Abortion in Cambodia I 3

ABORTION CAMBODIA IN Care seeking for abortion and family planning services:

Findings from a PEER study, conducted with women in Phnom Penh and Kandal provinces

Dr Joanne Hemmings and Ben Rolfe Reduction in Maternal Mortality Project Options Consultancy Services Ltd July 2008 Disclaimer The views expressed in this report are solely those of the authors and should not be taken to reflect the views of any third party.

For more information about PEER, or to download this document visit www.peer-method.com Contact: [email protected] Options Consultancy Services Ltd: www.options.co.uk RMMP: www.rmmp.org.kh

This work is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-sa/3.0/

You are free: • To copy, distribute and transmit the work • To adapt the work

Under the following conditions: • Attribution. You must attribute the work in the manner specified by the author or licensor (but not in any way that suggests that they endorse you or your use of the work). • Noncommercial. You may not use this work, or the PEER brand for commercial purposes without express permission of Options Consultancy Services. • Share Alike. If you alter, transform, or build upon this work, you may distribute the resulting work only under the same or similar license to this one. • For any reuse or distribution, you must make clear to others the license terms of this work. The best way to do this is with a link to the web page above. • Any of the above conditions can be waived if you get permission from the copyright holder. • Nothing in this license impairs or restricts the author’s moral rights.

Acknowledgements The research team would like to thank the peer researchers in Kandal and Phnom Penh who collected these findings. Also the management and staff at CHEMS, particularly Yvette Height, and, Siv Dalis and Sin Sovansorphea (PEER supervisors).

RMMP is funded by the UK Department for International Development.

Abortion in Cambodia I 3

c o n t e n t s

r e p o r t s u m m a r y 3

1. introduction 9 1.1 The context of reproductive health in Cambodia 9 1.2 Contraceptive prevalence and unmet need 10 1.3 Research objectives 12 1.4 Research method: Overview of PEER 12

2. b a c k g r o u n d : reproductive a n d m a t e r n a l h e a l t h in c a m b o d i a 14

3. m a i n f i n d i n g s 17 3.1 Who has abortions, and why? 17 3.2 Why don’t women use public abortion services? 24 3.3 What do women think about abortion? 32 3.4 Women’s perceptions and experiences of family planning 33 3.5 What prevents women from using family planning? 37 3.6 Other reproductive health issues for women 40 3.7 How do target groups receive information and communications? 41

4. implications o f p e e r f i n d i n g s 43 4.1 Implications for communications 43 4.2 Implications for supply side activities 46 4.3 Implications for interpreting quantitative survey data 50

5. implications f o r a d v o c a c y 52 5.1 Advocacy at provider level 52 5.2 Advocacy at higher level 53 5.3 Women’s right to access reproductive health services 54

6. bibliography 56

a p p e n d i x 1 p e e r m e t h o d in d e t a i l 57

Abortion in Cambodia I 3

- r e p o r t s u m m a r y

karsikSaenH)aneFVIeLIgkñúgeKalbMNgEsVgyl;TsSn³\riyabf nigGakb,kiriyabc©úb,nñTak;Tg nwgkar rMlUtkUn nigkareFVIEpnkar®KYsarrbs;®sþIenAtMbn;TIRbCMuCnmYykñúgTI®kugPñMeBj nigPUmimYykñúgextþkNþal én RbeTskm<úCa. lT§plrkeXIjBIkarsikSaenHnwgeRbIR)as;sRmab;karbegáItKeRmagkat;bnßymrNPaB mata. tamry³viFIsa®s PEERþ ®sþIenAmUldæan)anbegáItsMNYredayxøÜnÉgpÞal;sRmab;smÖasn_mitþP½kiþrbs; xøÜnnigpþl;mtiEklMGdl;lT§plenHCUneTA®kumRsavRCav. RsþITaMgenaH)anRbmUlTinñn½yEbbKuNPaBy:ag lMGitTak;Tgnwg\riyabfcMeBaHkarrMlUtkUn nigkareFVIEpnkar®KYsarenAkñúgsgÁmrs;enAEdlCaesvaRsþIFøab; TTYl)an nigCaktþaCH\T§iBlelIviFIsa®sþnigCeRmIsrbs;Gñkpþl;esva.

karsikSakñúgtMbn;TIRbCMuCn)aneFVIeLIgenAPaKxagt,ÚgénTI®kugPñMeBj. EdlkEnøgenHCakEnøgsñak;enArbs; kmµkarnIkat;edr nigGñklk;dUrtUctacCaeRcInnak; nigmanTItaMgsßitenACittMbn;]sSahkmµFunRsal. tMbn; enHmandg;sIuetRbCaCnx

®sþIBnüarkarEsVgrkesvarMlUtkUnEdlCajwkjab;bNþalmkBIRsþITaMgenaHBüayamrMlUtkUnedayxøÜn ÉgedayeRbIR)as;viFIsa®sþepSg²dUcCakarelathk; kareRbIR)as;»sfburaNfñaMRKab;TijBI»sfsßan. enAeBlEdlRsþIEsVgkesvakmµRsþIminniymeRbIesvakmµrdæsMrab;esvarMlUtedaysuvtißPaBeT edayRsþI mankarP½yxøaccMeBaHbuKÁlikEdlminrak;Tak;cMeBaHBYkKat;nigPitP½ykarxVHxatcMeBaHkarrkSakar sMgat;. CaerOy²®sþIniymeTAeRbIR)as;esvarMlUtkUnBIGñkpþl;esvaÉkCn nigminmanCMnajEdlRsþITaMgenaH yl;eXIjfaCakarpþl;esva }manPaBrak;Tak;rkSakarsm¶at;nigsäat}. »sfsMrab;eFIVkarrMlUtGacrk)anya:gTUlMTUlay nigniymeRbIR)as;CaTUeTA. b:uEnþKµannrNamñak;dwg c,as; faetI®sþI®tUv)anpþl;karpSMfñaM mIehVRBIsþÚn nig ®bUsþakøaDInsIuenTik edaysuvtßiPaBRsbtamGnusasn_ rbs;GgÁkarsuxPaBBiPBelakjwkjab;b:uNÑa ¬nigkñúgkRmitb:unµandUs¦ eLIy.

Abortion in Cambodia I 4 Abortion in Cambodia I 5 RsþIeRCIserIskarrMlUtedaysarmUlehtumYycMnYnEdlCaTUeTAbMput ¬minEmntamlMdab;¦³ suxPaBmin l¥ karmanépÞeBaHmunGaBah¾BiBah¾ karmankUnjwk karTTYlxusRtÚvkñúgRKÜsar nigPaBRkIRk. karTTYlxusRtÚvCaeRcInEdlCabnÞúkrbs;RsþIEdlCaehtueFVIeGayRsþIRtÚvkMNt;nUvkarmanK’t ®sþICaeRcInmankaryl;dwgtictYcbMputGMBImUldæanCIvsa®sþnigrUbviTüaénkarbnþBUC. eTaHbI®sþIPaKeRcInmankaryl;dwgGMBImeFüa)ayBnüarkMeNItTMenIbPaKeRcInk¾eday k¾RsþITaMgenaHCa jwkjab;mankaryl;dwgtictYcbMputGMBIkareRbIR)as;nigplb:HBal;CaGviC¢manénmeFüa)ayTaMgenH. BaküccamGaram nigkarP½yxøacCaeRcInGMBIplb:HBal;énmeFüa)ayBnüarkMeNItKWCa]bsKÁd¾FMkñúg karTTYlyk nigkareRbIR)as;CameFüa)ayBnüarkMeNIt[)anCaRbcaM.

KuNPaBminl¥énkarpþl;esvaBnüarkMeNIt minRtwmEteFVI[RsþIsÞak;esÞIrkñúgkareRbIR)as;esvaBnüar kMeNItEbbTMenIbb:ueNÑaHeT b:uEnþEfmTaMgbgá[manplvi)aky:ageRcIneTotpg. plrMxanBImeFüa )ayBnüarkMeNItEdlGacekItman nigBaküccamGaramEdlmanlkçN³GviC¢man eFVI[®sþICaeRcIn eRCIserIs ¬b¤gakeTAeRbI¦ viFIsa®sþburaNb¤eFVIkarrMlUtkUn. kargarCabnÞan;mYyKWRtUvpþl;esvaEpn kar®KYsarEdlmanKuNPaBx

Abortion in Cambodia I 4 Abortion in Cambodia I 5

tMélrbs;esvakmµrMlUtedaysuvtißPaBKW)anraMgsÞHenARKb;Epñk ehIyEdlCasBaØamYyd¾FMEdlCMruj eGayRsþIeTATTYlesvarMlUtKµansuvtißPaB. PaKeRcInénmUlniFismFm’suxaPi)alKWminKaMRTesvarMlUt edaysuvtißPaBedayEp¥ktameKalneya)ayCatiRBmTaMgc,ab; US “MexicoCity”. karbdiesFsiTi§ TTYl)ansuxPaBbnþBUCmUldæanTaMgenHCasBaØaEdlkMBugeFIVeGayRsþIkm<úCaCaeRcInrUbRbQmnwg eRKaHfñak;xøaMgdl;)at;bg;CIvitnigBikarPaB. Gs;ry³eBlCag 10 qñaMbnÞab;BIkarRbkas[eRbIR)as;c,ab;sþIBIkarrMlUtkUnenAqñaM 1997 esvarMlUtkUn edaysuvtßiPaBenAEtminTan;BRgIk[dl;®sþIkm<úCaPaKeRcInenAeLIyEdlmanRsþIcMnYn 85° rs;enAtMbn; CnbT. kñúgkrNIenHkargarcaM)ac;KWRtUvBRgIkvis½ysuxaPi)al[)anTUlMTUlayCaGtibrmarab;bBa©Úl TaMgGñkpþl;esvaÉkCnpg edIm,Ikat;bnßykarrMlUtkUnEdlKµansuvtßiPaBTaMgtamry³ kareFVIEpnkar ®KYsar nigkarpþl;esvaedayykcitþTukdak;CaBiesselIsmFm’én»kasTTYlesvatamry³karpþl; hirBaØvtßúedIm,IbMeBjtRmUvkarbs;RsþI.

b a c k g r o u n d

This study was conducted to learn about current perceptions, attitudes and behaviours relating to abortion and family planning among women in an inner city area of Phnom Penh and a village in Kandal Province, Cambodia. Findings will inform the development of a maternal mortality reduction project. Using the PEER method, local women developed their own research questions, interviewed their friends, and fed back findings to the research team. They collected detailed qualitative data on attitudes to abortion and family planning, the social context in which women experience these services, and factors influencing method and provider choice.

s t u d y c o n t e x t

The urban study was conducted in an inner city area in southern Phnom Penh. The area is inhabited by many garment factory workers and small scale traders, and located close to a light industrial area. The area has a relatively high population density and level of poverty typical of much of the periphery of central Phnom Penh. The women recruited for the study were broadly typical of the population in the area: of reproductive age, working in insecure employment, with poor labour conditions.

The Kandal study area was a village 50km from Phnom Penh. Despite its proximity to the city’s outer urban sprawl, the village relied on subsistence farming: rice farming, fishing in the lake, gathering wild products and keeping livestock. People could also commute to the city for work, and there were clothing factories from 10km away where many young women from the village worked. Most people were also engaged in small-scale commercial activities: selling food in the market, labouring, or sewing.

Abortion in Cambodia I 6 Abortion in Cambodia I 7 m a i n f i n d i n g s

Women delay abortion care seeking, often trying to self-induce an abortion using a variety of methods –jumping, traditional herbs, pills supplied by pharmacists. When they seek care, women prefer not to use Government providers for safe abortion care, where they often feel intimidated by unsympathetic staff and fear a lack of confidentiality. Women report having to persuade public providers in order to receive services. The private sector and often unsafe informal providers are preferred, perceived by women as “friendly, confidential and clean”. Medical abortion is widely available and in common use. However it is not clear how often, (and in what doses) woman are provided with the WHO recommended combination of mifepristone and a synthetic prostaglandin. Women choose to have an abortion for numerous reasons, most commonly (not in order): Ill health, pre-marital pregnancy, short birth interval, competing family responsibilities and poverty. The many responsibilities placed on women limit their options when considering whether to continue with a pregnancy. Many women have a very limited understanding of the basic biology and physiology of reproduction. Whilst most women are aware of the majority of modern contraceptive methods, their understanding of their uses and side effects is often very limited. Widespread rumours and fears about the side effects of contraceptives present a major barrier to uptake and consistent use.

m a i n implications

The poor quality of contraceptive service delivery not only limits access, but has wider implications. Unexpected side effects, and the resulting negative rumours influence many women to choose (or revert to) traditional methods, or abortion. There is an urgent need for better quality family planning provision, with effective counselling and a greater range of choice. This will enable women to find a method to suit them, and use it consistently. Improved delivery must be supported by evidence based and well designed communications strategies. These are required to inform and support consistent use, and to counter the widespread misinformation currently contributing to high rates of discontinuation. For those women who unfortunately find themselves with an unwanted pregnancy, communications need to stress seeking prompt pregnancy testing, and advice from at a safe and qualified provider.

Abortion in Cambodia I 6 Abortion in Cambodia I 7 To prevent deaths from unsafe abortion, services must be designed to be both safe, and responsive to the voiced needs of women. The government sector is currently failing women, who, in addition to technical safely, require kind and confidential services to encourage them away from the informal sector. Young, unmarried women are put at particularly high risk, both family planning and safe abortion services should be specifically tailored to their needs. Medical abortion is an important technology for woman’s reproductive health. Its introduction should be prioritised, in tandem with the development of appropriate training and referral mechanisms, especially for pharmacists who are often the first point of contact for women seeking help. Further research to understand the current use of traditional and modern abortificants (pharmacological tests, mystery client studies etc) to inform scale up of services is required. Additional research to explore the availability and use of pregnancy testing, and introduction of emergency contraception would also support efforts to reduce unsafe abortion. The cost of safe abortion services are prohibitive in all sectors, and a very significant proportion of women are unable to afford anything but an unsafe abortion. The majority of health equity fund implementers are unwilling to support access to safe abortion services in accordance with the national policy, due to the US “Mexico City” gag rule. This denial of basic reproductive health rights is putting many Cambodian women at significant increased risk of death or disability. Over 10 years since the introduction of the 1997 abortion law, safe abortion services continue to be out of reach to the majority of Cambodian women, 85% of whom live in rural areas. An urgent response is required, which should maximise the potential of the entire health sector, including private providers; to reduce unsafe abortion both through family planning and service delivery, with special attention to equity of access through demand side financing.

Abortion in Cambodia I 8 Abortion in Cambodia I 9

1. introduction

t h e c o n t e x t o f reproductive methods such as the pill and condoms. 1.1 h e a l t h in c a m b o d i a Long term methods such as sterilisation (tubal ligation or vasectomy) and The Khmer Rouge-led genocide of IUDs are a relatively new introduction. 1975–1979 had a significant impact on These services were either supported or Cambodia’s demographic and fertility operated by international NGOs. Only profile (de Walque, 2004; NIS & ORC in 1994, with support from the United Marco, 2001). Following the end of the Nations Population Fund (UNFPA), the regime, a baby boom occurred and con- Royal Government of Cambodia (RGC) tinued, with the result that 55 percent took the first steps toward implement- of the population is under the age of 20 ing its own FP program, which includ- (NIS & ORC Marco, 2001). This has sig- ed the introduction of services at health nificant implications for reproductive centers, FP education, and training of health and maternal mortality in Cam- public health sector staff. bodia (NIS & ORC Marco, 2001). Fam- ily planning (FP) services and modern In the same year, the Maternal and Child contraceptives were available for the Health Plan 1994-1996 was developed, first time in Cambodia in 1991. The introducing specific objectives for first methods available were reversible lengthening the interval between births.

Abortion in Cambodia I 8 Abortion in Cambodia I 9 To meet these objectives, the MOH ferentials between different sectors of approved its first Birth Spacing Policy for the population. For example, 30.6% Cambodia (MOH, 1995a), which advocat- of currently married women in urban ed the provision and use of a full range areas currently use a modern contra- of contraceptive services: “the availability ceptive compared with 26.5% of mar- of reversible and affordable contraceptives will be ried women in rural areas, and women increased so that all couples may have access to with secondary education or higher are them” (MOH, 1995a). The first mass 1.4 times as likely to use contraception media campaign about family planning than women with no education (31.9% was not until 19962. In 1997, the MOH versus 22.2 % of married women cur- issued its National Policy and Strategies for Safe rently using a modern method respec- Motherhood. This policy provides guide- tively) (CDHS 2005). lines to integrate all components of safe motherhood into all maternal and child Cambodia’s maternal mortality rate3 health and FP activities (MOH 1997a). is 472 per 100,000 live births (CDHS The implementation of all safe mother- 2005), a figure that has not improved hood and family planning activities is since the 2000 DHS, and is one of the overseen by the National Reproductive highest in the region. There is broad Health Program under the direction of agreement that a significant number Dr.Tung Rathavy. of maternal deaths could be averted by preventing unwanted pregnancies and eliminating unsafe abortion. At present, contraceptive p r e v a l e n c e there is very limited capacity in Cam- 1.2 a n d u n m e t n e e d bodia for the public sector to provide safe abortion care (IPAS 2007), despite The first recorded contraceptive Cambodia having one of Asia’s most prevalence in Cambodia was 7 per- progressive abortion laws, defined in cent in 1995 (MOH 1995b). The lat- the 1997 Abortion Act4. There is also est estimate of Cambodia’s Contracep- high unmet need for family planning, tive Prevalence Rate (CPR) is available with the 25% reported in the CDHS from the Demographic and Health 20055 likely to be an underestimate. Survey (DHS) 2005. The DHS puts the CPR at 27 percent for modern meth- This research was undertaken as part ods. There are, however, important dif- of the DFID6 funded Reduction in Mater-

2 Ministry of Health (undated) Reproductive health in Cambodia: A summary of research findings 1990-1998. 3 Maternal deaths are defined as any death that occurred during pregnancy, childbirth, or within two months after the birth or termination of a pregnancy. This time-specific definition includes all deaths that occurred during the specified period even if the death is due to non pregnancy-related causes. 4 The Act allows women to have an elective termination of pregnancy up to 12 weeks of gestation and, for pregnancies greater than 12 weeks, elective termination is permitted in the case of foetal abnormality, risk to the woman’s life, or rape. 5 Women who are currently married and who say either they want no more children or want to wait at least two years before having another child, but are not using contraception, are considered to have an unmet need for family planning (CDHS 2005). The likely unmet need amongst unmarried women is consequently not counted in this figure. 6 The UK Department for International Development.

Abortion in Cambodia I 10 Abortion in Cambodia I 11 nal Mortality Project (RMMP)7, which • Training providers in MVA, VSC and is working to increase the utilisation of IUD insertion. sustainable quality and affordable re- • Making minor upgrades to public productive and maternal health services health facilities to support provision by poor women. The project works to • Decreasing use of unsafe abortion improve the capacity of the Ministry of services through service provision, Health to provide, and to increase ac- communications and advocacy access to, safe abortion, and longer-term tivities, with a focus on the poor family planning interventions using and excluded. Manual Vacuum Aspiration (MVA), Vol- untary Surgical Contraception (VSC) Activities target the poorest and most and Intra Uterine Device (IUD) inser- vulnerable women, as they are at high- tions; in order to reduce maternal mor- est risk of maternal morbidity and tality in Cambodia. RMMP will work in mortality and face the greatest barriers 12 provinces8 principally by: to accessing services.

GBBELDG@H>C&'EGDK>C8:H

DYYVgBZVcX]Zn

EgZV]K^]ZVg 7VciZVn GViVcV`@^g^ BZVcX]Zn Hijc\IgZc\ H^ZbGZVe

7ViiVbWVc\ EV^a^c @Vbedc\I]db BdcYja@^g^ @gVi^Z

EjghVi @Vbedc\ 8]]cVc\ @Vbedc\8]Vb

@Vbedc\HeZj @d]@dc\ EgZn @VcYVa KZc\

HkVnG^Zc\ IV`Zd @Vbedi H^]Vcdj`k^aaZ Egdk^cXZhhZaZXiZY

7 For more information see www.rmmp.org.kh 8 Kampong Cham, Phnom Penh, Kandal, Prey Veng, Siem Reap, Takeo, Kampong Speu, Kampong Thom, Kampot, Svay Rieng, Kampong Chhnang, and Steung Treng.

Abortion in Cambodia I 10 Abortion in Cambodia I 11 r e s e a r c h semination, this study will contribute 1.3 o b j e c t i v e s to the wider pool of knowledge on reproductive health issues in Cambodia There has been limited research on and significantly progress understand- reproductive health services in Cambo- ing in a number of key areas. dia from the point of view of potential services users (the demand side). This qualitative study explores the current r e s e a r c h m e t h o d : context of abortion and family plan- 1.4 o v e r v i e w o f p e e r ning from the perspectives of women themselves, including: PEER is a participatory qualitative research method involving ordinary • Perceptions and practices relating to members of the target group. Fifteen family planning and abortion women aged 15-35 years from an in- • The broader social and economic ner-city area of Phnom Penh, and 15 context in which reproductive women of the same age range from health decision making takes place, a village in Kandal Province were se- including gender dynamics and lected to carry out the research. Known relationships as peer researchers, these women took • Treatment seeking and decision part in a four day participatory training making workshop where they developed inter- • Barriers to access and social exclusion viewing skills and designed an inter- • UnderstanDing of legal entitlements view schedule according to what they to services felt were the most important issues • Client/provider relationships. in their communities. RMMP worked with consultants from the PEER Unit9 The PEER method was chosen for its at Options, together with a local NGO, capacity to promote the participation CHEMS10, to supervise the workshops and capture the voice of hard to reach and collect data from peer researchers. populations around sensitive issues. Findings will support design of behav- After training, peer researchers car- iour change communications and ad- ried out in-depth interviews with two vocacy activities, and will also inform or three friends in their community on the supply side of the programme. three different topics (resulting in six Communications will be developed to to nine interviews collected by each inform people of the existence of, and peer researcher). Supervisors from the their entitlements to services, and to research team met with peer research- create demand for these services over ers to collect their findings in a series unsafe options. Through research dis- of debriefing sessions, making detailed

9 For more information see www.peer-method.com 10 Cambodia Health Education Media Service www.chems.org.kh

Abortion in Cambodia I 12 Abortion in Cambodia I 13 notes of the narrative data that peer tive topics and is suitable for groups researchers had collected. At the end with low literacy of the data collection period, peer re- • PEER builds capacity of local im- searchers reassembled in their original plementing partners to carry out groups and gave feedback on their ex- research in future periences, and helped analyse the data. • By participating in PEER, peer De-briefing notes and the outputs of researchers become “lay experts” in these workshops form the final dataset, the research issues, and form a pool which has been translated into English of expertise who can be involved and analysed by the social scientists in future stages of the programme from the PEER unit. Further details (e.g. materials and message testing, the PEER method, including selection monitoring service quality). of peer researchers, are provided in appendix 1. The PEER method has several advantages over other methods of formative The PEER method was chosen for the research. Focus group discussions often following reasons: produce normative statements (which refer to what people should do accord- • It generates in-depth, contextual ing to local norms) or reflect dominant data on a wide range of issues. voices within the group. Quantitative • Rather than collecting superficial sample surveys are useful for many facts about behaviour and prefer- purposes but cannot explain the how or ences, the method generates a de- why of social issues. In addition, people tailed “insider perspective” on are often unable or unwilling to talk the whole context in which deci- about sensitive issues openly in front sions are made. of focus group moderators or survey • Existing relationships of trust be- interviewers. tween peer researchers and their informants mean that findings are In Cambodia, the PEER studies re- more detailed and insightful than if ceived ethical approval from the Na- they had been gathered by an out- tional Institute for Public Health Na- side researcher. tional Ethics Committee for Health • PEER involves the participation of Research (NECHR) in October 2007. the target group from the early stages In the UK, the PEER method received of the programme, building owner- ethical approved from the University ship and involvement of Wales Swansea Ethics Committee in • The method is particularly suitable July 200711. for carrying out research on sensi-

11 For more information about the ethics of PEER visit www.peer-method.com/ethics

Abortion in Cambodia I 12 Abortion in Cambodia I 13

2. b a c k g r o u n d

reproductive a n d m a t e r n a l 3.1, compared with 2.5 in the capital h e a l t h in c a m b o d i a city.

This is a brief overview of the context Contraceptive use remains low, though in which the PEER study took place. For levels are increasing. The percentage of detailed background documentation on married women using any contracep- reproductive health in Cambodia, visit tive method increased from 18.5% in www.rmmp.org.kh. 2000 to 40% in 2005 (CDHS 2005), however, the percentage using a mod- The total fertility rate (TFR) in Cam- ern method is substantially lower at bodia has declined from 4.0 births 27%. The daily pill is the most com- per woman in the CDHS 2000 to 3.4 monly used modern method (11%), births per woman in CDHS 2005. There followed by injectibles (8%). The most are large fertility differentials between common traditional method is with- urban and rural areas with urban areas drawal (8%). having lower TFR: in Kandal the TFR is One reason for the low contraception

Abortion in Cambodia I 14 Abortion in Cambodia I 15 prevalence rate is discontinuation of The available research indicates that use. Recent research on the issue indi- although interest exists in family plan- cates the following: ning, the needs of women are not met by current services. Many of these find- What leads women discontinuing use ings suggest that current services are of contraception?12 not user-friendly; they offer limited reassurance and explanation around • Lack of easily accessible services and health concerns, are difficult to access, counselling leads to intolerance of side and offer limited choice. effects: drop-out rates reflect supply constraints Abortion was legalised in Cambodia in • Women saying they are “too busy” 1997 (see box below). However, there reflects limited availability of services is currently a lack of awareness regard- (far from women’s homes or short opening the legality of abortion among the ing hours) general population and providers, and • If contraceptives are discontinued, a lack of provision of safe abortion ser- women rarely switch to another method, vices (IPAS 2007). even though they mostly planned to use them again The CDHS 2005 reported that 8% of • Two thirds of women did not know women of reproductive age had had when they needed more pills or another at least one abortion during their life- injection. time. Among those who had an in- • Lack of knowledge about long term/ duced abortion, 44% had had more permanent methods than one. There is likely to be consid- • Female sterilisation thought to be too erable under-reporting of abortions as expensive, and many misconceptions most occur outside the formal health about its effects sector and abortion statistics are not • Side effects that interfere with work are currently reported to the Health In- unacceptable to women formation System (HIS)13. Additionally,

MAIN POINTS OF CAMBODIA’S 1997 ABORTION LAW Women may request abortion on demand for pregnancies up to 12 weeks Over 12 weeks they must meet certain criteria including serious health problems with the foetus, danger to mother’s life, or in cases of rape or incest (with woman’s consent and agreement of physicians) Abortions may only be performed by authorised secondary midwives, medical doctors or medical assistants, in facilities authorised by the MoH Medical personnel must counsel pregnant woman to the possible dangers of abortion, and the importance of “birth spacing” services

12 RACHA 2000 and Ministry of Health Cambodia Family Planning Survey 2005. 13 Routine data collection system from government facilities

Abortion in Cambodia I 14 Abortion in Cambodia I 15 women are often unwilling to report haviour such as treatment seeking. A re- abortion, particularly if they are un- cent summary of gender issues (DFID married. Further key findings from a 2005) described women’s status in recent needs assessment of abortion Cambodia as among the lowest in Asia. services in Cambodia are summarised Although they contribute dispropor- in the box below. tionately to the economy, women are disadvantaged in numerous ways: edu- KEY FINDINGS FROM ABORTION cationally, socially, legally, economically SERVICES NEEDS ASSESSMENT and politically. (IPAS 2007) The health sector in Cambodia is plu- Only 47% of hospitals, 10% of ralistic, with many types of provider high-level health centres and 5% and treatment: formal and informal, of low-level health centres report- regulated and unregulated, public and ed availability of abortion services private. The first port of call for information and purchasing medicines are Among the facilities that pro- pharmacies and drug sellers, especially vide safe abortion services, nearly for poor and young people, for whom half (42% of hospitals, 44% of private clinics or large hospitals may be health centres) refuse services to inaccessible for a range of reasons. adolescents A recent research summary has de- 40% of providers from hospitals scribed reasons for the lack of trust in believe that elective abortion is the public sector (MoH, undated). In not permitted by the Ministry of addition to charging user fees which Health. are set by the facility, public providers are also widely reported to charge Gender roles and norms affect many informal fees, making the total cost of aspects of women’s and men’s repro- treatment unpredictable. Drugs and ser- ductive health: they shape decision vices are often not available, and service making and communication processes, may be slow unless extra money is paid financial and social resources, and be- to get attention (Wilkinson 2003).

Abortion in Cambodia I 16 Abortion in Cambodia I 17

3. m a i n f i n d i n g s

The following chapters present key findings from the PEER study. Quota- tions from the peer researchers are included in italics to illustrate important points. They may have been edited for clarity.

w h o h a s a b o r t i o n s , were a recurrent and consistent theme 3.1 a n d w h y ? in the data, reflecting the low contraceptive prevalence rate and ineffective Understanding and characterising the use of contraception (discussed further changing dynamics of sexuality and re- in section 4.5.1). productive health in Cambodia presents Women reported making decisions major challenges, particularly in terms based on a range of factors, often of quantifying trends where cultural strongly related to social and eco- expectations and norms make underre- nomic issues beyond their immediate porting of certain behaviours likely. This control, such as poverty or social pres- study attempts to understand the nature sures. It is likely that socioeconomic and drivers of various behaviours, but changes in Cambodia are impacting makes no attempt to quantify them. significantly on youth sexuality and sexual behaviour, with implications The determinants of abortion are for unplanned pregnancies and unmet complex, and unplanned pregnancies need for contraception.

Abortion in Cambodia I 16 Abortion in Cambodia I 17 y o u n g u n m a r r i e d w o m e n women received support ranging from presents (teddy bears, mobile phones) The PEER data describe clearly the dif- to financial help with the rent. ferent types of relationships that young women had with men: while some Sex before marriage is good because the man will were said to avoid sex before marriage, be responsible and he will tell the parents about the others had a sweetheart (songsar), or relationship and when the parents know they have became a mistress to a man, or had sex had sex, they will arrange the marriage for them. for money. Factory workers and beer promotion girls were frequently said Although pre-marital sex is widely to be involved in commercial sex work. socially disapproved of in Cambodia, The overall picture from the PEER data the PEER data suggest that it is com- was that young women had a degree monplace for unmarried women to be of sexual freedom. However, this came involved in sexual relationships. Wom- with many risks, including damage to en described the enjoyment of sex, and the young woman’s reputation, con- the fact that having sex helped to secure tracting an STI, becoming pregnant, a relationship (women worried they and being exposed to risks of rape and would lose their partner if they did not violence. sleep with him).

Women had relationships with men Some get a phone number from a friend and ask for many reasons: they might be trying for love or start a relationship. For the man and to find a husband (in which case get- woman who know each other from talking on the ting pregnant could be a step towards phone, if they get, along they agree to be songsar (sweethearts) and agree a date, “Sex before marriage is good because and then after drinks they go to the guest house, because it’s the man will be responsible and will tell cheaper than hotel. the parents about the relationship...” There were many stories securing marriage, though this was in the data about women being coerced a risky strategy), they reported want- into sex through alcohol or drugs, or ing fun, sexual pleasure and romantic through persuasion, trickery or force. love, or relationships that provided eco- Risk of rape was a grave concern, and nomic support. In reality, it was usu- there were many stories about the dan- ally a mixture of these factors. Many gers of coercive sex even in long term young women have insecure, low paid relationships. employment in garment factories, bars or karaoke clubs (possibly all three), There are thus both external pressures or are involved in small scale trad- and internal motivations for women ing. Through relationships with men, to have sex before marriage. External

Abortion in Cambodia I 18 Abortion in Cambodia I 19 factors were often related to poverty, ily might abandon her or even become which could lead to women relying violent towards her. on male support. They could also be related to changing social norms and Pre-marital sex was said to be “against trends, which offer women alternative tradition” and some said it “angers the aspirations for romantic relationships ancestors”. Women also wanted to save and material possessions. Rural to ur- money and establish a stable relation- ban migration is another factor, as it ship before they had children. Child- leads to young people living away from care and the expense of bringing up the social structures and supervision a child without a husband or family that may have limited sexual activity in support, made the option of having the the past, while also increasing young child seem unfeasible. people’s need for financial resources. The urban group in particular report- Without effective contraception, pre- ed a period of time during late teens marital sex often leads to unwanted where they sought to establish a finan- pregnancies with the male partner cial foundation, and future marriage unwilling to accept any responsibility. partner; both objectives often involving “Abandoned” women in this context sexual relationships. However, to see have limited options. Pre-marital preg- this as purely transactional would be to nancy led to shame and loss of reputa- underestimate an important social, and tion for the woman and her family, and sometimes pleasure dimension to the problems in securing her a good mar- interactions. riage. In this situation a woman’s fam-

Abortion in Cambodia I 18 Abortion in Cambodia I 19 m a r r i e d w o m e n w h o and investments were required for each d o n o t f e e l a b l e t o h a v e a c h i l d child. Peer researchers described how in the following circumstances women According to the latest demographic might choose to abort rather than go and health survey, the more living chil- through with a pregnancy: dren that women have, and the older women are, the more likely they are to Short birth interval since have had at least one abortion (CDHS the last child: 2006). This partly reflects increas- This is thought to make the breast milk ing exposure to unplanned pregnancy sour, and to make the existing child fall over age. However, reporting biases, ill and lose weight. A short birth inter- particularly the fact that older women val reflects badly on the mother, who is are more prepared to report abortions thought of as ignorant or poor: make these data unreliable (the inter- pretation of surveys is discussed fur- When a woman is pregnant and also has a young ther in section 5.3). There was strong child, then her young child also has morning sick- evidence in the PEER data that married ness. She feels ashamed that other people might women might have abortions at any criticise her that the distance between the two stage of their childbearing career, for a children is short. In three years one woman had number of reasons. two children, and then her elder daughter got sick, and got thinner and thinner…The short birth The physical experience of pregnan- space is a reason for having an abortion. It isn’t cy was seen to be very tiring, difficult because they don’t want a child, it’s because they and risky. Women recognised that sig- think it will harm their youngest existing child. nificant short and long term resources

Abortion in Cambodia I 20 Abortion in Cambodia I 21 When other people know about this [short birth There was a woman who wanted a baby because interval], they mock the woman… ‘Why don’t she knows her husband loves children. But when you use birth spacing? pregnant her husband says, as an insult, that she is old and yet is still having another baby. Then she Mother has health problems: decided to abort although it was 3 months old… In many stories, doctors were said to She said she didn’t want another baby because advise women to have abortions for whenever she has a baby she argues a lot with the “pregnancy outside the womb”14 or husband, it wastes her time and money and it’s not if the foetus was thought to be dead good for her health because she is old. or deformed. Women might decide to abort if they feared the foetus had been Marital insecurity: damaged (e.g. if she had been beaten Stories about domestic abuse ap- by her husband, or if had already taken peared frequently in the PEER data an abortifacant which had not resulted and the problem is obviously of tre- in an abortion). Severe morning sick- mendous concern to women in both ness, being too old, weak or exhausted, urban and rural communities. There and heart complaints were also cited were many stories about troubled mar- as reasons why women had abortions: riages in which women felt compelled their bodies were thought unable to to abort their pregnancy. These includ- cope with the demands of pregnancy. ed husbands who beat or neglected In such cases, the decision to abort “Whenever she has a baby she argues a lot with was difficult, with the whole the husband, it wastes her time and money and family discuss- it’s not good for health because she is old...” ing whether the woman should risk having a child. their wives, as well as husbands who These reports must be understood in had other wives or girlfriends, or who the context of the harsh realities of ag- simply did not provide for the fam- ricultural life and the many simultane- ily. Some women worried that if they ous responsibilities placed on women. got pregnant their husbands would be more likely to have extramarital rela- In some cases, the doctors recommend that the lady tionships, and others worried that hav- should have an abortion because her womb is too ing more children would lead to argu- weak, if we don’t take the baby out it could injure ments at home. the woman’s health. So the parents try to get money Marital insecurity is closely linked to from relatives to pay for it. After she does the abor- economic insecurity: divorced or wid- tion, she is better, her health is in better condition. owed women struggled to make ends

14 Given the rarity of ectopic pregnancy, this is thought to result from a general lack of understanding about reproductive physiology. There were in fact many references to spurious sounding medical advice in the data, often resulting in recommended abortion.

Abortion in Cambodia I 20 Abortion in Cambodia I 21 meet. Women may not want to have a It is difficult for families with many children to child if it risked contributing to family feed them, and they don’t have enough money to problems: send them to school or higher education. They don’t have enough time to take care of them as When the lady comes back from work, she feels the husband and wife are busy working outside, very tired, and sleeps in bed. Her husband kicks so they don’t know whether their children go to her until she gives him money for drink. She is the school, or whether they escape school...so they stop breadwinner of the family so no one helps her to their children from studying and get their children earn money and the husband is a drunkard, so she to earn money to support their families. decided to do an abortion because there are a lot of problems in the family. Anxiety about costs did not only affect the poorest women, even better If the husband usually hits her, so she does the off women were concerned that an abortion secretly, without anyone knowing. unwanted pregnancy would reduce the living standards of the household. Fam- A woman was pregnant and her husband stayed ilies aspired to a certain level of material out late. She waits for him, but he is just drunk comfort and food security, which could and always hit and kicked her. She was very sick be threatened by the arrival of another and decided to abort the child, as she was afraid child. Women were often the main or she would be in danger in future. sole breadwinners in a household (men might be absent, ill, disabled or simply Poverty and “living standards”: not as successful at earning money) and There are numerous direct and in- thus pregnancy at an inconvenient time direct costs associated with having could have serious implications for the children. Poor women feared not be- household income. ing able to meet these costs. As well as school fees, food, and medical costs, w o m e n w h o h a v e c o m p l e t e d t h e i r there are opportunity costs in having f a m i l y children: not being able to work during pregnancy or the child’s infancy. Some There was widespread recognition households were in a situation where of the benefits of limiting family size. Having many chil- “They [women] don’t like services that are dren in the family was thought to lead expensive, so they abort on their own if their to arguments, being pregnancy is just early on by taking pills.” too expensive and time consuming. they could ‘only earn enough for each Women who had reached their desired day’ (i.e. could never put anything aside completed family size might therefore to save) which was a local expression to abort future pregnancies. refer to the poorest families.

Abortion in Cambodia I 22 Abortion in Cambodia I 23 a d d i t i o n a l r e a s o n s pregnancy and caring for a baby if a woman does not have good health or Work demands were frequently re- support from others. A nutritious diet ported as reasons for ending an un- with certain dietary exclusions is also planned pregnancy. There is high labour thought to be important during preg- force participation of women in low nancy, which may be difficult for poor paid, insecure, informal employment women to achieve. Rather than risking and agricultural production. Subsistence a miscarriage or unhealthy baby, wom- farmers may be unable to take time off en might choose to abort. work if other family members cannot help them. Garment factory workers may take “...she doesn’t want a baby is because she is three months maternity afraid her parents will criticise her, and won’t leave at highly reduced pay, but if no childcare let her come to work in Phnom Penh again.” is available, then women may not be able to continue working. There was a woman in the village who had two children with no problems. But she decided to A factory worker had a baby with her songsar, she abort the third child because her health was bad, asked a friend to buy a Chinese herb to abort. The and her husband couldn’t work much as his health reason she doesn’t want a baby is because she is was also bad… She had to work a lot, finding food afraid her parents will criticise her, and won’t let for the pigs she raised, and driving her moto a lot. her come to work in Phnom Penh again. She thought that if she kept her baby she wouldn’t be able to earn enough to support the baby and her A garment worker works right up to delivery, then health wasn’t good either. takes three months off, she will still get money from the factory, they give $5.50 for powdered Although some stories described cou- milk for the child for 3 months. ples making the decision to have an abortion together, women were pre- Peer researchers reported that women pared to have an abortion without tell- are meant to restrict their activities during their husband or family, as long as ing pregnancy (avoiding rough roads, they had the money. hard work etc.) to avoid harming the baby, and because they are thought Women with their own money don’t discuss it to be weaker than usual. Miscarriages with their husband just discuss it with their friend, are often blamed on the carelessness as if they discuss it with their husband he’ll want or overwork of women. Yet the work to keep her baby. They rarely discuss it with their women do in villages is hard: tending parents, as their parents think about karma. Some- fields, herding animals, and carrying times they have secret abortions without anyone loads. In rural areas, the need to keep knowing, because they think they’ll be criticised working is seen as incompatible with for not being able to support their baby.

Abortion in Cambodia I 22 Abortion in Cambodia I 23 Her family might also influence what might limit demand for these ser- whether women aborted or not: vices in future? There is a woman in this village whose husband The data indicated that there are many hit her a lot during her pregnancy. She discussed factors that influence women’s choice with her relatives whether to keep her baby or of provider: gestation of pregnancy, abort it. Her relatives and her parents suggested financial resources, perceptions of ser- that if you face difficulties, you should abort it, vices (confidentiality, cleanliness etc), because you have three children already. and recommendations from friends.

When she was five months pregnant she had heart Women buy pills from the clinic in the local disease…Her husband took her to check with the market. They use pills when the pregnancy is doctor, who said that if she wanted to keep the two months; after this they have to abort in the baby it was up to her, but she had to take care of hospital. her health. She shouldn’t work hard and should eat nutritious food. He recommended that if she kept They don’t like services that are expensive, so they that baby, it would be difficult for her to deliver abort on their own if their pregnancy is just early it… She discussed this with her husband and rel- on (e.g. 2 months pregnant), by taking pills. atives... They recommended that she should keep it, because they thought about karma: they thought As pregnancy went on, methods were that this baby had life, and if they killed it, it reported to get more expensive. Access would be a sin. So she decided to keep it. depended on women either having their own funds, or social networks from which to borrow or be given money. w h y d o n ’t w o m e n u s e For most health problems, women and 3.2 p u b l i c a b o r t i o n s e r v i c e s ? their families try to mobilise a wide range of sources of support to secure One of the objectives for the PEER the funds necessary for treatment. studies was to examine “barriers to ac- However, women seeking abortion are cess” to public abortion services. This unwilling to visit neighbours and family to borrow or ask for “Her relatives and her parents suggested money. The fees reported to be charged that if you face difficulties, you should abort it, for abortions ranged because you have three children already.” from under $10 for an abortificant pill, to must be understood in light of our $20-$40 for a surgical abortion in the findings that there is little evidence that first trimester, to over $100 for an abor- women want to access public sector abortion at the “big hospital” in the second tion services. A clearer way of framing trimester. Poorer women, women who the question is therefore, why do wom- cannot tell their friends or family about en not use public abortion services, and the abortion, and women unable to ask

Abortion in Cambodia I 24 Abortion in Cambodia I 25 or borrow money are therefore barred Sometimes women find it hard to decide which from accessing many services. way they should abort. If they do it in the clinic they will spend a lot of money. They try to run One scenario that emerged from both fast, reach high objects or carry heavy things. After rural and urban areas is that women try doing so the blood comes out and they feel pain. several methods in sequence, trying to No matter how hard it is, or even if they might find one that works: die, they will still try to abort. They chose that way to do the abortion because they don’t have A lady was pregnant and busy with work. She had enough money to raise the children and they have no money to abort at hospital. She tried to drink no choice. some kind of herb, and to run at high speed, and tried to reach high-up objects, but the baby was The data indicate that women com- still there. She did this again, and blood came out, monly go through a series of steps at- but the baby was still there. Finally she tried to tempting to induce an abortion before borrow money from people to abort the child. She seeking outside help. This has serious said that if she kept the baby she couldn’t earn implications as both attempting to self money, and she didn’t have enough money to raise induce, and presenting late, increases that baby. risks to women’s health. There is urgent need for communications ad-

Abortion in Cambodia I 24 Abortion in Cambodia I 25 dressing this issue, informing women Complicated systems: of the benefits of promptly seeking a Women did not know where to go, or pregnancy test and attention at an ap- what to do, at big hospitals (e.g. where propriate facility if they have an un- to register, pay, collect drugs). They said wanted pregnancy. these hospitals required accompaniment by an experienced friend or neighbour, l a c k o f d e m a n d f o r c u r r e n t often incompatible with abortion. p u b l i c a b o r t i o n s e r v i c e s Provider attitudes: Women were unlikely to attend a pub- Women said that poor people and un- lic health facility if they wanted exter- married women would not be treated nal help to abort. Local public facilities kindly at hospital, and might be shouted at and “blamed”. An- They have to say that they can’t keep the baby. other common problem was that some Because if they don’t lie, the doctor won’t providers were un- abort the baby... willing to carry out abortions without in rural areas are rarely appropriately “good reason”: staffed or equipped to conduct abortions [IPAS 2007], which may be why When they go to abort in large public hospitals in women in Kandal (the rural PEER site) Phnom Penh, they have to lie to the doctor because did not even mention the possibility of it’s a government hospital. They have to say that going to the local facility for an abor- they can’t keep the baby. Because if they don’t lie, tion. However, the urban peer research- the doctor won’t abort the baby. Some lie that their ers did occasionally mention women husband harms them very much. who visited rural health centres to have abortions so that no one would know They don’t like the big hospital or big clinic be- about their abortion. Most of the hospi- cause they are poor and they get blamed and there tals discussed in the PEER studies were is no special care for them. large hospitals in Phnom Penh. The main factors that dissuaded women A woman didn’t want to go to a hospital because from going there to abort were: it’s far away and the staff are impolite. They don’t pay attention to the customer, and they say the Poor women felt customer is dirty and unhygienic. they were not “for them”: There was a common view that hos- Lack of privacy: pitals were for rich people, who could Large public hospitals are not designed afford to pay the fees and extra money with confidentiality in mind. One of required to get attention from the doc- the strongest issues emerging from the tor and be attended too quickly. PEER data is that women do not want

Abortion in Cambodia I 26 Abortion in Cambodia I 27 to be recognised or seen when seeking the contraceptive pill to attempt abor- an abortion. tion. Women recognised risks associated with these pills, including hae- Some people rely on legal clinics with modern morrhage, incomplete abortion, and equipment, but they are afraid people will know weakness and fatigue. It is impossible them. The mother is afraid it will affect the family to tell in how many cases women used reputation and no one will marry her daughter. the WHO recognised combination of mifepristone and misoprostol, or what Lack of preferred method: regimens are followed. Taking abortificant pills was a widespread method of attempting abortion, With existing and proven technology these ranged from traditional herbs available elsewhere, the current lack to modern pills with various collo- of a formally registered, regulated and quial names. These were not available in supported medical abortion drug in hospitals. Cambodia is a serious concern, putting Cambodian women at unnecessary risk. Geographical distance: However, the data do indicate that were Women from the more rural group an appropriate product to be licensed in reported prohibitive transport costs to Cambodia, women would find it highly reach hospitals in Phnom Penh. acceptable. The unwillingness to seek professional help and advice from high- o t h e r a b o r t i o n m e t h o d s er level providers necessitates strong re- a n d p r o v i d e r s a r e p r e f e r r e d ferral networks with pharmacies.

In addition to low demand for public Medical personnel services, numerous competing services working informally: and products for abortion exist: Qualified medical personnel such as midwives, (or unqualified person- Abortificants: nel with medical experience) may Numerous types of pill were avail- visit people at home or operate out of able for women seeking medical abor- small private clinics to conduct surgical tion. Peer researchers were uncertain abortions. about their names: the following were mentioned but not consistently: 11- Small private clinics: Tiger, the Chinese pill and the French Many different types of private clinic pill. Drugs were usually bought from a were described. They were often small, pharmacist, but were also bought in the and might be located at the back of a market and from drug sellers. Some- pharmacy or in someone’s house. They times friends bought them for women could be staffed by an unqualified per- wanting an abortion. Some stories de- son, medical student or paramedical. scribed women taking high doses of They are the provider of choice for the

Abortion in Cambodia I 26 Abortion in Cambodia I 27 urban group if self induced abortion heightened risk awareness about this failed. However they were not men- method of abortion: tioned as frequently in the rural data, probably because there were no such Some women go to the traditional midwife to clinics in the village (the nearest was abort, but sometimes women die if she cannot get 10km away). Known by reputation, all the blood and foetus out. they provide a quick, confidential, no- questions-asked service at a cheaper They don’t like the traditional midwife because price than public hospitals. Women did she aborts by pressing into the abdomen to break not appear to distinguish providers at the neck in the womb, or penetrates the womb, and these clinics by qualifications or com- this can kill the mother. Sometimes she loses blood petencies: all paramedical staff are re- but the baby remains in the womb. Because of this ferred to as “Doctor”, even though they the baby becomes spoiled in the womb and causes may not be qualified. the mother to die as well.

The pharmacist’s wife will have an illegal clinic at Traditional herbs/alcohol: the back of the pharmacy. These are bought from the market or from a local person. The PEER data con- At the private clinic they don’t ask anything. If a tained numerous references to herbs woman goes to abort her baby, then they just do which make women “hot”, often taken it for her. in wine. Heat in the body is thought to induce abortion. These methods The small clinics do respect confidentiality, because form part of the attempts to self-induce there are only two people in the place, the doctor abortion that women might try before and the girl themselves. Mostly they are profes- seeking services at a clinic. sionally trained, they work for big hospitals. People think if they go to hospital, they may spend $30 For ladies who are poor, when they want to do for an abortion; at the doctor’s house, they only abortions, they just take traditional medicines. A pay $10 and the equipment is the same. lot of blood comes, for nearly the whole month, then the period comes and she also has that for Most women like to go to private clinics. They seven days, and then that lady is very exhausted rarely go to big hospitals because the doctor there because a lot of blood has been lost. doesn’t agree to abort them. But if they go to the private hospital and give money to the doctor, the Some people do abortions by taking hot and strong doctor helps them to abort. wine. Sometimes they put a herb (11-Tiger) with chilli and put into wine. Traditional nurse/midwife: Are said to conduct abdominal mas- Abortion versus post-abortion Care: sage to induce abortion. Stories about The pattern of abortion seeking as de- traditional midwives typically described scribed above means that women will them negatively, suggesting there is frequently present at a clinic having

Abortion in Cambodia I 28 Abortion in Cambodia I 29 already attempted an abortion one or also thought to affect quality of care more times. They may thus require post provided in public facilities, as women abortion care rather than initiating an perceived there to be a lack of respect abortion for the first time. Post abortion towards poor women (as discussed in care and abortion care are not distin- section 4.2.1). guished in women’s accounts. The quality of service is also consid- s u m m a r y o f ered. The following quotations describe f a c t o r s a f f e c t i n g c h o i c e what women like in abortion services:

Several factors affect women’s choice Women like to have abortion at safe places, where of abortion method and provider: people know how to console you, and make sure that the place is quiet, and no one can see, and the Financial resources are a key deter- doctor there must keep the secret. Some don’t like minant. In stories reported by peer to go to the clinic, it’s not a suitable place, and researchers, women raised the money people may not look after them if they don’t have needed for surgical abortion (which enough money for them. are more expensive than self induced or medical abortion methods) through A student went to a small clinic because she selling land, pawning jewellery, prom- thought people there would not know her and ising their future labour, or spending would keep her secret. She trusts them, it’s cheap capital from small businesses. However, and they don’t need her to register, so people will some women were unable to borrow not know her name. If she aborts in the hospital, money without “security” (e.g. land she will be ashamed. The place is big, and has a lot against which to borrow) or did not of people, and it’s not secret. She doesn’t want other want to get into debt. people to know she is pregnant, and she is afraid people there would know her. Some doctors are not Women borrow money with interest from others if friendly and they always take the money before they want an abortion without telling their hus- providing the service, and if the women don’t have band. They pay it back later. The interest rate for money, they don’t look after them. $25 is $5 a month. If they can’t pay the interest they have to help the owner harvest rice. They like services that their neighbour used to use, so they know that it won’t cause any problems to The price of abortion should not be their health. seen as a simple barrier to access, but as a determinant of provider and method They like going to the NGO clinic because the choice. Women do not necessarily con- staff pay lots of attention and are polite to the cus- sider going to an expensive hospital; tomer. They provide advice about how not to have rather, they consider which services are sex like that in the future and how to take care. appropriate for them and their finan- And they don’t pay much money. Some say that cial resources. Socioeconomic status is money does not matter if the staff are polite.

Abortion in Cambodia I 28 Abortion in Cambodia I 29 They like services that have good equipment, which them (e.g. for having sex too young). are cheap, with good and modern equipment, They do not want to wait for a long and safe to their health. If they have abortions, time to be seen, and do not want to be and a problem happens, the doctor there takes recognised at the facility. responsibility. In terms of women’s preferences around abortion, several aspects of KEY ASPECTS POSITIVELY abortion care were not mentioned in INFLUENCING DEMAND the PEER data: including physical pain, pain relief, or distinction between MVA Staff can be trusted (e.g. recom- (manual vacuum aspiration) and D&C mended by social network) methods. This is not to say that these Affordable might not be potentially important as- Look clean with modern equip- pects of care. At present women may ment accept pain management (or lack of Have a set-up that allows con- it) and type of surgical method used fidentiality as resting purely in the hands of the Staff should be polite, skilful, provider and therefore it is not seen as cleanly dressed, and responsible worthy of complaint or mention. Provide advice / consolation / reassurance The range of qualities both desired Provide a quick service without and undesired in relation to abor- asking questions tion services suggests that there is low awareness among women of how to evaluate the safety and quality of technical procedures. Rather vague crite- In summary, safety and other technical ria against which to judge risk and aspects of abortion care are not voiced safety include recommendations from as the primary concern: cost and con- friends, the “skilfulness” of the doctor fidentiality are seen as more critical. (which was not linked to qualifications Women have little sense of expectation or any other kind of official verifica- or entitlement to counselling, follow- tion), and “modernity” of equipment. up, or referral. Many spurious diagno- Women may want modern equipment, ses by “doctors” are accepted as fact. but their definition of modern may be very modest. More effective criteria for The aspects that women do not like judging safety need to be made available about abortion services are often the to women if they are to be encouraged converse of these. They do not like doc- to make use of skilled providers. In Ne- tors who only care for them properly pal for example safe abortion services when they have money. They do not are branded. Whilst branding for each want doctors to shout at them or judge clinical service may be confusing to

Abortion in Cambodia I 30 Abortion in Cambodia I 31 3.3

consumers, there is a need for effective There was no expectation that the poor quality assurance combined with commu- should be entitled to access to abortion nications to support decision making. services. If they could not pay, they sim- A quality assurance branding of a mini- ply did not expect to receive attention. mum reproductive health package may This was also a realistic expectation of be more appropriate. what women expected from providers of other health services. Women ex- l e g a l i t y a n d r i g h t s t o a b o r t i o n pressed frustration with this situation, and appreciated that they should not The abortion law was not mentioned have to pay additional fees in order to directly in the PEER data, though some be seen by providers, but they were also women were aware of the illegality of fatalistic about the likelihood of the certain clinics and abortificant drugs. poor receiving a good service. This did not deter women from using these services: abortion law was un- In summary, lack of knowledge about important in influencing whether and the legality of abortion does not stop how women chose to have an abortion. women seeking an abortion. However, Likewise, the concept of entitlement lack of knowledge about entitlement to abortion was absent from the data. to abortion is a barrier to women de-

Abortion in Cambodia I 30 Abortion in Cambodia I 31 manding and accessing abortion from al outrage or stigma around abortion qualified providers. Yet simply knowing from the perspective of peer research- that they are entitled to abortion ser- ers or their informants. However, wom- vices will not be sufficient for women still clearly want to “keep it quiet” if en to claim their entitlement unless they do have an abortion. There is a lot mechanisms are in place to ensure that of gossip about abortion which no one providers fulfil their duties to provide wants to be the focus of. There were services. a few examples of older members in families saying that it was ‘bad karma’ to abort. However, this did not deter w h a t d o w o m e n t h i n k women from having abortions if they 3.3 a b o u t a b o r t i o n ? felt this was their only choice.

s t i g m a Some still think about karma. Old ladies say that if the woman kills her baby, she will pay in the The PEER data were full of pragmatic, next life. everyday accounts of why women had Rather than being attached to abortion in itself, stigma “ They go in the day or in the morning... was described in relation If they have an abortion in the morning, to other parts of abortion they can go to work again at night.” stories: providers chastising women, families becoming abortions. Many women discuss abor- angered and women feeling ashamed. tion as a routine event, indicating a For unmarried women, pre-marital sex relatively high frequency, and low risk was shameful consequently so was any perception. subsequent abortion. For poor women, the fact that they could not afford They go in the day or the morning. If they go at to bring up another child, rather than night, the blood may come out more. If they have the fact that they ended the pregnancy, an abortion in the morning, they can go to work made them want to keep it a secret as again at night. they were ashamed to be so poor. In cases of ill health, or when women Peer researchers sympathetically de- were threatened by their husbands, or scribed difficult situations that women when women already had several chil- found themselves in, and abortion was dren, abortion had very limited stigma seen as a valid response to these cir- attached to it indeed. They could discuss cumstances. This widespread under- their decision openly with their family, standing indicates that abortion is not and could go to a clinic without shame highly stigmatised amongst women. or fear of being scolded by staff. There was very little evidence for mor-

Abortion in Cambodia I 32 Abortion in Cambodia I 33 r i s k p e r c e p t i o n such as morbidity, infertility and death, the main concern seemed to be general Women recognised risks associated weakness following abortion. In ad- with traditional abortion methods and dition, risks were thought to increase to a lesser extent with “modern” abor- with repeated abortions, rather than tions, although their risk perceptions being present in any one procedure. differ from biomedical models of risk. They recognised that unsafe or incomplete abortion risked wasting time and w o m e n ’s perceptions money, because you might end up in 3.4 a n d experiences hospital. o f f a m i l y p l a n n i n g

Commonly cited perceptions The PEER findings include insights of the consequences of abortion were into both the benefits of having children, and reasons why women want to • Womb cancer or thinning womb from space and limit births. Having children repeated abortion was seen to make the family happy, • Bleeding and complications keep the husband “faithful and steady”, • Weakness, fainting, becoming pale, and motivate children’s parents to work getting thinner and thinner hard. Children could be difficult to • Fertility problems in future care for while they were young, but were a source of support in sickness Some people’s wombs are affected by abortion and and old age. it can get thinner and thinner, and eventually get cancer. When faced with this they need to see the There was widespread recognition of doctor otherwise they will die. They should go to the benefits of limiting family size for hospital, but don’t want to go because they can’t social and economic reasons. Peer re- afford it. They just drink herbs, if it’s not effective searchers collected a large amount of or they don’t recover, they just let it be, or die. data about family planning behaviours and were familiar with a range of meth- Interestingly, many of these perceived ods and their perceived advantages and risks are similar to those associated disadvantages. Various reasons were cit- with various methods of contraception ed as to why women at different stages (discussed further in section 4.4): both of their lives used contraception: contraceptives and abortion are seen to have negative health consequences. • To ensure a healthy birth interval • Family complete: no desire for more Particularly in the rural area, there children was low risk perception of the specific • Delaying the first birth to earn some physical dangers associated with unsafe money before or after marriage abortion. Rather than recognising risks • To avoid pregnancy before marriage

Abortion in Cambodia I 32 Abortion in Cambodia I 33 • In the case of condoms: to avoid with more negative side effects. When STIs women talked about taking pills, they might not necessarily mean the daily The “injection” contraceptive pill: other pills are on was the most popular method. Wom- the market, under names such as Tiger en liked it because it only needed to be 11 and the Chinese pill, both of which administered once every three months were also named as abortificant drugs. and was thus easy to remember. It was A variety of pharmaceuticals were sold also said to be better for older women without women knowing exactly what (over 30 years old) because it caused they were taking.

In some instances, “When they use pill they become fatter and women used the pill on fatter, and they are afraid that fat will cover an ad hoc basis: for in- their heart and they will die.” stance, taking it for a few days immediately before menstruation to become irregular or expected sexual activity, or taking large to stop, which was not desirable for doses to try to induce an abortion. It younger women. was thought to be better for younger women as it made their periods regular. They prefer the injection as it lasts for three Becoming fatter was another common months. If they use the pill every day, sometimes side effect: sometimes this was seen in they forget about it because they are very busy a positive light, indicating good health, with work. whereas in other cases the fat was thought to accumulate around their Her sister uses the injection. Before, she had a heart, which was not good for health. womb problem – but after the injection her womb became better. Some women when they use the pill, it doesn’t work, and they can make her fatigued, thin, and It is good for their health and effective. After injec- pale, so they decide to use the injection instead. tion, they become fat. Before they were thin. Some women when they use the pill, it can make them fatter and fatter, and her health becomes Although this was the most popular good. method, there were still problems reported with its use, including irregu- When they use the pill they become fatter and lar bleeding, and women not feeling fatter, and they are afraid that fat will cover their themselves. heart and they will die.

Contraceptive pills A woman had four children didn’t want any more, were mentioned almost as regularly so she took the pill. Her husband worked far away. as the injection but were associated She bought a sheet of pills. After taking the sheet

Abortion in Cambodia I 34 Abortion in Cambodia I 35 of pills she stopped using it. One day her husband it to be the third most commonly prac- visited her home. Seeing her husband coming she ticed method of avoiding pregnancy bought more pills and took only two pills. As she (after the contraceptive pill and injec- was afraid of getting pregnant she took more pills tion). “Water out of the pot” was a term and continued using it every day, but one day her used covering both the withdrawal period didn’t come… she checked her urine and method, and douching the vagina after found out she was pregnant. sex to remove semen. These appeared to be very popular methods amongst sex- Traditional methods ually active young women in the in ur- The main methods mentioned were ban area, though “water out of the pot” drinking alcohol (home made wine), was often recognised as unreliable. often with herbs, and the withdrawal method. After having sex, the lady sits there to try to get the bodily fluids out of the vagina and clean it with They used traditional medicine in the past such water, which avoids getting pregnant. The prob- as papaya milk with alcohol, but now they have ability of having a baby is just small. stopped using that medicine because now you can find the pill. The older generations especially like to take water out of the pot. Now people like to use this method, People drink alcohol: they use a kind of fruit because when they take the pill or injections, they (popai) mixed with alcohol. They have to drink a glass of this alcohol three “...They have to drink a glass of this alcohol times per day. When they 3 times per day. When they drink it their body drink it their body feels feels warm and then they can’t be pregnant.” warm and then they can’t be pregnant. get a lot of flesh around the heart, and if they want to have a baby next time, it’s difficult to have, be- Many of the stories of traditional herbs cause of all the flesh. and wine relate to how things were done in the past. Traditional medicines IUD for contraception appear to play a rela- The IUD was a well known method tively small role in overall family plan- with numerous negative associations. ning practices and it seems unlikely Rural women saw it as being for rich that the traditional medical sphere women, and often did not actually would interfere significantly with ef- know anyone who had used an IUD. forts to promote modern contraceptive Rumours associated with the IUD in- methods. cluded it being uncomfortable for the man having sex, the woman not being The withdrawal method seems to still able to have sex or work hard, height- have greater relevance: the DHS found ened risk of ectopic pregnancy, damage

Abortion in Cambodia I 34 Abortion in Cambodia I 35 to the womb, and risk of the IUD be- in the arm”, it was seen to be for rich coming embedded in the body: people.

A woman saw another woman using an IUD who Compared to other methods of family got tired and thinner and also had a pregnancy planning, “tying the womb” (tubal outside the womb. So she did dare not use it. ligation) was not mentioned very frequently. Perceptions were varied. Some With the IUD they can’t work easily, and fear said it brought good health, whereas that if they use it for a long time it will hurt the others said that it was linked to cancer. womb, only a few people use this. It was widely seen as a method for rich women. When they use IUD, people get fatter and fatter so it’s difficult to take it out, their flesh covers the People don’t like it because it makes the womb IUD. smaller, and they may not be able to have children in the future.

The person who ties their “People don’t like it because it makes the womb has good health: they womb smaller, and they may not be able to rarely get sick. have children in the future. ” They are afraid that they In this village, there are only 2 or 3 people who use will become fat and that the fat will cover their IUDs, they use this method because they have rela- heart… most women don’t like tying womb, it’s tives working in an organisation (RACH)15 which a method only used by rich women. They’re afraid provides birth spacing. that they won’t be able to work heavily.

The Doctor gave her the option of an IUD in the Only a very few rich people do tying the womb, womb, but she has heard rumours from her neigh- because they are afraid of womb cancer. Some bours about them, like if you have an IUD in your women in this village used to tie their womb, some then you can’t get it out again, and it can cause became thinner and thinner, some become fatter womb cancer. She discussed this with her husband. and fatter, so they are afraid of it. Some get sick He understood her feelings and took her to the hos- when tying the womb. pital again. The doctor gave her another choice of having an implant in her arm. She discussed it The operation is for life. You have to go to the with her husband, who agreed. big hospital. It’s only for the rich people, it costs $200-300. For those who pay it’s no problem. Implants were only mentioned a few times. On Condoms one occasion described as “an IUD but were mentioned as being used by un-

15 Reproductive Health Association of Cambodia

Abortion in Cambodia I 36 Abortion in Cambodia I 37 married women, and women advocat- they think that there is virus in the condom. ed using them when sex with a hus- Vasectomy was only mentioned once band was risky (e.g. if he had another in the whole PEER dataset: “the doctor rec- partner: the risks of HIV and other STIs ommended that she tie her womb, or for her hus- were widely acknowledged though not band to have a vasectomy”. A very few peer necessarily understood in great detail). researchers had heard of vasectomy but In contrast to previous studies16, con- did not know what it was. dom use was not highly stigmatised, and was not strongly linked to com- Emergency contraception was not mercial sex work. However, many said mentioned in the PEER data. Probing that condoms were uncomfortable and at the peer researcher workshop con- caused womb thinning. firmed that women had no knowledge of this method. After her baby miscarried she didn’t have sex with her husband as she believed that having sex after miscarriage could cause health problems. During w h a t p r e v e n t s w o m e n f r o m these three months, her husband had sex with an- 3.5 u s i n g f a m i l y p l a n n i n g ? other girl outside. After three months, she had sex with her husband, but used condoms, because she c a m b o d i a a t a n e a r l y s t a g e was afraid that if she had another pregnancy it o f a d o p t i o n o f contraceptive u s e would abort again, and her husband could transmit a disease to her. Perceptions of family planning and behaviour around contraceptives suggest The method they don’t like to use is condoms be- that Cambodia is still at an “early adop- cause using condoms can damage their womb, and tion” stage, characterised by:

16 Sex Talk -Peer Ethnographic Research with Male Students and Waitresses in Phnom Penh. By David John Wilkinson Gillian Fletcher. PSI 2002.

Abortion in Cambodia I 36 Abortion in Cambodia I 37 • Low levels of knowledge and edu- d e n i a l o f s e x u a l a c t i v i t y cation about contraception, especially in u n m a r r i e d w o m e n for young and unmarried women. • Inaccessible or poor quality ser- Unmarried women are not well catered vices, and for long term or perma- for in family planning services and are nent methods (IUD and surgical able to access very little education about methods). Inconvenient services contraception or reproductive health in from public health centres and higher general. up-front costs for longer term meth ods meant that buying pills or injec- People don’t like single women to know about tions from pharmacies was com- contraception. mon. Services were particularly Some don’t want to go to the hospital [to get fam- inaccessible for unmarried women. ily planning] because they feel shy. If they aren’t • Numerous and often inconsistent married, they are afraid that other people will say stories and rumours about side that they aren’t married yet, yet they go to hospi- effects. tal, maybe they are having sex. • Incorrect use of contraceptives due to lack of knowledge and skills: The urban data in particular suggest e.g. cannot remember to take the pill that although many unmarried women every day. do have sex before marriage, they are • Sub-optimal provision of contra- not prepared for sex when it happens ceptives from pharmacies e.g. not in terms of contraceptive protection. providing full explanations and fol- This may be partly because preparing low-up, supplying contraceptive contraception in advance is to admit to pills in packets of ten, poor regula- yourself and others that you might have tion of pharmaceuticals. sex. This makes the pill and other lon- • Lack of choice: Difficulties for ger term methods less suitable for these women in finding a method that women. and there is a need for targeted “agreed” with their body. services and communications to cater for this group.

“People don’t like single women s i d e e f f e c t s to know about contraception. ” experienced a n d f e a r e d

These factors combine to contribute Side effects have already been discussed to high levels of doubt and suspicion in relation to specific contraceptive about contraception, a high discon- methods, but it is worth examining tinuation rate, and lowered levels of the overall impact of the large volume contraceptive protection if women are of misconceptions and negative expe- not using contraceptives effectively or riences of contraception circulating consistently. in the community. Most Cambodian

Abortion in Cambodia I 38 Abortion in Cambodia I 39 women rely on their physical health to she felt tired, became thinner and thinner, and her work and support themselves and their period flowed heavily. She had no strength to work, families. The perceived risk of ill health and she fainted. or weakness might be felt more acutely than the risk of unwanted pregnancy. Perceived side effects should not uni- The way in which women perceive and versally be dismissed as purely psy- balance risk, and use this to inform chosomatic or neurotic. Family plan- decision making must be addressed. ning services should provide a choice Communications must support women of methods in order to suit couples’ making informed choices around the needs, as women react to contracep- perceived costs and benefits of using tives in different ways. Some women contraception. experience side effects and need to try different brands, methods or dosages. In addition, a poorly regularly pharma- COMMON SIDE EFFECTS ceutical industry, combined with poor OF CONTRACEPTIVE METHODS usage and dosage delivered through FROM PEER DATA private, unregulated channels might predispose women to side effects. Getting hot inside Managing side effects more effectively, Getting fatter and fatter (fear of by providing greater support, advice, fat accumulating around heart) reassurance, and choice of methods, Getting thinner and thinner would help tackle the root cause of Irregular bleeding the rumours. Of course, many feared Womb thinning and cancer side effects are clearly misconceptions Infertility/future difficulties in and should be addressed as such, but becoming pregnant these issues should not be rejected as “imaginary”, as they are very real to the women concerned. Some women use birth spacing for a long time and later want children, but they can’t have them. In summary, there is demand from There was a woman in this village, first she used both women and men to control their birth spacing, she earned a lot of money and had a fertility, but there are also many ob- nice house, but then she stopped using birth spac- stacles to women achieving effective ing and now she can’t have children. contraceptive protection: low levels of knowledge; unstructured, unregulated She didn’t use family planning. She used to use private sector provision; inadequate birth spacing methods, but it didn’t work17. When provision within the public sector; and she used the pill or injection, she didn’t feel well, fear of side effects stemming from mis-

17 Here, contraception ‘not working’ does not mean that she became pregnant while using these methods, but that she has tried using them, but felt they were not suitable for her.

Abortion in Cambodia I 38 Abortion in Cambodia I 39 conceptions, rumours, and experiences Her neighbour asked her to send her again, but she of others. said that she didn’t have any money.

Unmarried women have a problem with dis- o t h e r reproductive h e a l t h charge… they get fever and vaginal discharge, 3.6 i s s u e s f o r w o m e n but they keep it to themselves and don’t tell their mother. They think that it is normal. Eventually During the PEER study, women dis- they will try a traditional herb to cure it. cussed other health problems that affected women like them. These included When unmarried women have discharge, they dare very common experiences of STIs, vag- not go to the doctor, because they are unmarried. inal discharge and irregular menstrual periods. There was a woman with a discharge, but was too afraid to tell anyone, because she thought the fam- There was an unmarried women who became thin- ily would blame her.. ner and thinner. She didn’t dare tell her parents she had discharge. One day her mother asked her “what An unmarried woman had irregular periods, and happened to your health?” She told her mother that was feeling hot inside, so she asked her mother to she had white fluid flowing out. Her mother didn’t take her to hospital. Village women said it was be- have money so she borrowed money from others cause of her having sex and pregnancy. and took her to the NGO clinic. Now her health became better, but she is still not well yet… The The data illustrated several important discharge is still flowing… and she became thin- points about reproductive health issues ner and thinner as her period flowed irregularly. for Cambodian women. Firstly, among

Abortion in Cambodia I 40 Abortion in Cambodia I 41 unmarried women in particular there is olence, gender, human rights, and HIV. reluctance to seek treatment for symp- Women have learnt to “talk the talk” toms such as vaginal discharge. This is about these topics, but with limited partly as discharge may be considered comprehension of what it all means. normal, but importantly also because During PEER data collection, it was clear of stigma related to revealing or seek- when women made superficial state- ing treatment for reproductive health ments that they had perhaps heard on problems, in case of “blame” or asso- television. Women are aware of “gen- ciation with pre-marital sex. der issues”, but this awareness did not seem to reach beyond agreeing that do- The issue of irregular periods may mestic violence is wrong. They knew of mean that it is harder for women to the existence of an issue called “human know whether they may be pregnant rights”, but could not elaborate further or not, another factor that may delay “...There was a woman with a discharge, seeking advice or services in the case but was too afraid to tell anyone, because of an unplanned she thought the family would blame her..” pregnancy. Further work to explore the availability and use on what this meant. There did seem to of pregnancy testing would provide be interest in programmes about repro- important insights into factors affect- ductive health, relationships and family ing delayed treatment seeking. planning. However, any communications need to take care not to patronise The problem of unmarried women is that their the audience, who are very accustomed period does not come because they sit and work to health messaging. One woman com- for a long time. plained that health educators shouldn’t tell women what to do when they aren’t Unmarried women say they don’t have regular pe- doing it themselves. There is a need to riod, and whenever they have it, they feel pain and move towards evidence based Behav- fever, exhaustion and loss of appetite. iour Change Communications, mov- ing on from outdated assumptions that information or education automatically h o w d o t a r g e t g r o u p s change behaviour. 3.7 r e c e i v e information ? Television and radio were highly val- m a s s m e d i a ued in both urban and rural data. Gar- ment factory workers listen to the radio Mass media (radio, TV and printed in during the day at work, and watch press) are already delivering content TV at weekends. The rural women only concerning issues such as domestic vi- watched TV in evening (as their village

Abortion in Cambodia I 40 Abortion in Cambodia I 41 only had electricity at night), but said helps, as people do not feel personally they might be too busy to watch it or exposed or threatened. In terms of tar- might fall asleep in front of the TV. get audiences at least, communications need not worry about communicating interpersonal c h a n n e l s in an open and honest manner with the target audiences about pregnancy, People trust what their friends tell sexual relationships, family planning, them. Poorer people in particular, lack- relationships with providers etc. The ing access to TV or radio, get their in- moral frameworks around these issues formation from gossip and talking to are complex, but within female peer friends. From the length, depth and networks, these issues are a central fea- detail of the PEER datasets, and the ture of day to day discourse. sheer amount of information women knew about other people’s lives, there Rural communities already have ad- is clearly a strong tradition of story- ministrative and organisational struc- telling and gossiping among friends. tures in place to deliver messages to Communications that capitalise on the their members (see box below for an popularity and relevance of testimoni- example from the study village). How- als and real life stories are thus likely to ever, these are not targeted structures: be well received. for instance, a village leader might call a meeting about reproductive health, but One finding that emerged clearly from any member of the household might PEER is that although there are strong attend to represent the family (e.g. cultural barriers to talking openly about grandfather, teenage girl). Representa- sex and sexuality, when women are talk- tion is also likely to be biased towards ing to their friends, conversation about the richest people in the village, who very personal issues can flow freely. are also likely to have affiliations to po- Employing third person techniques18 litical party.

18 In the form “what do other people say about x” or “have you heard any stories about x”.

Abortion in Cambodia I 42 Abortion in Cambodia I 43

4. implications o f p e e r f i n d i n g s

implications f o r i n c r e a s i n g a w a r e n e s s a n d a c c e s s : 4.1 communications k e y i s s u e s a n d m e s s a g e s

This section examines how the PEER There is a great need to increase aware- findings impact upon future commu- ness of, and access to, family planning nications work for the reproductive and safe abortion services. This is evi- and sexual health sector. Communica- denced by low levels of awareness of tions are made up of the messages that entitlements to services, inaccurate the programme wants to deliver, and risk perceptions, and risky behaviours. the channels through which these mes- Awareness needs to be raised not only sages are spread, in order to promote among the target groups (women of specific behaviour change. reproductive age), but also among people who influence them, such as community leaders and men. The old- est woman in the family (mothers and grandmothers) had a lot of influence

Abortion in Cambodia I 42 Abortion in Cambodia I 43 in family matters and relationships, Women do not demand or expect and may be key individuals to include. high quality abortion services in the Communications and advocacy activi- public sector. They do not know about ties should recognise the importance of their entitlements nor has there been relationships between men and women the opportunity for them to realise rather than conceptualising the two their entitlements. Quality of emotion- genders’ needs separately. al care, cost and confidentiality are at least as important as perceived quality Key issues to tackle suggested by the of technical/medical care. PEER data are as follows: Relationships between providers and Women do not seek safe abortion clients in the public sector need to be promptly. This delay is caused by try- improved, with particular attention to ing a variety of different methods first. counselling skills, attitudes to poorer clients and eliminating financial mal- Women require greater support to at- practice which contributes to low levels tain reproductive and sexual health. of trust in services. Women in these communities find themselves in very difficult circum- Numerous misconceptions and low stances due to violent relationships, levels of knowledge about contracep- economic hardship, stigma around is- tive methods need to be addressed. sues such as STIs and pre-marital sex, These must include what side effects to and other social and economic pres- expect and what to do, together with sures. They may feel compelled to end a improving women’s understanding of pregnancy in secret, or be afraid to seek the basic factors of reproduction and treatment for suspected infections. contraception. This lack of understanding contributes to poor uptake and Communications and other interven- limited effective use of family plan- tions must work to tackle the social po- ning. In particular, longer term and less sition of women which underlies these frequently used methods (IUD, ster- problems, and work towards changing ilisation) are poorly understood and attitudes of men and other people who viewed negatively. Efforts to scale up have an influence on women’s health. access must involve coordinated efforts The Support to Safe Motherhood Pro- on both the supply side and demand gramme in Nepal is widely recognised side (eg. communications, financing) as a leader in the use of demand side to be effective. empowerment approaches to support safe motherhood19.

19 House of Commons International Development Committee Maternal Health Fifth Report of Session. For more information on SSMP see www.safemotherhood.org.np

Abortion in Cambodia I 44 Abortion in Cambodia I 45 c h a n n e l s o f communication of curiosity and many questions to put to health professionals when they have The following communications ac- the chance (as evidenced by experi- tivities and channels present different ence during Q&A sessions in the peer opportunities and challenges: researchers’ workshop). The political and socioeconomic profile of commu- Community based communications: nity networks in communities should Can take different forms (eg. meetings, be considered as they are unlikely to be drama) in rural and urban areas and fully representative of the population, should reflect the target group. Some and the most vulnerable members of should be aimed at the primary tar- the target group may be missed by forget audiences (women of reproductive mal activities. age) whereas others should be aimed at influencing wider community at- Referral options: Women using fam- titudes, especially community leaders. ily planning or safe abortion services Women’s availability to attend such could be provided with a referral card events may be limited by work load ... Women have a high level of curiosity and and season, as rural women continually many questions to put to health professionals stressed how busy when they have the chance ... they were. Suggest- ed locations for meetings include non- to pass on to friends. Because word of clinical community sites (classrooms, mouth recommendation is highly val- house of women’s chief) and in or near ued, this is likely to be an effective way work places (venues near factories or of increasing trust. However, it is likely markets so that they are convenient for to be a relatively ‘slow burner’ for safe working women to attend). Materials abortion in particular, as abortion seek- and messages for use in such meeting is not a very frequent activity. ings should be developed and tested in collaboration with members of the Printed materials: leaflets, posters, target group. Materials should include stickers. These should use simple and a suggested topic guide and points for authentic language and phrases, and discussion, life-like (or real life) case where appropriate include case stud- studies and testimonials, and interac- ies and testimonies from women who tive activities prompting discussion. have used services. Technical language Meetings are ideal for women to have should be avoided. an opportunity to ask questions: with low levels of general knowledge and Mass Media: the majority of house- many misconceptions about reproduc- holds own a TV and/or radio (see be- tive health, women have a high level low) so these channels are potentially

Abortion in Cambodia I 44 Abortion in Cambodia I 45 an effective way of reaching a large implications f o r s u p p l y s i d e number of people. Although outputs 4.2 activities should focus on the key programmatic messages, they should also ad- In order for communications ac- dress other salient themes for women tivities to be effective, there must be a to raise and keep interest. Domestic corresponding increase in the quality, violence is a hugely important issue availability and accessibility of family for women, as are the difficulties and planning and safe abortion services. challenges of making a living, and the The supply side needs to be able to ful- resourcefulness that women employ to fil whatever benefits relating to services address these problems. Other issues that have been promised to the target of concern are alcohol abuse and gam- audience. Therefore communications bling, husbands working far away from relating to specific improved services home, and STIs. must be made locally and only when services are in place.

MASS MEDIA REACH a b o r t i o n s e r v i c e s IN CAMBODIA At present, public facilities are not the 63% of urban and 47% of rural main provider of abortion services, and households own a radio in Cam- for this to change, the supply side will bodia. have to address the following key issues An even greater proportion own a successfully: television: 52% in rural areas, ris- ing to 72% in urban areas. > Confidentiality: ensuring women However, this still means that are able to access facilities outside nearly half of households do not their home area. Ensuring women have a television. are not readily identifiable as seek- 69% of women in Cambodia are ing abortion in public areas of the literate, although this is higher facility. in Phnom Penh (87%) and Kandal (79%). > Improving access: • For vulnerable groups: this may include addressing financial bar- Open days: Once facilities have been riers both for fees and associated refurbished, an effective way of getting expenses such as transport (see box the word out to local women is to in- below for a summary of Health vite groups of women to visit the new Equity Funds). Even very small facilities, show and tell them what is upfront fees can restrict access. available, and give them information to • Providing services outside conven- take away for their friends. tional clinic hours

Abortion in Cambodia I 46 Abortion in Cambodia I 47 > Improving provider attitudes: Summary of current status Ensuring that clients receive emo- of Health Equity Funds tional care and are treated with respect, as well as receiving techni- Health facilities are encouraged to exempt cally expert care poor patients from paying user fees. One way of doing this is through Health Equity > Catering for adolescents: Funds, which reimburse facilities for poor At present there is a lack of services patients’ health care expenses and as- for young and unmarried people. sist poor patients with indirect costs. They A model has recently been devel- currently cover 1% of the population, 35% oped for adolescent friendly of whom are women of reproductive age. reproductive health services in There are currently 35 HEFs in operation, Cambodia [Wilkinson 2003] which with 13 further funds planned for the start provides detailed recommendations of 2008. Problems relating to the use of for those interested in developing HEFs for abortion services include confi- this area. See box below for an dentiality; women’s access to the HEF card; example of how the clinics of an and crucially US funded NGOs controlled NGO in Phnom Penh have res- by the Mexico City Policy refuse to facilitate ponded to their clients’ needs. access to safe abortion services. Currently the vast majority of HEF implementers fall > Providing quick, affordable, into this category. accessible and confidential pregnancy testing: > Considering If women are to seek a qualified role of private sector: practitioner as soon as possible, they At present the various private need to be able to have a pregnancy sector players are a very significant test quickly. It is worth consider- provider for women. Expanding ing how to do this in facilities, access for women necessitates a con without women necessarily having centrated and integrated effort to to queue, register etc. work with the private sector. The inevitable increasing availability of > Contraceptive counselling MA makes this particularly urgent. and services: Women using abortion services > Developing a process for MA: should be provided with contracep- There is an urgent need to explore tive advice and supplies if required service delivery options, and man- from the same provider and in the age the systematic introduction of same location. Post abortion con- MA to ensure the benefits of this traception counselling and services important reproductive technology can help raise the overall contracep- are maximised. tive prevalence rate of an area, as

Abortion in Cambodia I 46 Abortion in Cambodia I 47 they are an opportunity for service to make MA available to Cambodian providers to provide information, women, with work in early stages. advice and services to women. Women would benefit from being > Emergency contraception: able to access recognised and regulated Given the findings of the PEER study, medical abortion services. Issues such particularly the unmet need and as training pharmacists where to refer widespread use of the withdrawal clients for prescription of drugs, ap- method, urgent work is required propriate referral procedures, and se- to increase access to emergency curing quality controlled drug supplies contraception. need to be considered in detail. All par- ties need to consider options to work m e d i c a l a b o r t i o n more closely with pharmacists for referral, as they are often the first point of PEER data show that use of pharma- contact for many women. Additionally, ceutical abortificants bought from public providers should consider mak- pharmacies is widespread, such was ing medical abortion widely available the frequency and consistency with at their facilities. There is already great which they were mentioned. At present demand for medical abortion in Cam- internationally approved drugs used bodia, which will continue to be met to induce abortion are not registered by unsafe provision without post abor- for this purpose in Cambodia. It is not tion family planning counselling, un- clear what drugs are being supplied, less alternative safer choices are given and the frequency and dosage of pills to women. seemed to vary widely, suggesting little consistency in drugs or methods used. l o n g t e r m f a m i l y p l a n n i n g Further research is planned by RMMP to investigate these matters; a number Many of the implications of PEER find- of organisations are looking at options ings for developing long term family

AN EXAMPLE OF YOUTH FRIENDLY SERVICES

RHAC clinics in Phnom Penh have separate entrances for adolescents leading into a library, so nobody knows whether they are accessing services or simply visiting the library. Their first point of contact is the librarian, which helps main- tain confidentiality. Information in the library is provided on issues of concern to service users. A recent survey showed them to be concerned and interested in drug and alcohol misuse, and the management of sexual violence. Girls wanted to learn how to protect themselves from rape, particularly beer promotion and bar girls.

Abortion in Cambodia I 48 Abortion in Cambodia I 49 planning services are similar to those > Considering the role for improving abortion services. Fam- of the private sector: ily planning services need to be client- private providers such as pharma- centred, and respond to the needs of cies are a leading source of women and men who want to access contraceptives, but women do not the services. These needs can be sum- always receive adequate counselling, marised as follows: instructions, and follow-up, contributing to ineffective use of con- > Confidentiality: traception and anxiety about side is required in all services, especially effects. for young and unmarried people. This means addressing the physical > Improving provider attitudes: layout and organisation of facilities helping providers to understand as well as increasing levels of trust. and respond to the needs of cus- tomers is vital, emotional care and > Improving access: respect are equally important to as with abortion services above – technical competency for family the financial costs and inconvenient planning providers. opening hours of services limit access.

Abortion in Cambodia I 48 Abortion in Cambodia I 49 > Catering for adolescents: implications f o r tackling the lack of services for 4.3 interpreting quantitative young people, including in rural s u r v e y d a t a areas, has great potential for reducing the number of unwanted PEER data allow exploration of how pregnancies and maternal deaths. attitudes may affect reporting in surveys on abortion in Cambodia. Eight percent > Improving perceptions of women are currently estimated to of long term family planning have ever had an abortion. This would methods: seem an underestimate, considering the at present surgical methods of high frequency with which abortion contraception and the IUD are was discussed by the peer researchers, viewed very negatively. Clients at the nature and content of their stories, family planning services should be and, the fact that many women are like- offered a wide range of methods, ly to be disinclined to admit to abortion. The high level of detailed knowledge ...Women talk freely [...] about issues around about abortion pro- abortion, [...] but want to keep it a secret viders and methods, when they seek an abortion for themselves... the numerous reasons and situations mentioned described for and providers should be equipped why women had abortions, and the with the skills to explain and reas- matter of fact way in which women sure women and men about all talked about abortion, suggest that it methods, including the less popular is a much more common occurrence. ones. PEER data do not allow a quantitative estimate of prevalence of abortion, but > Reducing financial burden of it is worth bearing in mind that survey long term family planning: data may only represent the tip of the one common belief about lon- iceberg. ger term methods was that they were much more expensive than Why might women under-report the injection/contraceptive pill. abortion? Although in the third per- If longer term methods are to be son (i.e. what do other women say promoted, the price differentials about....), women talk freely and com- and financing options will need to fortably about issues around abortion, be addressed. PEER data show that many women want to keep it a secret when they seek an abortion for themselves. However, older women who had abortions because they

Abortion in Cambodia I 50 Abortion in Cambodia I 51 had had all the children they wanted, or both rural and urban areas, reported because their health was at risk, were use of public facilities for abortion was not generally ashamed about having an very rare. Older women with a socially abortion, as these are more socially ac- acceptable reason for having an abor- ceptable reasons. This may help to ex- tion might be more likely to go to pub- plain why in the DHS, the proportion lic service providers as they are not as of older women reporting having had concerned about feeling ashamed or one or more abortions is much higher judged. than among young women (of course, one would also expect this figure to be PEER data also highlight a pattern of higher among older women due to the attempting abortion by trying to self- fact that they have had longer exposure induce (through herbs, alcohol, run- to risk of abortion too). ning and jumping), only seeking attention at a hospital or clinic if these Another finding from PEER is that attempts fail. For women reporting younger women are far more likely to abortion in a survey, when asked what go to private or informal providers for method they used, they might answer abortions than to public facilities. The with the final method that resulted in most recent data on abortion in Cam- the successful abortion. Surveys may bodia come from a survey of public thus underestimate greatly the preva- facilities, and thus again represent an lence of attempts at unsafe abortion us- underestimate of the total number of ing traditional, self-induced, or medi- abortions occurring in the country. In cal abortion methods.

Abortion in Cambodia I 50 Abortion in Cambodia I 51

4. implications f o r a d v o c a c y

Advocacy: is a process of working to gain the support of decision makers and people in power. These include services providers, policy makers, and community members. In the case of RMMP, advocacy will aim to build a sup- portive environment for reducing maternal mortality through the activities of the programme.

a d v o c a c y professional and social position is very 5.1 a t p r o v i d e r l e v e l different when they meet in a clinical setting. The provider may think that Any drive to improve safe abortion they have insight into the problems and and family planning services has to needs of their clients, when in fact they ensure that providers (doctors, nurses, may know little about the everyday re- midwives and facility staff such as reality of their lives. ceptionists) recognise women’s right to a safe abortion in a government facility, Examples of key messages for advoca- and all women’s rights to family plan- cy work with providers might be: ning services. PEER findings will help develop tools for this purpose. Even • Women are in danger from unsafe if providers come from a similar cul- providers: women use unregulated, tural background to their clients, their ineffective and unsafe abortion

Abortion in Cambodia I 52 Abortion in Cambodia I 53 methods and providers. If you do not • Accountability: to improve services, provide them with a service, some- clients need to be able to feedback one else will. their experiences of providers, and • Women’s right to choose: women providers held responsible for do not need to justify WHY they want responding to this feedback. Explain an abortion: if they are required to importance of whatever feedback do so, they will either lie about the mechanisms are introduced, such as reason or go elsewhere. Respect feedback cards. their right to choose. • Importance of family planning: women with unwanted pregnancies a d v o c a c y 5.2 have a need for family planning ser- a t h i g h e r l e v e l vices, which should be offered along with explanations of how the meth- The pragmatic and non-moralistic tone ods work, and reassurances about in which peer researchers talked about their safety. abortion is quite different from how • Importance of equity of access: Young people may be sexually active before poor women face the greatest dan- marriage, but they should not be treated any gers with un- differentlyfrom married people when safe abortion: they come to access services. help them to avoid this danger by welcoming abortion is dealt with at a policy level. them to services and treating them Abortion is a very sensitive political is- with the same respect as any other sue in the national and international client. arena; it is interesting that Cambodian • Making services accessible to young women are not similarly perturbed by people: young people may be sexu- dealing with and discussing the practi- ally active before marriage, but they cal and moral issues around abortion. It should not be treated any differently is very important in terms of advocacy from married people when they activities to present a realistic picture of come to access services. abortion in Cambodia. De-stigmatising • Quality of care: women want quick, abortion and other aspects of sex and polite, affordable, reassuring and con- reproduction is an essential step in im- fidential services. They do not want proving reproductive health and reduc- to be judged or told off by providers. ing maternal mortality and morbidity. If they cannot get these qualities from public facilities, they will go else- Advocacy at government and policy where, where they may receive sub- level and above will be important in standard and unsafe services. promoting commitment to access to

Abortion in Cambodia I 52 Abortion in Cambodia I 53 safe abortion, and of the right to have the safety of abortion in Cambo- an abortion in a government facility, dia. As well as developing structures with equitable access for all women. such as drug registration, a support- At this level, the following issues will ive environment in terms of at- be important to tackle: titudes among decision makers and implements will have to be fostered, • Financial barriers to access: services using the strong arguments for pro- must be affordable. Even if quality viding regulated medical abortion of services improves, and knowledge presented in this report. about availability of services increases, financial costs will discourage • Provider incentives: informal pay- and prevent women from using ments, long waiting times and poor public services. This will be a serious provider attitudes are all barriers to barrier to reducing maternal mor- high quality service provision. Staff tality, as the poorest women are at must be adequately rewarded for the highest risk, yet will not be us- good quality services, training and ing services intended to help them. “sensitisation” will not be effective. Urgent work is required to explore pro-poor health financing solutions • Commitment to equity of access: which include safe abortion in a particularly for young and unmar- basic package of services without ried people. stigmatising those who use them. All stake-holders, and particularly the donor community have a w o m e n ’s r i g h t t o responsibility to minimise the im- 5.3 a c c e s s reproductive pact of Mexico City on Equity funds h e a l t h s e r v i c e s and other pro-poor mechanisms. Advocacy activates should fall under a • More structured provision of medi broad “rights based” framework20. This cal abortion: as mentioned in holds that reducing maternal deaths cannot be accomplished by technical health in- Health services are often male dominated terventions alone, but and struggle to hear the voices of the women requires work to chal- they serve. lenge the political and social status quo, in- section 4.2.2, developing safer cluding actions within and beyond the medical abortion provision presents health sector. a huge opportunity for improving As PEER data have shown, women’s

20 For more information see the DFID ‘How to’ note at http://www.dfid.gov.uk/pubs/files/maternal-how-to-final.pdf 13/2/08

Abortion in Cambodia I 54 Abortion in Cambodia I 55 health is affected by a number of cess to other interventions working on factors: gender, empowerment, livelihoods etc. • Low levels of knowledge and lack of In short, the issue should be addressed education from a gendered social development • Gender relations: lack of power and perspective rather than as a purely tech- decision making in relationships nical health systems issue. • Lack of power in relationships with Health services are often male domi- health care providers nated and struggle to hear the voices • Lack of money to pay for services of the women they serve. In additional to advocacy, systems of accountability Opportunities should be found to link must be developed which put women work around safe abortion and LTFP ac- at the centre of service delivery.21

21 For more information see DFID Briefing note: Voice and Accountability Matters for Better Education and Health Services 2007 13/2/08

Abortion in Cambodia I 54 Abortion in Cambodia I 55

bibliography

De Walque, D. 2004. “The Long-Term Legacy of the Khmer Rouge Period in Cambodia.” World Bank

National Institute of Public Health, National Institute of Statistics [Cambodia] and ORC Macro. 2001/2006. Cambodia. Demographic and Health Survey 2000/2005. Phnom Penh, Cambodia and Calverton, Maryland, USA

Policy Research Working Paper 3446. Washington, DC: World Bank. DFID (2005) Country Assistance Plan www.dfid.gov.uk/pubs/files/capcambodia.pdf

IPAS (2007) “Ready or Not?” A National Needs Assessment of Abortion Services in Cambodia.

Ministry of Health (undated) Reproductive health in Cambodia: A summary of research findings 1990-1998.

Ministry of Health (2005). Family Planning Survey. Domrei Research and Consulting.

Reproductive and Child Health Alliance (RACHA) (2000) Birth Spacing Dropout Study, Siem Reap province.

Wilkinson, David (2003) A Model for Adolescent-Friendly Reproductive Health Services in Cambodia. MoH/WHO.

Abortion in Cambodia I 56

a p p e n d i x 1: p e e r m e t h o d in d e t a i l

Selection of Peer Researchers

CHEMS was asked to select 15 peer researchers from both of the study sites according to the following criteria: • They should be a woman aged 15-35 years • They should represent typical local women as far as possible • They should not have been involved in programmatic activities before (e.g. peer education schemes, village health volunteer etc) The urban location (within Phnom Penh) and rural location (a village in Kandal Province approximately 50km from the centre of Phnom Penh) were selected because RMMP’s ini- tial activities are taking place in these areas, and insight into differences between urban and rural communities were required.

Monitoring work involving qualitative operations research will be carried out as RMMP’s activities stretch into other provinces. Findings from Kandal and Phnom Penh should not be viewed as representative of the whole country, as there are likely to be many socioeconomic differences between Phnom Penh and its immediate surrounds and more remote rural areas. Ongoing dialogue with local communities about their needs will therefore be necessary. However, many structural issues such as financial barriers to access, client/ provider relations and the factors that shape decision making around abortion are likely to be important across the country.

Peer Researcher Training Workshop

The peer researcher training workshops were held separately for the urban and rural groups. Informed consent for participation in the study was obtained from all peer researchers, and they were paid a per diem and any expenses incurred. During the four-day workshops, peer researchers: • Discussed and identified important issues in their community (with an emphasis on women’s health) • Developed their prompts on three different topics, to guide their in-depth interviews, shaped by what they felt to be the most important issues (see below) • Practiced asking open-ended questioning, probing, and asking for stories • Practiced and were observed asking for consent from their friend to take part • Learned about ‘third-person interviewing’ (using no names, and asking about ‘what other people say’ rather than personal questions)

Abortion in Cambodia I 56 • Field tested their interview guide with a friend, fed back to the group and made any necessary adjustments

Peer researchers were provided with pens and notebooks, and were advised that they could write down key words or phrases to help them remember what their friends told them if they wanted. Some peer researchers found this to be a useful practice, whereas others did not have the requisite literacy levels. Experience of PEER in other countries has shown non-literate groups to be highly skilled at remembering the stories they hear when interviewing their friends, and the depth and detail of peer researchers’ responses confirmed that this was also the case in Cambodia.

Data Collection

After the training workshop, peer researchers started to carry out in-depth, conversa- tional interviews with two of their friends (rural group) or three of their friends (urban group) on three different topics, using the prompts developed at the workshop to guide the conversation. In the urban group, PEER data were collected from peer researchers by supervisors from CHEMS at regular intervals over a period of around five weeks. At the end of this period, the PEER advisor met up with each peer researcher individually to de-brief them, referring to the notes provided by CHEMS.

The rural PEER study used a more rapid data collection timetable, whereby peer researchers interviewed one friend on one topic every day, and fed back every other day to the PEER advisor and CHEMS supervisor who took detailed notes in English, using an interpreter. All peer researchers successfully remembered detailed information from their friends, some requiring very little prompting from the research team.

Data analysis

Data were analysed in two stages: By peer researchers: at the end of data collection, peer researchers were brought back together for a workshop to provide feedback on their experiences, discuss their findings, and answer emerging questions from the PEER advisors. By PEER advisors: Narrative data were entered into Microsoft Word in English, translated from the original data which were in Khmer. Data were read and re-read, and key themes were identified. Data were thematically analysed according to the pre-existing analytical framework (developed according to the objectives of the research). Emerging themes, cat- egories and insights were incorporated into this framework. PEER researchers feedback on their experience:

What they found difficult:

• First time we thought we couldn’t do it • If we were not clear about the topic, they couldn’t answer in detail • Some of the single peer researchers felt shy to ask and answer some questions (about marriage, reproductive health) • Some of our friends felt jealous as we got a lunch allowance • A friend rejected the interview saying researchers are too young to talk about sex • Some of the interviewees didn’t want to talk for a long time • Some of our unmarried friends didn’t want to talk about sex • Some people went off the topic • Some people complained about snack provided, that it was too little

What they liked:

• To know about the relationships with men • To know about health problems and women’s problems • Our friends were friendly • To get more experiences, share ideas • That there were incentives but no force • We are excited, never thought they’d become a researcher • We learned to express our opinion, and not be shy • To know more about the women • We feel proud to have been a researcher • Before, felt shy to talk about sex, now we dare to face talking about sex • Dare to talk about everything because we know a lot • Surprised because I can be a researcher • Happy to become a researcher, and people are friendly • We became famous (in the village - met lots of people)