- with 1997. Neither group had cases of endophthalmitis, tish Executive Office granted the trust a one-off retinal tear, or return to theatre.

150 BMJ | 20 JANUARY 2007 | VOLUME 334 PRACTICE

side contractors would have been £360 000 at a costing Cataract surgery of £600 per operation. Thus over two years we have recouped the capital cost of the rebuild.

Old pathway New pathway Referral with or without optometric report Referral with optometric report Medical staff and training issues No net increase occurred in senior medical oph- General clinic One stop cataract clinic thalmology staff during this period. The increased surgical training opportunities attracted an extra spe- Cataract theatre cialist registrar. High turnover has not been at the Preassessment Biometry

Three month wait, two staff expense of compromising training or surgical stand- clinic clinic ards. Over three months in 1997 trainees carried out 43 operations compared with 100 in 2004, thus more One year wait, four staff Further assessment clinic if wait > 1 year than doubling our training opportunities. Our goals were helped by the almost complete conversion to General theatre suite small incision phacoemulsification cataract surgery resulting in fewer intraoperative complications and postoperative visits. Changes over time for cataract Lessons learnt Vision at time of listing for surgery is the single most surgery pathway Streamlining referrals, general practitioners, and opto- important feature in threshold for cataract surgery.16 metric input Studies have estimated that 1280 cataract operations are We found that streamlining cataract referrals sig- required per annum for a population of 400 000 with nificantly reduced waiting times (figure). The gen- a listing vision of 6/18 or worse. We have exceeded eral practitioners and optometrists were canvassed this number and are therefore now in a position to and were in full support of providing an updated increase our trainee numbers and offer our services to optometry report. This was crucial in reducing false neighbouring trusts. positive referrals. Twenty per cent of patients still had no optometry report. We believe this was because of Improvements in quality of care a small group of transient trainees in both general The redesign has meant that quality of care has practice and optometry who are not familiar with improved for patients requiring cataract surgery in local referral practices. This problem has been cir- Fife. The waiting time for cataract surgery decreased cumvented by the cataract nurses telephoning the from 72 to 14 weeks in 2004. Twenty per cent of our patient, general practitioner, or optometrist for an patient cohort were aged 85 years or more and 20% updated optometry report. had a preoperative vision of counting fingers before cataract surgery in 1997 compared with less than 8% Specialist nurse role for both indices in 2004. We interpret this as elderly The recruitment of nurses who specialise in cataract people in Fife now having cataract surgery earlier surgery, working to protocol in accordance with Royal in their lifespan with fewer presentations of dense College of Ophthalmologists guidelines, has dramati- cataracts. cally increased the numbers of patients assessed for We strive towards continuous quality improvement. cataract surgery.19 Postoperative care has traditionally This year we achieved a further target—patients hav- been the remit of medical staff, but in our centre this ing cataract surgery 18 weeks from their optometry was almost completely devolved to specialist nurses. In visit. We hope soon to pilot electronic optometric turn this allowed more operations to be carried out. referral directly to the one stop cataract clinic. We audit all aspects of our cataract surgery and this year Purpose built cataract unit we were able to show the trust our performance data The redesign helped shorten the patients’ journey and secure additional theatre sessions. through the hospital thus allowing for more operations We have shown that modest, long sighted invest- in the same theatre session. Critical to this was the time ment in the NHS can have a dramatic effect on the previously lost in between operations by the patients’ quality and throughput of cataract surgery while travel and transfer between different groups of nurses. maintaining surgical training standards and consoli- The cost of tendering the 600 extra cataract opera- dating continuity of care without recourse to provi- tions (carried out in 2004 compared with 1997) to out- sion of care from sources outside the NHS. Patients with cataracts may benefit from surgery earlier in KEY LEARNING POINTS their lifespan and before they reach severe visual • Increased productivity for cataract surgery can be impairment. This study gives an example to other achieved by streamlining referrals with optometric input, trusts of excellence that can be achieved within the specialist nurse clinics, and space rebuild NHS providing cataract surgery, the commonest elective surgical procedure in elderly people.4 • Modest NHS investment in redesign can increase throughput for cataract surgery by 60%, with maintenance Contributors: AT and RS proposed the study. AT and BG carried out data of quality of care and training opportunities collection and interpreted the data. BG, DH, SS, PK, and RS formed part of the redesign committee and think tank for the redesign of the cataract service.

BMJ | 20 JANUARY 2007 | VOLUME 334 151 PRACTICE

Funding: None. 9 UK Department of Health. Action on cataracts: good practice guidance. Competing interests: None declared. Jan 2000 www.dh.gov.uk. 10 Guly C, Sidebottom R, Hakin K, Bates K. Treatment centres and their 1 Raferty J, Stevens A, eds. Cataract surgery. London: NHS Executive, effect on surgical training. BMJ 2005;331:1338. 1994. 11 Ferris JD. Independent sector treatment centres (ISTCS): early 2 Desai P, Reidy A, Minassian DC, Vafidis G, Bolger J. Gains from experience from an ophthalmology perspective. Eye 2005;19:1090-8. cataract surgery: visual function and quality of life. Br J Ophthalmol 12 Harrad R. Local funding would reduce waiting lists for cataracts. BMJ 1996;80:868-73. 2002;325:1033. 3 Brenner MH, Curbow B, Javitt JC, Legro MW, Sommer A. Vision change 13 Kelly SP. Recurring policy errors: blind spots over cataracts. Lancet and quality of life in the elderly. Response to cataract surgery and 2005;366:1691. treatment of other chronic ocular conditions. Arch Of Ophthalmol 14 Desai P, Reidy A, Minassian DC. Profile of patients presenting for 1993;111:680-5. cataract surgery in the UK: national data collection. Br J Ophthalmol 4 UK Department of Health. Hospital activity and episode statistics. www. 1999;83:893-6. doh.gov.uk/public/stats1.htm. 15 Desai P, Minassian DC, Reidy A. National cataract surgery survey 1997- 5 Reidy A, Minassian DC, Vafidis G, Joseph J, Farrow S, Wu J, et al. 8: a report of the results of the clinical outcomes. Br J Ophthalmol Prevalence of serious eye disease and visual impairment in a north 1999;83:1336-40. London population: population-based, cross sectional study. BMJ 16 Taylor H. Cataract: how much surgery do we have to do? Br J Ophthalmol 1998;316:1643-6. 2000;84:1-2. 6 Minassian DC, Reidy A, Desai P, Farrow S, Vafidis G, Minnassian A. The 17 Fife Acute Hospitals Trust. http://intranet.faht.fife. deficit in cataract surgery in England and Wales and the escalating 18 Information and Statistics Department, Scotland. www.isdscotland. problem of visual impairment: epidemiological modelling of the population dynamics of cataract. Br J Ophthalmol 2000;84:4-8. org/isd. 7 Gray CS, Crabtree HL, O’Connell JE. Waiting in the dark: cataract surgery 19 Royal College of Ophthalmologists. Cataract surgery guidelines. in older people. BMJ 1999;318:1367-8. London: RCO, 2004. 8 Jakko L, Laatikainen L. The decrease of visual acuity in cataract patients waiting for surgery. Acta Ophthalmol Scand 1999;77:681-4. Accepted: 3 November 2006

BMJ UPDATES Web based monitoring improves glycaemic control in type two diabetes

Research question What did they find? Can web based enhancements to self monitoring Adults using the internet based system of monitoring

improve glycaemic control for patients with type 2 and feedback had lower mean concentrations of HbA1c diabetes? during the study than did controls (mean over the whole study period 6.9% v 7.5%, P=0.009). They were also Answer more stable, as measured by the standard deviation of

Yes. Glucose monitoring and feedback via the web help each individual’s results (HbA1c fluctuation index 0.47 v

reduce concentrations of haemoglobin A1c compared 0.78, P=0.001). The benefits were evident in subgroups

with traditional face to face consultations. with both higher HbA1c concentrations (!7%) and lower concentrations ("7%) at baseline. Why did the authors do the study? Tight glycaemic control helps prevent complications What does it mean? in patients with type 2 diabetes, who mainly combine This relatively small trial suggests that self monitoring self monitoring with visits to a diabetic clinic or primary combined with regular feedback over the internet care doctor. These authors wanted to test a new system can help patients with type 2 diabetes improve their based on monitoring and feedback via the internet. glycaemic control—possibly because the web based system gave them faster and more timely access to advice What did they do? from doctors and dieticians. Glycaemic control was better Eighty Korean adults with type 2 diabetes took part in a for the intervention group from about three months, randomised controlled trial over two and a half years. All and the difference persisted for the full two and half participants monitored their own serum concentrations years of the study. Most of the time the doctors simply of glucose regularly at home, and visited a diabetic gave encouragement, but about 12% of web based outpatient clinic every three months for blood tests interactions led to changes in drug regimen. Another 12% and advice. Forty of them (the intervention group) also were about lifestyle issues. uploaded their results regularly on to a website. “Clinical All the participants in this study had access to the instructors” accessed the website daily and gave internet and knew how to use it. The system may not work feedback and advice to each participant fortnightly. so well for patients who are less computer literate.

The authors compared glycaemic control between Cho et al. Long-term effect of the internet-based glucose monitoring

intervention and control groups over 30 months, system on HbA1c reduction and glucose stability. Diabetes Care using intention to treat analysis. They also looked at 2006;29:2625-31 the stability of control using a fluctuation index—the This summarises a paper that has been selected by bmjupdates. standard deviation of each participant’s haemoglobin To register for bmjupdates (free email about high quality new papers in your favourite subjects) go to http://bmjupdates.com/ A1c (HbA1c) concentrations.

152 BMJ | 20 JANUARY 2007 | VOLUME 334 PRACTICE

10!MINUTE CONSULTATION Collapse with loss of awareness

Dougall McCorry,1 Angela McCorry2

1 A 24 year old woman attends your surgery after having Department of Neurosciences, The DIFFERENTIATING SEIZURE FROM SYNCOPE Walton Centre, Liverpool L9 7LJ collapsed and blacked out the day before at the department 2Mather Avenue Surgery, Liverpool store where she works as a sales assistant. Colleagues told Unhelpful features—often mistakenly thought to indicate seizure but can occur in syncope Correspondence to: D McCorry her that she had twitched and jerked. [email protected] • Twitching and jerking Incontinence (reflects a full bladder at the time of the event) BMJ 2007:334:153 General points • doi:10.1136/bmj.39070.390961.DE • Vasovagal syncope (simple faint) accounts for most • Pallor collapses with loss of awareness in patients of good • Bitten tip of tongue general health. Vasovagal syncope can be confused • Fatigue after the event with other, more serious causes of collapse with loss Helpful features—indicate a seizure of awareness, such as seizures and cardiac syncope • Confusion after the event lasting >2 minutes • Accounts of the incident from witnesses, as well as • Deeply bitten lateral border of the tongue from the patient, are important in reaching a diagnosis • Tonic then clonic movement lasting >1 minute • A collapse is a dramatic event, and terms such as • Deep cyanosis “fitting” and “seizure” may be used. It is important to get a description rather than a label • Develop a mind’s eye view of events before, during, What you should do and after the attack Examination—Patients who have had vasovagal syncope or a seizure usually have no abnormal examination findings. What issues you should cover The purpose of examination is to reassure the patient and What was the patient doing when it happened?—To to look for signs of cardiac disease that indicate cardiac make a confident diagnosis of vasovagal syncope, syncope. Check pulse, lying and standing blood pressure, expect a precipitating factor such as prolonged standing, and heart sounds. dehydration, heat, fear, or pain. Cardiac syncope and Investigation—In cases of vasovagal syncope investigation seizures are usually spontaneous; collapse during exercise results should be normal. Only occasionally is there an is a “red flag” sign for cardiology referral. underlying cause. Consider checking full blood count What was the patient’s last memory?—Expect a and urea, electrolytes, and glucose concentrations and prodrome in vasovagal syncope. Common precursor doing a pregnancy test. The guidelines of the UK National symptoms are light headedness, feeling hot, nausea, and Institute for Health and Clinical Excellence suggest dizziness. Just before losing consciousness the patient may electrocardiography in all collapses with loss of awareness experience dulled hearing and changed vision (blurring, to check for abnormalities such as arrhythmias and long QT spots, or dull vision). Cardiac syncope usually has no syndrome. warning, although palpitations or light headedness may Management—A clear explanation and reassurance are be recalled. Similarly, seizures often have no warning, important, as vasovagal syncope can be frightening. Advise although they can be preceded by an aura such as a déjà vu her to lie down should warning symptoms occur again and experience or a rising sensation in the abdomen. to avoid the precipitants. If you suspect cardiac syncope What did the eyewitnesses see?—Vasovagal syncope or seizure then refer her to the appropriate specialist and occurs in a standing or sometimes a sitting position. It is advise her not to drive until she has been seen. associated with pallor and brief loss of consciousness. Twitching and jerking are often seen; this can be prolonged USEFUL READING if during the collapse the head remains above the level of the heart. Falling stiffly followed by sustained, rhythmical Lempert T. Recognising syncope: pitfalls and surprises. Journal of the Royal Society of Medicine 1996;89:372-4 jerking of all the limbs indicates a tonic-clonic seizure. A National Institute for Health and Clinical Excellence. collapse with loss of awareness but no motor phenomena The epilepsies: the diagnosis and management of would be highly unlikely to be a seizure. Pallor and brief the epilepsies in adults and children in primary and twitching may be seen in cardiac syncope. secondary care. www.nice.org.uk/page.aspx?o=CG020 What was the patient’s first memory on coming (Appendix A, “Differential diagnosis of epilepsy in round?—Tonic-clonic seizures are followed by a period of adults and children,” and Appendix E, “Key clinical confusion, disorientation, and amnesia. Determine this questions”) from a witness and clarify the patient’s first clear memories. After vasovagal syncope individuals may feel sick and This is part of a series of occasional articles on common problems in sweaty, then tired for many hours, but immediately on primary care coming round they can recollect conversations and events The BMJ welcomes contributions from general practitioners to the that took place before the event. series

BMJ | 20 JANUARY 2007 | VOLUME 334 ` 153 PRACTICE

CHANGE PAGE Don’t use minocycline as first line oral antibiotic in acne

Paul McManus, Ike Iheanacho

Drug and Therapeutics Bulletin, The clinical problem KEY POINTS BMA House, London WC1H 9JR Most people develop acne vulgaris at some point in life. ! Oral minocycline is no more effective than other oral Correspondence to: P McManus Typified by inflammation of hair follicles and accompanying [email protected] tetracyclines in treating acne References and search methods are sebaceous glands, acne may require oral antibacterial ! The risk of rare but serious unwanted effects with on bmj.com. treatment, particularly in those with moderate to severe minocycline makes it less suitable for use than other Competing interests: None declared. disease (an estimated 11% of adolescents1). Minocycline is drugs in its class a common choice of drug and is often wrongly assumed to be ! Oral minocycline is more expensive than most other oral BMJ 2006;334:154 tetracyclines doi: 10.1136/bmj.39048.540394.BE more effective, easier to take, and less likely to cause bacterial resistance than other tetracyclines. ! Patients who need treatment with an oral tetracycline should be prescribed doxycycline, lymecycline, or The evidence for change oxytetracycline Efficacy A Cochrane review assessed 27 randomised trials, involving a hypersensitivity reactions, and benign intracranial total of 3031 patients with acne vulgaris affecting the face or hypertension.6 However, minocycline seems to be unique in upper trunk, which compared oral minocycline with placebo causing potentially irreversible slate-grey hyperpigmentation or other active treatment.2 Although minocycline seemed to of the skin. The drug also seems much more likely than other be an effective treatment for acne, there was no convincing tetracyclines to lead to lupus-like syndrome. evidence from good-quality studies that it was superior to other oral antibacterials. Convenience These findings were echoed in a pivotal randomised trial The once daily administration of minocycline and the fact involving 649 patients with mild to moderate acne randomised that it need not be taken on an empty stomach are theoretical to one of the following regimens: oral minocycline, oral advantages over oxytetracycline and tetracycline.7 However, oxytetracycline, topical benzoyl peroxide, or topical benzoyl lymecycline and doxycycline are also taken once daily, and peroxide plus topical erythromycin.3 At 18 weeks, the only their absorption is not affected by food.7 significant difference between treatments was that more participants rated themselves at least moderately improved Cost with benzoyl peroxide plus erythromycin (the regimen with the The cost to the NHS of six months’ treatment with minocycline highest response rate) than with minocycline (which had the 100 mg daily (the licensed dose) as either tablets or modified lowest response rate; 66% v 54% of patients, odds ratio 1.74 release capsules is around £69 (€103;$134). By comparison, (95% confidence interval 1.04 to 2.90)). six months’ treatment with doxycycline 50 mg once daily costs Skin colonisation by erythromycin resistant propionibacteria £27, lymecycline 408 mg once daily costs £46, oxytetracycline did not affect reported response to erythromycin based 500 mg twice daily costs £40, and tetracycline 500 mg twice treatments, but, crucially, colonisation with tetracycline daily costs £114. resistant propionibacteria reduced the effectiveness of both minocycline and oxytetracycline. We can find no published Barriers to change evidence to support claims that minocycline is less likely than Years of effective marketing and preferential use in secondary other tetracyclines to cause propionibacterial resistance or care have helped to establish minocycline as the oral Change page aims to alert that switching to minocycline from another tetracycline will antibacterial of choice for acne. Despite serious unwanted clinicians to the immediate improve response. effects, the drug remains widely prescribed. The British need for a change in practice to make it consistent with current Further evidence comes from a randomised trial involving National Formulary singles out minocycline as offering “less evidence. The change must 134 patients with acne, which reported no significant likelihood of bacterial resistance,” without discussing its be implementable and must difference between modified release minocycline and unfavourable risk to benefit profile in acne.8 Unawareness of offer therapeutic or diagnostic 4 advantage for a reasonably lymecycline in the mean reduction of inflamatory lesions. the lack of evidence for preferring minocycline, coupled with common clinical problem. Evidence that oral minocycline might be more effective than practitioners’ and patients’ experience of the drug’s undoubted Compelling and robust evidence other tetracyclines is, at best, weak, being limited to a few, efficacy in acne, represent the greatest barriers to change. must underpin the proposal for 2 5 change. poor quality trials with highly questionable results. Flaws in these studies include a lack of blinding, failure to specify the How should we change our practice? Series editor: Joe Collier power of the study, reporting of data for only a proportion of Minocycline should no longer be prescribed as the first line ([email protected]), professor of medicines policy, St participants, only graphical presentation of data, and failure to oral tetracycline for patients with acne. There is no compelling George’s Hospital and Medical cite confidence intervals for key results. evidence that it is more effective or less likely to produce School, London. Anyone wishing antibacterial resistance than safer, less expensive tetracyclines to propose a change in clinical 7 practice should discuss the Unwanted effects that are just as easy for patients to take. Where such treatment proposal with Joe Collier at an Like other tetracyclines, minocycline can cause is indicated, doxycycline, lymecycline, or oxytetracycline is a early stage. gastrointestinal upset, candidiasis, photosensitivity, much better option.

154 BMJ | 20 JANUARY 2007 | VOLUME 334 Pompous and patronising? p 157 VIEWS & REVIEWS

Who is responsible for “do not resuscitate” status in patients with broken hips? PERSONAL VIEW Rahij Anwar, Azeem Ahmed

ip fractures are on the rise in hand, find themselves the ageing population of the in a “tight spot” as United Kingdom. It has been they are unclear of the estimated that about 30-40% of consequences of such these patients die within the first a decision. Some argue Hyear. Although actual perioperative causes of that the “do not resus- mortality are difficult to ascertain, it is gen- citate” status is usually erally believed that a vast majority of early misunderstood as “dis- deaths occur as a result of medical condi- continuation of active tions such as chest infection and fluid and treatment,” especially electrolyte imbalance. Better outcomes have by the nurses, which been reported in centres where the manage- significantly compro- ment of these patients is carried out jointly mises patient care. under the medical and orthopaedic teams. There are also fears However, lack of resources, poor co-ordina- that such an approach L P S

tion between the orthopaedic surgeons and may become indefen- / D T L medical teams, and ever increasing patient sible in a court of law. G N I load often affect the quality of treatment that Looking at this G A M I

these frail patients deserve. problem from the rela- L A C I

When elderly patients with hip fractures tives’ perspectives also D E M

become acutely unwell during the periop- raises certain issues E N A C erative care and their chances of survival which can often prove U seem bleak, the whole medical workforce quite challenging for D including orthopaedic surgeons, anaesthet- doctors. Most relatives, when approached for relatives. It must be remembered that the tim- ists, and physicians unanimously agree that a decision, assume that the patient is likely ing of this decision is extremely important, cardiopulmonary resuscitation should not be to die, although this may not always be the for any delay will result in confusion and mis- performed in the event of an arrest. How- case. They also, therefore, regard the “do not interpretation, which can seriously compro- ever, none is ready to make a firm decision resuscitate” status as a decision not to treat. mise the quality of treatment. This decision in relation to individual patients and as a As a result, some of them have even objected should be clearly documented in the medical result sometimes patients as old as 95 years to the administration of intravenous fluids or notes, and all consent forms for hip opera- are subjected to the trauma of cardiopulmo- antibiotics to patients after they had agreed to tions should have a statement confirming the nary resuscitation, which they the “do not resuscitate” status. resuscitation status of the patient. This can would never have agreed to if Physicians and Such misinterpretations are also be universally applied to all other opera- they were mentally competent orthopaedic surgeons often the result of poor com- tions in elderly patients. This would ensure to make a final decision. should join hands munication, which can easily that each patient gets what he or she (or the The physicians, who get to provide the best be improved. Rarely the rela- relatives) has opted for and active treatment only temporarily involved in possible care to these tives may mislead the doctors, would continue to be delivered irrespective their care for many reasons, especially if they have vested of the resuscitation status. are not always willing to elderly patients interests. Physicians and orthopaedic surgeons establish the “do not resusci- All elderly patients should should join hands to provide the best pos- tate” status themselves. They strongly believe make an informed decision regarding resus- sible care to these elderly patients, who need that this decision is actually the responsibility citation status in the presence of a doctor, much more than just a simple hip operation. of the admitting orthopaedic team, who are on admission. If the patient is too mentally Rahij Anwar is specialist registrar, South East Thames considered to be in overall charge of the care confused to give informed consent, this deci- Region [email protected] of these patients. sion should rest with the team responsible Azeem Ahmed is senior house officer, West Middlesex The orthopaedic surgeons, on the other for the patient’s care, in consultation with the University Hospital, Middlesex

BMJ | 20 JANUARY 2007 | VOLUME 334 155 VIEWS & REVIEWS

NETLINES Referral management schemes are The excellent site www.prodigy.nhs. uk has been around for some time and is well known, but for those who don’t damaging patients’ interests know it this is a wonderful knowledge base. The topics covered are now called PERSONAL VIEW Peter Lapsley clinical knowledge summaries, and 200 topics are covered in depth. This database forms an excellent online ome years ago I visited my general hospitals, alongside their consultant colleagues) textbook, particularly for primary care practitioner to ask why the beds of or, more usually, by GPs with a special interest professionals. There is evidence of a most of my fingernails and toenails in the specialty concerned, whose knowledge recent update, and the site is easy to were thickening and whitening, and experience are extremely variable. In navigate. Importantly, the data provided raising the nails, which were many cases GPs are being offered financial are practical and designed for frontline Sthemselves becoming heavily pitted. She incentives to participate in the schemes. primary care. This is a gem of a resource and deserves to be around for many more admitted that she did not know and referred Where dermatology is concerned—and years to come. me to a consultant dermatologist. this almost certainly holds true for other The consultant diagnosed nail psoriasis, specialties—referral management schemes An impressive medical biochemistry resource that is effectively an online explaining that it was difficult to treat, and pose a serious threat to patients’ interests. textbook is at http://web.indstate. prescribed a protracted programme of hand They introduce an extra step in the patient’s edu/thcme/mwking/home.html. and foot PUVA (psoralen and ultraviolet A journey, delaying the diagnosis and treatment Called the Medical Biochemistry Page, treatment). Although this did not cure the of often complex and difficult skin diseases. it is linked with a university course. psoriasis, it improved it considerably and What is more alarming is that some primary It has an impressive subject list, the halted the psoriatic arthritis that was beginning care trusts now deliberately delay outpatient content of which is accessed by clicking to affect the joints of two of my fingers. appointments, refusing to fund routine paper on simple hypertext links. There is a Had that episode occurred this year in one referrals seen within eight weeks of the date good smattering of illustrations, and the volume of information is impressive. of the many primary care trusts that have of the referral letter. In contrast, patients who introduced a referral management scheme, can be booked into clinics directly through the The history of medicine makes a rich and it is most unlikely that I would have had an Choose and Book electronic booking service interesting story, but it can be difficult appointment with the consultant. can be seen within two to three weeks, no to find resources, especially when you Referral management schemes, otherwise matter what their complaint. are on your travels. The website www. historyofbiologyandmedicine.com known as clinical assessment and treatment Furthermore, the schemes remove is an appealing offering. It is based schemes (CATS) or tier 2 services, are any vestige of “patient choice,” another on a book written in 1986 and was springing up across the NHS as a means of government mantra. subsequently updated much later reducing primary care trusts’ spending on Dermatology is a complex specialty with by another author. Read the most secondary care services. In summer 2006 the more than 1000 potential diagnoses. Although interesting and well written introduction British Association in Britain about 15% of GPs’ consultations on the home page. From here you can Some primary of Dermatologists relate to skin disorders, the average find the medical and biological places care trusts now conducted an informal undergraduate curriculum has only six days of of historical interest in a number of countries, including France, Italy, the deliberately internal survey of dermatology, and only 20% of GP vocational United Kingdom, and the United States. delay outpatient its members. More training schemes include a dermatological appointments than half (55%) of component. Practice nurses receive no such It is not always easy to obtain good respondents said that training. Referral management schemes quality advice and information when looking for ethical guidance. For those their primary care trusts had plans for their therefore create a real risk that patients with in the UK, at least, there is one helpful referral management scheme to reduce the skin diseases will be seen by clinicians who lack port of call: www.rcgp.org.uk/extras/ number of referrals to secondary care by the necessary training and experience, greatly ethics/query.asp. This ethical guidance rerouting them back into primary care. reducing the likelihood of prompt and accurate database produced by the Royal College The justification given for the introduction of diagnosis, not least in respect of skin cancer. of General Practitioners is free to access the schemes is that they bring services “closer The schemes are also insulting to GPs, and contains both a search engine to to home”—a mantra repeated often by the second guessing their decisions. They try to locate the information you need government at present. Trust managers admit undermine the viability of secondary care and a simple hypertext alphabet index. Although primarily of interest to a British privately that the true purpose of the schemes dermatology, which is an essential component audience, the site has a lot of advice and is to reduce costs in the face of the budget of a coherent, integrated service. And they wisdom waiting to be tapped. deficits so many of them are confronting. remove any incentive for secondary care Typically, such schemes require that 80% of specialists to support or develop the role of the Harry Brown general practitioner, GPs’ referral letters be reviewed in primary GP with a special interest in dermatology. Leeds [email protected] care and that 60% of cases should be retained The schemes may provide a short term We welcome suggestions for websites to within the trust. In some cases the reviews are solution to a short term financial problem. The be included in future Netlines. Readers conducted by consultants contracted by the risk, though, is that they will do lasting damage. should contact Harry Brown at the above trust; in others they are conducted either by Peter Lapsley is chief executive of the Skin Care Campaign, email address clinical assistants (GPs who work part time in Highgate, London [email protected]

156 BMJ | 20 JANUARY 2007 | VOLUME 334 VIEWS & REVIEWS

REVIEW OF THE WEEK Shopping for patients on the high street Sending doctors out on the streets of our most unhealthy cities to cure the sick? The result is pompous and patronising television, finds Margaret McCartney

Street Doctor The idea of this series is to send doctors out onto the Even some men who are at work despite the fact that BBC One, Thursdays at streets of Britain to meet members of the public wher- they think they have flu must be questioned and exam- 7 pm ever they happen to be—in shopping centres, at their ined, apparently to make sure they have nothing sin- Rating: 0 stars office, on the bus home—to diagnose their illnesses ister wrong with them. In other words, even if you are and treat them. The BBC press release says that these sensibly at work, self managing your mild viral illness, “unique open air surgeries offer instant reassurance and you may not be safe unless a doctor checks you for the treatment to patients with a wide number of common conditions you may have but don’t yet know about. complaints and help those unsuspecting members of You may have been given a false alarm about your the public who may have been previously unaware health in front of the BBC audience, but you should they had a health problem.” be grateful for all the conditions that you have been The four “street doctors” are visiting some of our reassured you (probably) don’t have. “unhealthy cities” (the first two programmes are in Then there is the medical advice. Back pain can be Liverpool and Glasgow), where “there are countless sorted “once and for all” with the help of a “physiother- people living with undetected illness and health prob- apist, osteopath, or chiropractic,” even though the evi- lems.” But do not fear, for “it’s up to the dence that the second and third help long street doctors to turn the city around.” The Are people in such term back pain is lacking. And before we doctors carry large signs announcing their dire need of medical even get to the medical advice, the dull presence and are furnished with a patient’s care that they but important things—history taking, clini- trolley, skeleton, and luminous orange should interrupt cal examination, and a review of previous medical bags. They unashamedly tout for investigations—are either abbreviated or business by quizzing innocent passers-by. their shopping? absent. Even before we get to that stage, (“Anyone got any medical problems?”) how possible is it for the “patients” to give The doctors, with film crew, take abbrevi- fully informed consent? Everything is prey ated medical histories and do partially or completely in the Big Brother era, but we don’t have to degrade fur- clothed examinations in full view of whoever is pass- ther the concept of confidentiality. ing. As the voiceover proudly proclaims: “Nothing’s too Are people in such dire need of medical care that private, nothing’s too personal.” they should interrupt their shopping, watching horse In television this might be true. The more personal, racing, or minding their own business to be seen by a dramatic, private, and preferably tortured the revela- street doctor (and television crew)? No. In fact, many of tions, the better the expected ratings. Hence people the people who featured had already seen their doctors are warned that they “might” have diabetes before an for their various conditions. Chronic diseases have a anxious wait reveals that they don’t. Young women are habit of lasting a long time, after all. A woman with a “Anyone got any medical cautioned that they “could” have tuberculosis, before long term ear condition—who, she told us, had been in problems?” a chest radiograph is subsequently declared normal. hospital for treatment with intravenous antibiotics and had undergone several surgical procedures—was seen by a street doctor. He had the idea of taking a swab and testing it for antibiotic sensitivity, which we were told resulted in the problem being cleared up for the first time in 13 years. Imagine, it took a BBC programme to think of that! Except that I rather doubt it was a new suggestion: what we see, of course, is the edited, cut, spliced, and reassembled version of events, especially made for short attention spans. The perky aura of quick fixes and instant solutions fizzes effusively from the show. Viewers, unfortunately, might just believe it. We are left believing that doctors and nurses in Glasgow and Liverpool have been doing nothing useful till now and thus hail the heroic street doctors who have arrived to save their city. Except that they have left town. This programme is pompous and patronising. Margaret McCartney is a general practitioner, Glasgow [email protected]

BMJ | 20 JANUARY 2007 | VOLUME 334 157 VIEWS & REVIEWS

Law and disorders

FROM THE We have no body armour, no air cover, no weapons— This is a “bad” idea. Firstly, it is simply against notions FRONTLINE only our wits to protect us. General practice is door of liberty that someone who has committed no crime to door, street fighting medicine. Personality disorders should be locked up without trial or hope of release. Sec- Des Spence are our stock in trade. The dependent, the paranoid, ondly, it is based on the religion of “risk assessment.” the histrionic, the borderline—we jolly them along, not This stems from the same belief system that responding to their ill placed health anxieties. Our job predicted weapons of mass destruction in Iraq and is to keep them away from the vortex of secondary care dictates that you need a certificate to use stepladders. investigation that would only feed their introspection. I can think of 20 patients capable of extreme violence, One group of patients, however, is different: those but I doubt any of them will strike. Are we to lock with antisocial personality disorder, defined by lack of them all up? Besides, the single most dangerous group respect, remorselessness, recklessness, and deceit, but are the gangs of under 16 year olds, “the young team.” most of all by violence. My well practised and calm As their victims are largely confined to other mem- veneer conceals the sweat that trickles down my back bers of the urban underclass, there is never any press when patients produce knives for my interest and off- coverage. Are we to lock these children up too? Risk handedly talk of stabbings and beatings. Many have assessment just doesn’t work. been stabbed, countless have been beaten, and the odd Lastly, this isn’t our job. Society and the legal pro- one has been or has murdered. These patients are not fession pour all the woes into a large funnel labelled ill, and no treatment can be offered. In the past your “medicalisation,” then ram the responsibility down grandmother would simply have called them “bad.” our throats. Frankly we have enough responsibility The government now wants the Mental Health Act just caring for the ill and dying. I am sick of lawyers to cover these patients, with powers detaining them passing the buck to us. If any of these patients truly indefinitely to protect the public (see News, p 113)— are such a threat, then let the legal profession sentence even though they may not have committed any crime. them and keep them off the streets. This follows pressure in the wake of some high profile Des Spence is a general practitioner, Glasgow murders. [email protected]

What goes around, comes around THE BEST It is a great time to be Irish; this is by upper class Brits, sheep with eastern Europeans and Brazilians MEDICINE the first generation for hundreds no road manners, mawkish on my list. of years to be free of the scourge playwrights, and miserable The young couple in front of me Liam Farrell of emigration, of the flight from memoirs. This depressed isolation had brought in a complicated form. snipe grass and poverty, of our had its own peculiar appeal: Their English was halting, and they young people dispersing on the four gorgeous, reclusive French film were diffident and shy, like the winds. stars would retreat to our uttermost sad heart of Ruth, sick for home When I qualified in the 1980s west because they wanted a winter amid the alien corn. I helped them most of my class were already of peace, guilt, inner spiritual complete the form, which included preparing for flight. When my best torment, and more guilt. And they 30 minutes of negotiating a helpline pal in college, Mick Fitzpatrick, left wouldn’t be disappointed: guilt was of labyrinthine complexity (press for Canada, we were not overly our specialty, all the fault of the button 2 if you are gibbering with sentimental, but 20 years later he’s English, from whom we had learnt frustration, 3 for murderous rage). still there and has made a new and Victorian prudery; before that we It wasn’t strictly medical, but I happy life for himself. Mick was a were a bawdy folk. wanted them to know that here was great guy, and looking back over But Alain Delon needn’t bother a friend, someone who would look the years I know that his leaving any more, unless he wants to catch out for them and treat them like was an immense personal loss for the Tullylish Burning Goat Festival I’d have wanted our own exiles to me. We stay in touch and we meet or wishes to make a new cult be treated. And then she coughed occasionally, but it’s not the same. horror film in Ballyboggy. Thanks theatrically, brilliantly mimed an If you’ve seen The Field or to the Celtic Tiger, Ireland has awful sore throat, and ventured, Ryan’s Daughter you’ll know what been transformed and has become “Antibiotic?” Some cultural mores Ireland was once like: rain, incest, prosperous and confident. are quickly absorbed. loneliness, poteen, village idiots, And so the wheel turns; we have Liam Farrell is a general practitioner, graveyards, tears, beauty, doomed our own blossoming immigrant Crossmaglen, County Armagh love affairs, being patronised population now, and I have many [email protected]

158 BMJ | 20 JANUARY 2007 | VOLUME 334 VIEWS & REVIEWS

A historical whopper MEDICAL CLASSICS Doctor in the House By Richard Gordon Does it matter—in so BETWEEN suspected until the First published 1952; film released 1954 far as anything lit- 1920s, after George Everyone has heard of Doctor in the House. It has been erary matters these THE LINES and Gladys Dick a film, a play, and a television series, but it started as days—if historical Theodore Dalrymple established that a novel by “Richard Gordon,” the pseudonym of a BMJ fiction is inaccurate? streptococcal infec- assistant editor. The book, published in 1952, has Does anyone mind tion was responsible acquired the reputation of being a collection of old if Richard III is a for scarlet fever. The medical jokes. It is in fact a picture of medical student tendentious, even connection between life in London in the late 1940s—comical, yes, but also full of what one reviewer called “shrewd, sympathetic a sycophantic and sore throat and understanding.” opportunist, justifi- rheumatic fever It was the film, released in 1954, that became cation of the Tudor was not proved a classic. The producer, Betty Box, got Nicholas dynasty, and that beyond all doubt Phipps to write the script and later commented that the real Richard until 1934. “there wasn’t a great deal of the book in it, except III of history was a Does the error for the characters.” This was untrue as many lines Jolly Good Thing (in affect the value appear verbatim, but it was the actors who gave it the sense of British of the book? If magic. The film won Kenneth More a Bafta, set Dirk Bogarde on the road to his knighthood, and made humourists Sellar and we answer in the a generation of us fall in love with nurse Muriel Yeatman of 1066 and The connection between affirmative, to be Pavlow. All That fame), or at streptococcal sore throat consistent we must Box made six more Doctor films of diminishing least a very much less and rheumatic fever was downgrade the quality. The only character to appear in all seven Bad Thing? history plays. If we was Sir Lancelot Spratt, the irascible surgeon Not long ago, I read not even suspected until answer in the nega- played by James Robertson Justice. With a voice Sebastian Faulks’s the 1920s tive, we are giving “like an educated foghorn,” bullying his juniors and patronising his patients, he became the archetypal novel Human Traces historical novelists consultant, still (reviewed in BMJ licence to play fast reviled by ageing 2005;331:1029). It brought out the inner and loose with history, sometimes no doctor bashers and pedant in me: there was more rejoicing doubt for sinister ends. Here is a subject by bright young in my mind over one historical mistake for the most delicious of all kinds of doctors who think than over 99 true facts. vituperative debate: one in which noth- they are more As it happens, there is a whopper at ing is at stake, and where no definitive empathic than he the very heart of the book, which con- answer can be given. was. They’re kidding cerns the relationship between two 19th The strange thing is that rheumatic themselves. The century alienists, one French and the fever has all but disappeared. It still book describes other English. Their friendship under- existed in my youth, and when I was a how Sir Lancelot goes strain as one of them remains faith- medical student a disproportionate part devoted his life to his ful to the organicist, strictly medical of my time seemed to be spent trying charitable patients, view of psychiatric disease and the to distinguish between its many and Medical magic how he was president other adopts the new fangled Freudian, various cardiac sequelae. of almost every students’ club, and how, through psychological approach. A boy in my class at school had rheu- the war, he had slept every night in hospital during the bombing, playing cards or sharing a pint with the Their disagreement comes to a head matic fever, one of the last children, I sup- students living in with him. In those days, patients over the case of a young woman. Her pose, to develop it. Looking back, he had understood this. Doctor in the House showed his symptoms are regarded as hysterical never had a healthy or robust appear- humanity, particularly in his dealings with the po- by the Freudian, the result of the usual ance, but we always suspected him of faced dean of the medical school; but as the series childhood miseries, but the organicist manufacturing his symptoms and exag- progressed a stereotype took over. points out that in fact hers is a classic gerating his illness so that he could avoid Now on DVD with the unappealing subtitle “classic case of rheumatic fever, for various unpleasant strenuous physical exertion, British comedy,” the film is selling well in Britain symptoms followed a streptococcal such as rugby. He had enviable amounts and has just been released in Germany as Aber, Herr Doktor. You may dislike its sexism (nurses are sore throat. of time off, too, so we thought that he chased but chaste) or disbelieve a world where Any reasonably competent doctor was a pathetic sissy. He probably needed working clinicians could receive knighthoods and should have been able to draw the open heart surgery later in his life. Of students could conduct home deliveries, but you will appropriate conclusion. The two men, course, I now regret our joyful malignity enjoy its warmth. In 1954, it was seen as an exposé having once been as close as brothers, and casual cruelty towards him. of medical reality, but unlike today’s edgy hospital don’t speak to one another afterwards. It is strange how historical fiction, dramas, what it revealed was tolerance and good Unfortunately, these events take place even when it is inaccurate, can evoke humour. No wonder people are still buying it. at the end of the 19th century, and the memories that might be morally James Owen Drife, professor of obstetrics and gynaecology, connection between streptococcal sore instructive Leeds [email protected] throat and rheumatic fever was not even Theodore Dalrymple is a writer and retired doctor

BMJ | 20 JANUARY 2007 | VOLUME 334 159 OBITUARIES For the full versions of articles in this section see bmj.com

Dunedin Hospital he spent three years, in 1968, of the London Royal College of from 1959 to 1961, working for the New Physicians in 1972, of the Royal College Zealand Medical Research Council. of Ophthalmologists in 1989, and of the During this time he was warden of Knox Academy of Medical Sciences in 1999. College, an Otago University student He was awarded a dozen honorary residence. He then became medical fellowships, 15 prizes for multiple sclerosis registrar at Dunedin Wakari Hospital. He research, and honorary membership of was awarded his PhD in 1962. 10 overseas neurological associations. He He came to Britain in December 1963 held six visiting professorships and 30 as senior house physician to Macdonald named lectureships. He served on many Critchley at the National Hospital for multiple sclerosis committees in the United Nervous Diseases in Queen Square, where Kingdom and internationally. he was to spend most of the rest of his He successfully brokered the move of the career. He started his experimental research journal Brain from Macmillan to Oxford on demyelination and remyelination at University Press. This was a financial this time. From 1965 to 1966 he was a success, and the profits are now used to neurology research fellow at Harvard, assist young neuroscientists. He later, from William Ian McDonald working on muscle spindle function and 1991 to 1997, became Brain’s editor. demyelination. Ian was courteous, charming, and Neurologist who refined the He returned to Queen Square in 1966 as dashing, always immaculately dressed, and diagnosis and understanding of consultant physician, and from 1969 was had an absurdist sense of humour. He was also consultant neurologist at Moorfields immensely cultivated, played the piano multiple sclerosis Eye Hospital. With Professor Tom Sears, to a professional standard, loved music he often worked until daybreak, driving and opera, and collected art and antiques. William Ian McDonald (“Ian”) greatly home for a couple of hours’ sleep before In his “retirement” he was Harveian advanced knowledge of multiple returning for a 9 am clinic. librarian at the Royal College of Physicians sclerosis and was also an expert in neuro- In the 1970s he researched evoked from 1997 to 2004. He was an honorary ophthalmology and, in retirement, an potentials, a laboratory method that member of the Royal Society of Musicians, outstanding Harveian librarian at the Royal supplemented the clinical diagnosis of and was delighted when he was elected a College of Physicians. multiple sclerosis. foreign member of the Venetian Institute of He transformed the study of multiple When magnetic resonance imaging and Science, Arts and Letters. sclerosis from a series of disjointed spectroscopy were introduced in the 1980s In October 2004 he and his friends approaches to a coherent and scientific Ian realised their potential for visualising noticed an inability to read music and entity, and the current diagnostic criteria the course of inflammatory brain diseases. play the piano expressively. A magnetic are called the McDonald criteria. He At the suggestion of Professor Alan resonance scan in December showed that was one of the few neurologists to do Davison, he developed a research unit he had had a minor stroke. He later wrote substantial laboratory research. When funded by the Multiple Sclerosis Society a marvellous description of it (Musical magnetic imaging was introduced he and directed it from 1984 to 1995, along alexia with recovery: a personal account, quickly recognised its potential value in with his other duties. Brain 2006;129:2554-61). He made a diagnosis, monitoring, and establishing He held a personal chair from 1974. substantial recovery over 18 months, and the pathogenesis of multiple sclerosis. He In 1987, when Roger Gilliat retired as he continued working. carried out studies that showed his hunch departmental chairman, many people He died in his sleep of a heart attack, was correct, and, thanks to him, imaging regarded him as the natural successor. having gone to bed early—after a full day’s is now used to speed up the monitoring of However, the chair went first to David work—to read Ruskin’s The Stones of Venice. clinical trials. He was a great mentor and Marsden and then to Anita Harding. When He is survived by his partner of 25 years, an excellent lecturer. the latter retired in 1995, McDonald finally Stanley Hamilton. Ian was born into a Presbyterian family succeeded to it. He retired three years later, Caroline Richmond in Wellington, New Zealand, the son of aged 65. an insurance manager. From St Andrew’s During these years he did a huge William Ian McDonald, consultant neurologist Queen College he went to Otago University, number of other things. He lectured Square (1966-98), professor of neurology (1974- earning a BMedSci in 1955 and qualifying widely in the United Kingdom and 98), consultant neurologist Moorfields Eye Hospital MB ChB in 1957 with distinctions in abroad, including giving many named (1969-96), editor of Brain (1991-7), (b 1913; q Otago pharmacology, microbiology, medicine, lectures. He was awarded fellowship of 1957; FRACP, FRCP, FRCOphth), died from a heart and gynaecology. After house jobs in the Royal Australian College of Physicians attack on 13 December 2006.

160 BMJ | 20 JANUARY 2007 | VOLUME 334 OBITUARIES

develop comprehensive services 1959 he was appointed consultant Moyna Gladys Clark to the NHS and industry in central Aloysius (“Lou”) Michels dermatologist at the Royal Free Scotland. His last paper, completed Hospital, where he worked for 30 when he was unwell, was in years. His publications numbered Occupational Medicine—for which nearly 300. His many contributions he was a regular reviewer. Sandy to dermatology and medicine was a regional specialty adviser included examining for the MRCP of the Faculty of Occupational and being president of the British Medicine, and the main author Association of Dermatologists of one of its recent position and member of the American statements, as well as helping to Dermatological Association. He produce two Scottish intercollegiate leaves a wife, Helen, and three evidence based (SIGN) guidelines. children. Former clinical medical officer Carlisle He was also active in the Society of Former consultant anaesthetist Robin Graham-Brown and designated doctor for child Occupational Medicine. He leaves a Stirling Royal Infirmary (b 1962; protection North Cumbria (b 1943; q wife, Alison. q Utrecht 1987; FRCA), died from Leeds 1967; DCH), died from recurrent Ewan B Macdonald, Ian S Symington, recurrent oligodendroglioma on 18 Peter Thomas John glioblastoma on 10 September 2006. Richard Soutar September 2006. Christopher Plumbly Moyna Gladys Clark entered general Aloysius Michels (“Lou”) Warner practice in Dunstable. In 1982 she grew up and qualified in the and her family settled near Carlisle, Gwilym Penrose Hosking Netherlands. He started his career and Moyna worked as a school in anaesthetics in south-east doctor. During 1988 she began England, was a senior registrar working as a medical examiner in in Leeds, and spent a year child and adult sexual abuse, in working in Western Australia, 1998 becoming the designated where he enjoyed windsurfing doctor for child protection in and qualified as a private pilot. North Cumbria. Moyna continued He was appointed consultant to to provide local child protection Stirling Royal Infirmary in 1996. He training after her retirement in developed the acute pain service 2003, and even during her final and became college tutor. He was Former chief occupational medical illness she was keen to be involved diagnosed with a brain tumour officer Province of Manitoba (b in the new national child protection Consultant paediatric and in 1998 and was able to return 1920; q Middlesex Hospital, London, training. She leaves a husband, developmental neurologist Cromwell to work for five years. When his 1944; MD, PhD, FRCPath), died from Keith, and two children. Hospital, London, and Thornbury tumour recurred in 2003 he retired pulmonary fibrosis and Waldenstrom’s Nicola Pritchard Hospital, Sheffield (b 1945; q Charing from medicine. Lou was a talented macroglobulinaemia on 9 January 2006. Cross 1969; FRCP, FRCPCH), died from trumpet player and loved to play Peter Warner was awarded the a heart attack on 22 October 2006. jazz. He leaves a wife, Jo, and two Canadian Centennial Medal for Alexander (“Sandy”) Gwilym Penrose Hosking was daughters. “valuable services to the nation.” Gordon Elder attracted to paediatric neurology Kenneth Lamb, Jonathan Falla, He was director of the department partly because of a desire to work Josephine McGettigan of bacteriology at Winnipeg in multiprofessional teams, and General Hospital (1958-67), and he pioneered new models of care associate professor, bacteriology during his first consultant post. Imrich (“Emery”) Sarkany and immunology at the University In 1990 he joined the Wellcome Emeritus consultant dermatologist of Manitoba (1958-80). During Foundation as a senior medical Royal Free Hospital (b 1923; q London 1967-72 he was director of health adviser and subsequently became a 1952; FRCP), d 22 November 2005. services with the Manitoba research physician. In 1997 Gwilym Imrich Sarkany (“Emery”) government; assistant deputy set up an independent healthcare overcame great difficulties to minister, public health; and first company to provide services for become one of Britain’s leading chairman of the Clean Environment children with neurological and dermatologists. Born in Slovakia, Commission. He was assistant Consultant occupational physician neurodevelopmetal problems. In he was sent by his mother at regional director of Health and Salus, NHS Lanarkshire (b 1962; recent years he became interested the outbreak of war to England, Welfare Canada in Manitoba for q Dundee 1985; MRCGP, MPH, in the links between untreated where he enrolled in the Czech seven years before returning to FFOM), died from an epithelioid attention deficit disorders and brigade of the British Army. His the government. He worked part haemangioendothelioma on 11 offending behaviour and was on the whole family was lost in the time as a consultant until his death November 2006. independent monitoring board at Holocaust. He began his medical and was editor of the Manitoba Alexander Gordon Elder (“Sandy”) Wakefield prison at the time of his studies in Prague but, with the Medical Review. He leaves a wife, was an experienced general death. He leaves a second wife, five advent of communism, escaped Iona, and his six children and four practitioner before training in children, and five grandchildren. to England, an opening at St grandchildren. occupational medicine. He helped Emma-Jane Hosking, Suzie Mitchell Thomas’ reigniting his career. In Iona Warner

BMJ | 20 JANUARY 2007 | VOLUME 334 161 MINERVA

Maintaining adequate oxygen supplies to the Before you can sell a drug, you have to sell brain protects other organs from trauma, so when a disease. An article in Common Knowledge patients undergo cardiac bypass surgery it may (2006;12:379-87) discusses how, in the 1980s, make sense to actively monitor brain oxygen everyday feelings of bashfulness and shyness saturation. Two hundred bypass patients were were reinvented as the medical conditions of randomised to actively displayed intraoperative social phobia and social anxiety disorder. At one cerebral oxygen saturation monitoring or blinded point, bus shelters across the US were plastered oxygen saturation monitoring. Significantly more with posters containing the eye catching slogan, patients in the blinded monitoring group showed “Imagine being allergic to people.” Pictures of a prolonged cerebral desaturation and had a longer forlorn yet attractive man staring morosely into duration of stay in intensive care. The incidence a teacup illustrated the headline, “You blush, of major organ dysfunction and death was also sweat, shake—even find it hard to breathe,” higher in this group (Anesthesia and Analgesia A 66 year old woman presented with a long the copy continued “that’s what social anxiety 2007;104:51-8). history of a dry cough and hoarse voice. disorder feels like.” Magnetic resonance scanning showed that she Before the G8 meeting at Gleneagles in 2005, had a non-secreting right sided parapharyngeal Phototherapy has been used for over 30 years glomus vagale tumour abutting the skull base thousands of people in the developed world to treat neonatal hyperbilirubinaemia. A small and displacing the internal carotid artery demonstrated their belief in the idea of making anteriorly. After excision of the tumour, she had a study from France raises a slight query about Archives of Dermatology poverty history by marching. Politicians transient, well demarcated left hemifacial flush, its long term safety ( responded by pledging an extra £50bn probably caused by neuropraxia of ascending 2006;142:1599-604). A group of children aged (€75bn; $97bn) in foreign aid, half of which sympathetic nerve fibres. The phenomenon is 8 or 9 years who had been treated in this way was to go to Africa. But an analysis in African sometimes known as harlequin syndrome. had significantly more naevi than a comparison Affairs (2007;106:133-40 ) doubts whether Jason Y K Chan preregistration house officer group. Because higher numbers of acquired rich countries are really serious about tackling [email protected] benign naevi are associated with an increased African poverty. Little progress has been made Richard Oakley specialist registrar risk of melanoma the authors wonder if children in resolving the basic paradox of aid; namely, Michael J Gleeson professor exposed to neonatal phototherapy should have because of conflict, mismanagement, and department of otolaryngology and skull base surgery, regular dermatological surveillance. Guy’s Hospital, London SE1 9RT corruption, the countries that need aid most are the ones least likely to use it effectively. The enzyme methylenetetrahydrofolate different races. Not any longer. A commentary reductase plays a central part in folate As long as cerebrospinal fluid circulates freely, on recent epidemiological studies points out metabolism. A common polymorphism of the surely it follows from the laws of physics that the that the disease is more common in men than gene that encodes it results in reduced enzyme pressure measured by lumbar puncture equals in women; that some migrant groups have activity. Although the evidence is disputed, intracranial pressure. But it is always good to strikingly high rates; and that risk is associated possession of this allele has been linked to check. A series of patients being investigated with older paternal age, winter or spring birth, an increased risk of cardiovascular disease. A for normal pressure hydrocephalus had their living in higher latitudes, and complications systematic review now suggests that the same intracranial pressure measured directly with during pregnancy and birth (Archives of General polymorphism increases the risk of psychiatric a catheter in the roof of the lateral ventricle as Psychiatry 2007;64:14-6 ). This ought to provide disorders including depression, bipolar disorder, well as indirectly by lumbar puncture (Neurology useful clues about aetiology. and schizophrenia, raising the possibility of 2007;68:155-8 ). The correlation was almost using folate for treatment and prevention of these perfect. A breathtaking five million central venous diseases (American Journal of Epidemiology catheters are inserted every year in the US 2007;165:1–13 ) The World Question Center at the Edge Foundation alone. So, perhaps it is inevitable that weird asked 150 or so scientists, philosophers, and complications occur from time to time. A woman Carpal tunnel syndrome, caused by compression thinkers what they felt optimistic about. Anyone admitted with injuries after a road traffic incident of the median nerve at the wrist, is one of the in the northern hemisphere who currently has developed suspected catheter related sepsis. most common peripheral neuropathies. A seasonal affective disorder should have a look at After a replacement catheter was inserted, it systematic review confirms that jobs requiring the results (www.edge.org/q2007/q07_index. proved impossible to remove the old one. The repetitive flexion and extension of the wrist, html), and they will discover many reasons to feel introducer needle and guide wire of the second such as hammering or painting, or those in more cheerful. catheter had impaled the first and trapped it. which hand held vibrating tools are used for Although the authors of the case report offer prolonged periods, greatly increase the risk of Schizophrenia was once thought to be one an algorithm for detecting and dealing with the developing the condition (Occupational Medicine of those unusual diseases whose incidence problem, let’s hope it won’t be needed very often 2007;57:57-66 ). In contrast, no association was varied little between different countries and (British Journal of Anaesthesia 2007;98:89-92 ). seen with using a keyboard or computer.

162 BMJ | 20 JANUARY 2007 | VOLUME 334