From Research to Reality: The Challenges of Introducing Medical Abortion into Service Delivery in Vietnam

Abstract

Following the inclusion of mifepristone + misoprostol for early pregnancy termination into the Vietnam National Reproductive Health Guidelines in 2002, a team from the Ministry of Health, World Health Organization and Ipas assessed how best to move from clinical research to widespread public sector availability. After field visits to hospitals and discussions with stakeholders, the team endorsed the phased introduction of medical abortion alongside vacuum aspiration services to expand choice. They stressed the importance of patient-oriented information on what to expect as the abortion process takes place and the need for follow-up, also crucial in training of providers. Routine use of ultrasound to detect ectopic pregnancies or determine that abortion was complete was considered unnecessary. The mandated four-hour observation period following misoprostol administration could be reduced. The National Guidelines, appropriately conservative at the start of the programme, should be amenable to modification as experience grows. Introduction is not possible without a steady supply of drugs at affordable prices. Ways to reduce the high price of mifepristone and deal with provider expectations of extra allowances, as with surgical abortion, are needed. Making medical abortion a viable option for most Vietnamese women will require provision at commune-level clinics through mid-level providers, and with home use an option. Other challenges include use of misoprostol alone and regulating provision in the private sector.

Résumé

Après l’inclusion du mifepristone + misoprostol pour l’interruption précoce de grossesse dans les Directives nationales de santé génésique au Viet Nam en 2002, une équipe du Ministère de la santé, de l’OMS et de l’Ipas a étudié comment passer de la recherche cliniqueàune large disponibilité dans le secteur public. Après des visites aux hÁpitaux et des discussions avec les intéressés, l’équipe a approuvé l’introduction progressive de l’avortement médicamenteux parallèlementàl’avortement par aspiration. L’information est importante pour que les patientes sachentàquoi s’attendre pendant l’avortement et soient conscientes de la nécessité d’un suivi, mais aussi pour la formation des prestataires. L’utilisation systématique de l’échographie pour détecter les grossesses extra-utérines ou s’assurer que l’avortement est complet a été jugée superflue. La période obligatoire d’observation de quatre heures après l’administration de misoprostol pourrait Átre réduite. Les Directives nationales,àjuste titre prudentes au début du programme, devraient s’ouvrir aux changementsàmesure que l’expérience s’accroît. L’introduction dépend d’un approvisionnement régulier de médicamentsàun prix abordable. Il faut parveniràréduire le prix élevé du mifepristone et satisfaire les prestataires qui s’attendentàdes subventions supplémentaires, comme avec l’avortement chirurgical. Pour que l’avortement médicamenteux devienne une option pour la plupart des Vietnamiennes, il devra Átre pratiqué dans les dispensaires communaux par du personnel de niveau intermédiaire, avec l’option de l’utilisationàdomicile. D’autres points concernent l’utilisation du misoprostol seul et la réglementation des prestations dans le secteur privé.

Resumen

Tras la inclusión de mifepristona + misoprostol para la interrupción del embarazo en etapas iniciales en las Directrices Nacionales de Salud Reproductiva de Vietnam en 2002, un equipo del Ministerio de Salud, la Organización Mundial de la Salud e Ipas evaluaron la mejor forma de hacer la transición de la investigación clánica a una amplia disponibilidad en el sector público. Después de visitar los hospitales y hablar con las partes interesadas, el equipo aprobó el lanzamiento por fases del aborto con medicamentos junto con los servicios de aspiración endouterina para ampliar las opciones. Recalcaron la importancia de proporcionar información orientada hacia las pacientes sobre qué esperar al efectuarse el proceso de aborto y la necesidad de dar seguimiento, fundamental en la capacitación de los proveedores. El uso rutinario de la ecografáa para detectar los embarazos ectópicos o para determinar la finalización del aborto se consideró innecesario. El peráodo obligatorio de cuatro horas de observación después de la administración del misoprostol puede reducirse. Las Directrices Nacionales, debidamente conservadoras al inicio del programa, deben prestarse a modificaciones según se adquiera más experiencia. El lanzamiento no es posible sin un suministro constante de medicamentos a precios asequibles. Se necesitan formas de disminuir el alto precio de la mifepristona y lidiar con las expectativas de los proveedores respecto a descuentos adicionales, asá como con el aborto quirúrgico. A fin de lograr que el aborto con medicamentos sea una opción factible para la mayoráa de las mujeres vietnamitas, los profesionales de la salud de nivel intermedio deben prestar estos servicios en las clánicas comunitarias, y debe permitirse el uso en el hogar. Otros retos son: el uso del misoprostol solo y la regulación de su suministro en el sector privado.

Keywords:

• medical abortion
• abortion providers and services
• abortion law and policy
• informed choice
• drugs and drug companies
• Vietnam

Abortion in Vietnam is legal and available on request.¹ Data on the number of abortions in the country are limited and vary widely by source of information. Abortion rates in the 1990s were estimated to be in the range of 83 per 1000 women aged 15—44, which translates to an average of 2.5 induced abortions in a woman’s reproductive life.², ³ Although the introduction of pregnancy testing has helped to reduce unnecessary procedures on non-pregnant women, approximately 700,000—800,000 abortions are still reported each year (Prof. Tran Thi Phuong Mai, Ministry of Health, personal communication, January 2003). This does not include the largely unreported abortions taking place in an ever-growing private sector. While reported contraceptive prevalence is high, a large proportion of users rely on withdrawal or periodic abstinence.⁴ The majority of women who have abortions are married, but the proportion of unmarried adolescent abortion seekers has been increasing in recent years. 60—70% of induced abortions occur within the first six weeks of a missed period.², ⁵

Abortion services for first and second trimester are available at central (Ho Chi Minh City and Hanoi) and provincial levels, and for first trimester at district level hospitals. Commune health centres provide services up to six weeks. However, a recent situation analysis found that while all provincial level facilities and 95% of district hospitals provided abortion services, only 66.8% of commune level health centres were functional.⁶ Manual vacuum aspiration (MVA) is commonly used till eight weeks of pregnancy but D&C is still prevalent above eight weeks.⁷ Apart from doctors and assistant doctors, secondary midwives are authorised to do MVA up to six weeks. Though wide variations exist, the average cost of an early abortion was 30,000—150,000 VND (US$2—10) in 2003.

Following several in-country clinical trials in the 1990s that documented the safety, efficacy and acceptability of medical abortion [mifepristone + misoprostol] for early pregnancy termination,⁸, ⁹ the Ministry of Health (MOH) was interested in adopting the method on a wider scale. However, a 1997 strategic assessment of the abortion situation in Vietnam recommended further exploration of the conditions needed to introduce medical abortion into the existing health care system.¹⁰

The National Reproductive Health Standards and Guidelines, which came into effect in September 2002, included mifepristone + misoprostol as an appropriate method for termination of pregnancy up to 49 days from the last menstrual period. The recommended regime was 200 mg of mifepristone followed 48 hours later by 400 mcg of oral misoprostol, administered at a facility with a four-hour observation period after misoprostol. A follow-up visit two weeks later is booked to check completeness of the procedure. Unlike vacuum aspiration, medical abortion use was permitted only for obstetric—gynaecology specialists at central or provincial level hospitals and for women living within 30-minute access to the hospital. Mid-level providers were not authorised to provide medical abortion.¹¹

Assessment of conditions for introducing medical abortion

In order to identify what would be needed to implement the recommendations of the National Reproductive Health Guidelines, the Ministry of Health (MOH), the World Health Organization and Ipas conducted an assessment in Vietnam. A multi-disciplinary team of five individuals, including a representative of the MOH, formed the core assessment team. A translator accompanied the team in order to assist the two non-Vietnamese speaking team members. Fieldwork was done over a three-week period in January 2003.

All central hospitals involved in medical abortion research trials or with past experience with the method were included in the site visits. These included two obstetric—gynaecology hospitals and one Maternal and Child Health/Family Planning (MCH/FP) Centre in Ho Chi Minh City and two hospitals in Hanoi. A visit to the provincial hospital and the MCH/FP centre in one province in the south of the country was also included. Visits included interviews with doctors and nurses and observations of hospital infrastructure. Providers at a hospital in Ho Chi Minh City, the only site where medical abortion was part of service provision at the time of the assessment, asked patients for permission to be interviewed by the assessment team. Five women who agreed and were available at the time of the visit were interviewed in-depth.

In addition, a group discussion with providers and researchers was held in Hanoi including representation from provincial hospitals not visited. The team also had extensive discussions with representatives of the MOH Reproductive Health and Drug Departments, and a group discussion with representatives of NGOs and multilateral agencies. An interview with the pharmaceutical company that marketed mifepristone was not possible. Instead, the mystery client approach was used with the company and 13 pharmacies in Hanoi and Ho Chi Minh City in order to gauge its availability.

Availability of mifepristone

Although mifepristone had already been approved, it had not yet been registered. The MOH permitted one domestic pharmaceutical company to import the drug from China and to market it in bulk quantities only to hospitals carrying out research studies for VND 270,000—300,000 (US$18—20) per 200 mg tablet. Misoprostol was available as a gastric ulcer medication for VND 2250—6750 (US$0.15—0.45) per 200 mcg tablet. Its licence had expired in 2001 and supplies were due to run out by the end of 2004. Its use as an abortifacient was off-label.

The two mystery clients were unable to obtain mifepristone over the counter at the 13 pharmacies visited. Phone calls to the company in Vietnam were also unsuccessful in obtaining the drug. However, a number of providers and NGO representatives reported that many restricted drugs in Vietnam can be obtained over the counter and the potential for mifepristone being available in this way could not be ruled out. One provider also mentioned that while obtaining the drug may be difficult for patients, providers could obtain it easily even if they were not linked to the authorised hospitals.

Adaptability of services to providing medical abortion

Abortion caseloads at all the central hospitals in Ho Chi Minh City and Hanoi that the team visited were high. Tu Du Hospital, one of the largest ob-gyn hospitals in the country, performs about 30,000 first trimester abortions each year and the Hung Vuong Ob-Gyn Hospital performs 18,000—21,000 per year. The provincial hospital and the MCH/FP centre we visited averaged 20 abortions per day. All the hospitals reported that two-thirds or more of their patients come for termination before eight weeks. In Ho Chi Minh City, providers estimated that 30—40% of women were unmarried, while providers in Hanoi reported a somewhat lower proportion who were young and unmarried.

Staff shortages were not uncommon, especially at the provincial general hospitals. Only the FP unit in the ob-gyn departments performed abortions, and doctors were not always keen to be assigned to that unit because of competing interests to work in more surgically-oriented areas.

The organisation of abortion services varied from hospital to hospital. While some facilities had large waiting rooms, in others women waited in a corridor or other open space, sometimes alongside women awaiting antenatal care or other ob-gyn services. While the larger hospitals did have designated counselling areas, ensuring privacy was not always possible. In the province we visited, the MCH/FP centre was better designed to ensure privacy and space for counselling than the general hospital, and we were told the situation was similar in most provinces. The recovery room in most of the hospitals we visited was a small room with only 3—4 beds. These conditions have a significant effect on confidentiality, as women risk being recognised by friends or neighbours.

Counselling and information provision

High caseloads and inadequate staff make it difficult for many hospitals to spend much time on abortion-related counselling. Additionally, many providers felt that as surgical abortion had been available in Vietnam for many years and was familiar to both providers and women, providing information about medical abortion and what to expect from it was less of a need.

The team did not find any patient-friendly informational material about medical abortion except for a brochure brought out (illegally) by the pharmaceutical company that marketed the drug. We did not see any material with information on both abortion methods to help women make an informed choice.

Several of the sites had detailed consent forms for women who participated in the medical abortion trials. These forms mostly contained comprehensive information on the side effects of the method but at most places the consent form remained with the provider. On the other hand, sites without prior research experience often had no consent forms for induced abortion, only an admission form covering miscarriage and abortion.

Follow-up and unscheduled visits

Some hospitals advised a routine follow-up visit after MVA or D&C, but in most, women were told to return only in the event of complications. A medical abortion procedure as per the National Standards and Guidelines requires three hospital visits, one and in some centres two visits more than currently required. While lost-to-follow-ups were low in research studies, they are likely to be higher in the service delivery setting. Most hospitals admitted that drop-out rates for follow-up visits were high and tracing women who did not return was difficult as many provide false addresses. It is also expected that some women seek further treatment at other facilities.

As the abortion process mostly takes place outside the hospital with medical abortion, some women may want advice or reassurance prior to the two-week follow-up visit. During the research trials, most centres set up emergency hotlines or other access mechanisms. In normal service delivery, however, women would be seen by on-call emergency staff at the hospital. A woman who reports with problems outside of routine clinic hours may see a doctor who is not well versed in medical abortion and may, for example, advise an unnecessary surgical intervention.

Links with contraceptive services

At several service delivery sites, especially at the provincial level, sterilisation and IUDs were the only available contraceptive options. Women wanting condoms or pills needed to buy them at pharmacies or get them through the Population and FP Collaborators within their communities. Condoms are also provided free by Provincial AIDS Prevention Committees, but women expressed reluctance to go there for fear of being thought to have STIs or HIV. Providers mentioned that some active MCH/FP centres got condoms from the Provincial AIDS Prevention Committees and provided them free to patients, while others purchased condoms and pills and re-sold them to women, but this was the exception rather than the norm.

Private sector services

All the interviewed providers were reluctant to talk about medical abortion provision in the private sector but several admitted that it was happening. They also mentioned that given the high cost of mifepristone and its limited availability, abortion with misoprostol alone may be more prevalent in the private sector. Many public sector providers have private practices outside their working hours, and women get the drugs in public sector clinics and come for follow-up to the private sector, whether for convenience or at the physician’s request.

At one Hospital in Ho Chi Minh City we were told the hospital itself ran both a “general” service, subsidised by the government, and a “special” service, with fees similar to private care. The cost differentials between the two can be considerable. For example, a surgical abortion at the general service cost approximately 30,000 VND (US$2), while in the special wing it cost 150,000 VND (US$10). Waiting times were shorter at the special service because the case load was lower and services were available outside routine working hours. For women having a surgical abortion, the special service included the option of general anaesthesia which, although technically unnecessary, was perceived by many women to mean higher quality care.

Protocols vs. practice

Although the National Guidelines recommend a four-hour observation period following misoprostol, doctors at most research sites admitted that women went home within half an hour, both because women often requested this as soon as they took the tablet and because with high caseloads, waiting space was insufficient for women to stay that long.

Some providers expressed a desire to use a misoprostol-alone regime as a way of overcoming the high costs of medical abortion. Several providers were of the opinion that the upper limit with medical abortion should be 8—9 weeks instead of the present 7 weeks of pregnancy LMP.

The biggest difference between policy and practice was with respect to routine use of ultrasound to rule out ectopic pregnancy prior to giving women the drugs and during follow-up to check the abortion was complete. Many providers believed in routine use of ultrasound, partly because they had used it during research trials. Some did not. One doctor related her past experience wherein, at one centre, overuse of ultrasound for follow-up led to excessive surgical intervention. The intervention rate dropped and the success rate with medical abortion increased after routine use of ultrasound at follow-up was discontinued.

Advantages and disadvantages for providers

Doctors with prior experience in clinical trials generally felt that medical abortion was easier for women, less risky and would save hospital expenditure on surgical equipment and operating theatre costs. Some said the research studies had already generated a demand for the method among their patients that it would be unethical not to meet.

Some providers in the southern part of the country, where religious influence is stronger, mentioned that medical abortion shifted responsibility for the abortion from themselves to the woman, which was in keeping with their own ambivalence about doing abortions. As the head of one unit told us:

“Our staff will be more willing to provide medical rather than surgical abortion because the woman takes the medicine and the provider does not have to touch the [abortion] instrument.”

Knowledge about the method or Vietnamese experience with it in clinical trials had not spread beyond those actively involved in the research. National Guidelines had also not percolated down to all providers, especially those outside Ho Chi Minh City and Hanoi. At the time of the assessment, apart from training related to clinical trials, there had been only one large-scale training course for doctors (September 2002, by the MOH and Population Council), and there had been no training for midwives. As the training had not been followed by drug availability, attrition of knowledge appeared to be high among those who had been trained:

“I’ve forgotten how medical abortion is administered; it was quite a long time ago (last year).”

Almost all providers raised the issue of extra allowance payments. These are additional payments that providers receive as compensation for surgical procedures. While the actual amounts vary from province to province, given the high abortion caseloads and relatively low basic salary, extra payments constitute an important source of income — as high as a quarter of total monthly earnings for some paramedical staff. Many were concerned that an extra allowance would not be applicable to medical abortion and without it, it might be difficult to sustain enthusiasm for it. There was also concern that counselling (generally done by nurses), could be neglected if the extra allowance was given only to the doctor.

Women’s experiences

The five women interviewed all felt the method was more natural and less likely to interrupt their daily lives. Two chose the method because of their fear of surgery and associated pain.

“I am afraid of pain but can stand the pain [caused by the tablets]. I am so scared of being on the table. I am afraid even of the sound of an instrument." (30-year-old woman, two children)

The young, unmarried girl liked the idea of an abortion that did not involve any instruments. One woman thought it less risky for future fertility. But the women were not always happy about the time and number of visits involved:

“The first time I went to the hospital, I spent about 3 hours. I was told to have ultrasound but I could not make it because working hours ended. I had to return the next day and spent about 2—3 hours for the ultrasound and buying the drug. I took the tablets at home. Two weeks later I returned for a follow-up visit. It took about an hour and a half as I had to wait for ultrasound again."

All the women experienced side effects such as diarrhoea, nausea, fatigue or abdominal pain. One of them felt worried that she was still pregnant because she had nausea for two weeks after she took the tablets, but she was not. One woman expressed fear of long-term effects of the drug in the bloodstream (which is not a problem). How women responded to the side effects depended on their expectations and the counselling they had received. One woman who had been well informed said:

“I was so tired during those days, but I didn’t call the doctor because my symptoms were exactly what the doctor told me.”

But some women had difficulty recalling having received information about symptoms or what they could expect during the course of the abortion.

“It was explained only that … the menstrual period would be longer than usual. They did not tell me about other symptoms. I took the first tablet and felt dizzy and had a headache for about an hour. In the afternoon, I took the second tablet and felt abdominal pain. I was frightened. I told my boyfriend and asked him to go to the hospital to ask the doctor. ” (20-year-old unmarried girl)

Two of the women said they did not receive sufficient information about vacuum aspiration, and that the choice of method was made by the provider. Interviews with providers also confirmed that many of them had strong opinions as to which women were suitable for which method and recommended these. One provider admitted she steered unmarried women to medical abortion because she felt it was a more confidential method. But the unmarried girl we interviewed did not return to the hospital herself when she was troubled by side effects for fear of being identified so the need for confidentiality is not always what this doctor intended.

Costs to women were often considerably higher than the costs of the drugs. One woman we interviewed reported that in addition to the VND 300,000 (US$20) she paid for the drug, she paid VND 90,000 (US$6) for three ultrasounds. Several also spoke of the costs of each required visit. Nonetheless, all five women said that if they were to have another abortion they would opt for the medical method.

“I would choose medical abortion again even if I have to pay more money because it is less painful."

Key issues arising from the assessment

The assessment team presented their report to the government in July 2003. It not only examined the challenges in implementing the National Guidelines, but also made some comments on the appropriateness of the Guidelines themselves. The team appreciated the need for the Guidelines to be conservative and allow for a wide margin of safety to begin with, but emphasised the need to be able to adapt to new and emerging evidence and greater experience.

Areas where the team felt existing protocols could be modified included reducing the waiting time at the facility following misoprostol administration to half an hour, in keeping with ongoing practice. There was no evidence that a shortened waiting time increased complications or in-transit abortions and may in fact have decreased the load on overburdened facilities while meeting women’s needs for confidentiality and convenience.

Similarly there is research evidence (including from studies in Vietnam) that home administration is preferred by many women for reasons of convenience and cost,⁹, ¹² and this should be an option that facilities should consider offering. However, the team also felt that as loss to follow-up would be more common, counselling should prepare women who did not come back to recognise warning signs of incomplete abortion or other problems.

The team also stressed the need to build a local evidence base and suggested consideration be given to increasing the time limit for which mifepristone + misoprostol are permitted from the current seven weeks to eight or even nine weeks LMP, as recommended by the WHO.¹⁰

On the other hand, there is interest in a misoprostol-only regimen for early abortion since it is cheaper. However, the team recommended that the regimen outlined in the National Standards and Guidelines not be overwritten, as at the time there was no agreement on the optimum dose of misoprostol alone or the best interval between doses, and there were concerns about the relatively higher failure rate, risk of fetal abnormalities if pregnancy continued and the incidence of side effects.¹³, ¹⁴

The National Guidelines do not mandate the routine use of ultrasound and the team felt that in spite of provider opinion to the contrary, the routine use of ultrasound would only impede access and increase costs while providing no additional benefits over clinical examination for diagnosing ectopic pregnancy and incomplete abortion. The team also did not find evidence of a high rate of ectopic pregnancies in Vietnam but given the concerns expressed by numerous providers, data on the incidence of ectopic pregnancy among Vietnamese women, especially among those seeking early abortion, would be useful to obtain, to lay the debate to rest.

The team stressed the need for phased introduction of medical abortion, with the central hospitals becoming functional before moving on to the provincial. At the provincial level it was felt that MCH/FP centres could adapt to medical abortion more easily than general hospitals because of their existing emphasis on counselling.

Secondary midwives are allowed to perform vacuum aspiration and should be allowed to provide medical abortion pills as well. At a minimum, even within the framework of the existing guidelines, it needs to be recognised that as nurses provide counselling, they are critical partners in medical abortion service delivery. Thus, all training programmes should involve doctors and paramedical staff as a team. But doctors too need to be familiar with the basic principles of counselling, the abortion process with medical abortion and appropriate follow-up care. The team also recommended the need to increase training efforts, utilising the pool of in-country experience built up in Vietnam during the last decade of clinical research.

Information sharing to enable the woman to choose a method is a key element that has not been adequately addressed in abortion service delivery more generally. Verbal counselling must be supplemented with patient-friendly fact sheets that can be taken home. Given the high levels of female literacy, reliance can be placed on the print media in Vietnam in a way that many other developing countries cannot. Nevertheless, the use of internationally available materials translated into Vietnamese is not adequate and materials suitable for Vietnam need to be developed.

Many providers expressed concern whether there was a need for extra staff and infrastructure to cope with increased counselling needs. However, in many hospitals reorganisation of existing space to allow for privacy, use of group counselling techniques (being experimented with at one hospital), reducing waiting times, doing away with ultrasound requirements and better preparedness for side effects would all be effective measures to allow services to be integrated without increased investments.

Contraceptive counselling must be an obligatory part of all abortion service provision and a choice of methods should be available where women access abortion services, both before and after abortion.¹⁵

Policy for reimbursing providers should deal with their expectations of extra allowances, as with surgical abortion. Efforts must be made to ensure the same quality of care at both general and special services in hospitals.

Introduction is not possible without a steady supply of drugs at affordable prices, and this was emphasised as the single, most immediate action the government needed to take. Various mechanisms including price negotiations with existing pharmaceutical companies to lower public sector prices of mifepristone and/or importing it from other countries like India or the active ingredient from China, were suggested.

The aftermath: follow-up activities

The assessment was presented to the Ministry of Health, who not only accepted the findings but felt they needed wide dissemination and discussion among stakeholders. They held a national dissemination meeting in August 2003, where the key concern expressed was reducing drug prices and addressing the issue of extra allowances.

The government took the question of drug availability seriously and in the following months, mifepristone was registered and a local drug company working in collaboration with a German multinational was allowed to import the active ingredient and manufacture mifepristone tablets in-country, bringing the cost per tablet down to 85,000 VND (about US$5.60). The re-registration process of misoprostol was also started. Sales remain restricted to authorised pharmacies and neither mifepristone nor misoprostol are on the essential drugs list as yet. The question of extra allowances also remains to be resolved at this writing.

Training activities received a boost when the MOH on its own and in collaboration with NGO-led efforts organised several training programmes throughout the country. By March 2004, over 100 providers (both doctors and midwives) from over 20 sites had received basic training. More recently, home use has been introduced by some central hospitals as part of medical abortion service delivery.

The future

Vietnam has a relatively well developed public health infrastructure and facilities for early abortion using vacuum aspiration already exist at all levels. Integrating medical abortion into this system will expand women’s ability to choose the method more suited to their needs. But choice is a challenge to achieve. The risk of individual facilities or the health system abandoning one method in favour of the other is real. To ensure choice, both methods must be available and providers must allow the counselling and decision making to be women-focused and not provider driven. This is not easy to achieve in a setting like Vietnam, where an established method exists and counselling has not been always been seen as important.

Both providers and women must know more about medical abortion. Recent efforts at training have seen knowledge about the method and the guidelines percolate down to lower levels of the health system but an even wider awareness base is essential. At the moment, there is only word-of-mouth for women to learn of this option apart from provider information.

Economic considerations must not be allowed to make choice a theoretical one. Thus the recent efforts to allow local formulation of mifepristone are welcome as they will reduce costs. But for medical abortion to be a viable option for the average Vietnamese woman, it will eventually have to be available at the lowest level of the health system (i.e. commune level) through mid-level providers, and with home use an acceptable option. Also, unless the issue of extra allowances is resolved, it will prejudice providers against medical abortion in the longer run.

The fact that abortion is legal in Vietnam and services are widely available means that widespread over-the-counter use is less likely to occur. Nevertheless, self-medication and direct purchase of drugs from pharmacies is common practice in Vietnam. Once these two drugs become cheaper and more widely available, controls on drug availability and appropriate education for pharmacists and the public will be needed. The implications of public—private sector interactions in terms of access and costs to women also need to be studied.

The Vietnam experience has important lessons for introducing medical abortion into the public health system. Documenting this experience is important not only for Vietnam but for other developing countries which are considering introducing this technology.

Acknowledgements

The paper is an abbreviated and slightly modified version of the full report of the assessment submitted to the Ministry of Health. The assessment was supported by the Ford Foundation Vietnam and the World Health Organization. The active support of the Reproductive Health Department, Ministry of Health, especially Dr Nguyen Dinh Loan, Director; Dr Tran Thi Phuong Mai, Vice Director; and Dr Nguyen Duy Khe, Vice Director; and of Ipas colleagues Do Thi Hong Nga, Ann Leonard, Traci Baird, Janie Benson, Alyson Hyman, Ronnie Johnson, Niki Jagpal and Jeremy McCamic is acknowledged. Nguyen Phuong Mai assisted in interpreting and translating between Vietnamese and English. Jane Hughes, Vu Quy Nhan and Jennifer Blum, Population Council, shared findings of ongoing research and commented on drafts of the report.