The introduction of first trimester medical abortion in Armenia

Abstract

In Armenia, abortion is the main means of fertility regulation; however, before research activities were initiated only surgical methods were available and the quality of services was low in some areas. Our clinical study from 2008–2011 aimed to show that early medical abortion is an acceptable and feasible option. A total of 700 eligible women with pregnancies up to 63 days LMP presenting for abortion were recruited for the study in five locations. Participants took 200 mg mifepristone and 800 μg buccal misoprostol 24–48 hours later. They returned for a follow-up visit two weeks after mifepristone administration. 95% of the women had successful abortions and 95% were satisfied with the method. In 2012–2013, we conducted a follow-up assessment to examine the ongoing provision and quality of medical abortion services at the former research sites. Medical record reviews, interviews and observations were carried out three times approximately six months apart. The assessment found that all five sites had continued providing medical abortion, with about half of eligible women choosing the medical method. Four of the five sites were achieving high success rates. Staff turnover and the lack of trained providers likely contributed to the higher failure rate at the fifth site. These findings provide evidence that first trimester medical abortion is an acceptable and feasible option for Armenian women and providers, and that high quality services are being delivered.

Résumé

En Arménie, l’avortement est le principal moyen de régulation de la fécondité ; néanmoins, avant le début des activités de recherche, seules les méthodes chirurgicales étaient disponibles et la qualité des services était faible dans certaines régions. Notre étude clinique de 2008 à 2011 souhaitait montrer que l’avortement médicamenteux précoce est une option acceptable et faisable. Un total de 700 femmes réunissant les conditions requises, avec une grossesse de moins de 63 jours depuis la DDR, qui s’étaient présentées pour un avortement ont été recrutées pour l’étude dans cinq sites. Les participantes ont pris 200 mg de mifépristone et 800 μg de misoprostol par voie buccale 24–48 heures après. Elles sont revenues pour une visite de suivi deux semaines après l’administration de la mifépristone. 95% des femmes avaient avorté et 95% étaient satisfaites de la méthode. En 2012–2013, nous avons mené une évaluation complémentaire pour examiner la prestation et la qualité des services d’avortement médicamenteux dans les précédents sites de recherche. Un examen des dossiers médicaux, des entretiens et des observations ont été réalisés à trois reprises à environ six mois d’intervalle. L’évaluation a révélé que les cinq sites avaient continué à pratiquer des avortements médicamenteux et que près de la moitié des femmes pouvant y prétendre choisissaient cette méthode. Quatre des cinq sites obtenaient des taux élevés de réussite. Le roulement du personnel et le manque de prestataires formés ont probablement contribué au taux plus élevé d’échec dans le cinquième site. Ces conclusions montrent que l’avortement médicamenteux du premier trimestre est une option acceptable et faisable pour les femmes et les prestataires arméniens et que des services de qualité sont assurés.

Resumen

En Armenia, el aborto es el principal medio para regular la fertilidad; sin embargo, antes de que se iniciaran las actividades de investigación, solo había métodos quirúrgicos disponibles y los servicios eran de baja calidad en algunas regiones. Nuestro estudio clínico realizado de 2008 a 2011 tuvo como objetivo mostrar que el aborto con medicamentos temprano es una opción aceptable y factible. En cinco lugares, se reclutó para el estudio un total de 700 mujeres elegibles con embarazos de hasta 63 días desde la FUM, que acudieron a la unidad de salud para tener un aborto. Las participantes tomaron 200 mg de mifepristona y 800 μg de misoprostol por vía bucal 24 a 48 horas después. Regresaron para una consulta de seguimiento dos semanas después de administrada la mifepristona. El 95% de las mujeres tuvo un aborto completo y el 95% estuvo satisfecha con el método. En 2012–2013, realizamos una evaluación de seguimiento para examinar la prestación continua y calidad de los servicios de aborto con medicamentos en los lugares donde se realizó la investigación. Se llevaron a cabo revisiones de los expedientes médicos, entrevistas y observaciones tres veces, a intervalos de aproximadamente seis meses. La evaluación encontró que los cinco lugares habían continuado prestando servicios de aborto con medicamentos, y que aproximadamente la mitad de las mujeres elegibles eligieron el método con medicamentos. Cuatro de los cinco lugares estaban logrando altas tasas de aborto completo. La rotación de personal y la falta de profesionales de la salud capacitados probablemente contribuyeron a una mayor tasa de aborto fallido en el quinto lugar. Estos hallazgos ofrecen evidencia de que el aborto con medicamentos en el primer trimestre es una opción aceptable y factible para las mujeres y profesionales de la salud en Armenia, y que se están prestando servicios de alta calidad.

Keywords:

• medical abortion
• mifepristone
• misoprostol
• first trimester pregnancy
• Armenia

Abortion has been legal and widely available in Armenia for over 50 years. Currently, the law allows for termination of pregnancies up to 12 weeks without restrictions. However, the quality of abortion services has left much room for improvement. As documented in a 2009 United Nations Population Fund report, abortions accounted for 12.5% of maternal deaths.¹ Prior to the initiation of our clinical study, induced abortion was synonymous with surgical abortion and could only be performed by obstetrician–gynaecologists at inpatient units.² Most surgical abortion procedures were performed by vacuum aspiration, mechanical dilatation of the cervix prior to aspiration was a common practice, and pain management protocols were not available.³ Although mifepristone and misoprostol were approved in 2007 and 2008, respectively, no distribution system for the medications was in place, national medical abortion guidelines were not established, and clinical providers lacked training in the use of evidence-based regimens.

Mifepristone–misoprostol medical abortion could provide a safe and effective option for abortion in Armenia. We conducted a clinical study to examine the acceptability and feasibility of introducing a medical abortion regimen of 200 mg mifepristone followed by 800 μg of buccal misoprostol for pregnancies up to 63 days since the last menstrual period (LMP) in the capital city, as well as in two regions of Armenia. Following this introduction, we completed an assessment of the ongoing provision and quality of medical abortion care at participating sites.

Methods

Clinical study

The first stage of the clinical study was conducted from April 2008 to March 2009 at three clinics in the capital of Yerevan (Republican Institute of Reproductive Health, Perinatology, Obstetrics and Gynaecology (RI); Centre of Family Planning and Sexual Health (CFP); and State Medical University (SMU)). The second stage was conducted from December 2010 to November 2011 at sites in two regions of Armenia (Vanadzor Hospital #1 (VH), Vanadzor and Gyumri Maternity Hospital (GMH), Gyumri).^* Clinicians experienced in providing surgical abortions were trained in mifepristone-misoprostol medical abortion provision and study implementation. The Allendale Investigational Review Board, Old Lyme, CT, USA reviewed and approved the study protocol. All women gave written informed consent.

Women who desired medical abortion to terminate an intrauterine pregnancy of 63 days LMP or less were recruited to participate. To enrol in the study, they had to be 18 years or older, were expected to live or work within one hour of the study site and needed to be willing to return for a follow-up visit two weeks later. Women with suspicion of ectopic pregnancy, concurrent anticoagulant therapy, history of haemorrhagic disorders, concurrent long-term corticosteroid therapy, adrenal insufficiency, inherited porphyria, or known allergy to mifepristone or misoprostol were not eligible. Menstrual history, physical exam and/or ultrasound were used to determine the duration of pregnancy.

Medical abortion patients in Yerevan who chose to participate swallowed one 200 mg mifepristone pill (Pentcroft Pharma or Mirpharma, Russia) in the clinic; all but 30 women were given the option of taking misoprostol (Cytotec, Pfizer, USA or Mirolut, Mirpharma, Russia) either at the clinic or at home. Because providing medical abortion was new to study staff, the first ten women enrolled at the three sites in the capital were instructed to return to the clinic for misoprostol administration. In the second stage, women participating in the regions were offered the option to take both the mifepristone and misoprostol at home. Women were told to take the misoprostol 24 to 48 hours after ingesting the mifepristone. Follow-up visits were scheduled two weeks after mifepristone administration.

Women who chose to administer the misoprostol at home were given four 200 μg pills of misoprostol and informed that they could return to the clinic to take the pills if they changed their minds. Participants were instructed to take the misoprostol buccally by placing two tablets between the cheek and gum on each side of the mouth and letting the pills dissolve for 30 minutes, after which they were to swallow any remaining fragments. Those choosing in-clinic administration were scheduled to return to the clinic two days later to take the second set of pills in the presence of study staff. Women were advised to take 500 mg of paracetamol as needed for pain. Participants were asked to document their abortion experiences for two weeks following mifepristone administration on a home study card. Women in Yerevan recorded when the misoprostol was taken, the level of bleeding, the occurrence of expulsion and any side effects, and if extra drugs were taken for pain or other symptoms. Women from the regions recorded when mifepristone and misoprostol were administered, number of missed days of work or school, and hours that they needed help with childcare. All were told to call the clinic if they had any questions, and to return to the clinic if heavy bleeding occurred (defined as bleeding through two large sanitary pads per hour for more than two consecutive hours) or if they experienced fever lasting six hours or more.

At the follow-up visit, providers determined women’s abortion status by using clinical examination and ultrasound, as needed. Surgical intervention, either electronic or manual vacuum aspiration, was recommended to women with ongoing pregnancies (defined as embryonic or fetal growth since the first visit with the presence of cardiac activity). Those women with heavy bleeding or an incomplete abortion (defined as having intrauterine debris or retained products of conception) for whom treatment was recommended were offered either surgery or waiting another week with or without an additional dose of 800 μg buccal misoprostol. If women still had heavy bleeding or an incomplete abortion after the additional week, aspiration was performed. Participants could request surgery at any time during the study. If the provider determined the abortion to be complete, that is, the pregnancy had been aborted and no further treatment or follow-up was recommended, women were interviewed and discharged from the study.

The clinical study was intended to assess the acceptability and feasibility of mifepristone-misoprostol medical abortion. The sample sizes for both stages of the study were selected to allow staff to gain expertise in providing this new method. Based on previous reports,^4–6 we expected an efficacy rate of about 96%.

Follow-up assessment

After the clinical study was completed, we returned to the research sites to assess ongoing provision of medical abortion services and outcomes. Data were collected by a research consultant over one week at each site, three times at roughly six-month intervals, beginning April 2012 and ending June 2013. At each time point, we collected the following data. A retrospective medical record review of medical abortion patients was performed, beginning with the date of data collection and proceeding backwards until 50 cases were reviewed. Key data recorded included medical abortion regimen used, mifepristone and misoprostol place of administration, abortion success, mode of follow-up, and whether any serious adverse events occurred. Early first trimester surgical abortion patients were approached in the recovery room over the week and consenting patients were interviewed about their experience with abortion counselling, awareness of the medical abortion method, and reasons for choosing aspiration abortion. All abortion providers at each site, all of whom were obstetrician/gynaecologists, were also interviewed. They were asked about the volume of surgical and medical abortions they performed monthly and the limits to the duration of pregnancy of the procedures, their beliefs on abortion safety, whether they offered medical abortion to all eligible women, the medical abortion training they had received, and if they were not providing medical abortion, the reasons why. All those interviewed provided written informed consent.

To supplement these data, facility records, policies and protocols were reviewed for information on total abortion volume, medical abortion regimen, eligibility criteria and costs of services. Pharmacists and administrators, after consenting, were also interviewed to obtain similar information. Lastly, abortion counselling sessions provided in-person or over the phone were observed to document if medical abortion was being offered to all eligible women. Any perceptions of provider bias towards one or another abortion method, inaccuracies of abortion information provided, any concerns voiced by the women, and any other notable observations were also recorded.

All data collected for the clinical study and follow-up assessment were entered into SPSS version 13.0 at Gynuity Health Projects in Tbilisi and sent to Gynuity Health Projects in New York for analysis in SPSS version 15.0.

Findings

Clinical study

The five study sites enrolled a total of 700 women into the clinical study: 450 (64.3%) in the capital and 250 (35.7%) in the regions. Two participants were lost to follow-up. No serious adverse events were reported. The median age of the women was 29 years, and the vast majority were married and had had at least one live birth. Just over half had had a previous abortion. Women from Yerevan were more educated than the women from the regions, with 65.3% vs. 38.8%, respectively, completing a university-level education (Table 1).

Table 1 Characteristics of clinical study participants by site (n = 700).^a

The following abbreviations are used:

RI (Republican Institute of Reproductive Health, Perinatology, Obstetrics and Gynaecology, Yerevan)

CFP (Centre of Family Planning and Sexual Health, Yerevan)

SMU (State Medical University, Yerevan)

VH (Vanadzor Hospital #1, Vanadzor)

An overall success rate of 95.3% was achieved (95% CI: 93.5%, 96.7%) with mifepristone and misoprostol, with success defined as abortions completed without surgical intervention at any point during the study (Table 2). Success rates were similar across sites as well as between home mifepristone users and clinic mifepristone users. As length of pregnancy increased, success rates remained high, with 94.4% of women with pregnancies of 57–63 days LMP having successful abortions. Seventeen of 18 women who took an additional dose of misoprostol for incomplete abortions had successful abortions without surgical interventions. In total, 33 women had aspirations, 12 of whom were diagnosed with ongoing pregnancies. Rates of ultrasound use at follow-up were higher in the regions than in the capital, at 75.7% vs. 48.4% respectively.

Table 2 Treatment outcome (n = 698)^ab.^c

Around 63% of women who had the option chose to take the mifepristone at home rather than in the clinic. Women who worked outside the home were not significantly more likely to choose home mifepristone (66% vs. 62%, p = 0.31) than women who did not. More than half chose home mifepristone because they found it more compatible with their household duties or thought that it provided them flexibility with their schedules (Table 3). The majority of clinic mifepristone users liked the presence of clinic staff and believed it would help them feel less anxious about the process. Nearly all women who chose home mifepristone took the pill the same day as their initial appointment; no one waited longer than three days to take the pill after seeing the provider. All except one woman in the home user group took the mifepristone at the scheduled time (defined as within a window of ± 6 hours of the time they had selected with the provider) (Table 4). This woman took the mifepristone two days later than planned and accordingly adjusted the time she took misoprostol; she had a successful abortion.

Table 4 Timing of mifepristone and misoprostol administration^a.^b

Table 3 Top reasons for choice of mifepristone location among women in the region: n (%).^a

Aside from the first ten women enrolled at each of the Yerevan locations, the option to take the misoprostol at home was offered to all women in the capital and in the regions. Ninety-seven per cent of women chose to take the misoprostol at home; this did not vary by location (Table 4). Though all women were told to take the misoprostol 24–48 hours after mifepristone, two took the misoprostol just after that time frame (and had successful abortions).

Participants were asked about the side effects they experienced during the abortion (Table 5). Women from the capital documented them daily on the home study card, while those from the region were asked to recall side effects at their follow-up visit. Almost all women in the capital reported having at least one side effect, whereas 29% of women in the regions reported no side effects. Women most commonly experienced nausea, followed by vomiting, and then fever and chills. Eighty-nine per cent of participants found the side effects acceptable or very acceptable (Table 6). More than half (55.3%) did not use any analgesics during the abortion. Among those who did, most (92.9%) said that the medication was adequate to alleviate the pain.

Table 6 Women’s reported experience and acceptability: n (%).^a

Table 5 Side effects reported: %^ab.^c

Overall, almost all women (94.8%) thought the abortion counselling they received from their provider was sufficient regarding the amount of pain, bleeding and time to complete abortion. Six per cent of women called the clinic and five per cent made an unscheduled visit due to concerns about the abortion. The proportion of women making these extra contacts was not greater among the home mifepristone group than the clinic mifepristone group. However, women from the regions called the clinic more often than those from the capital, 10.4% vs. 2.9%, respectively. The majority of unscheduled visits were made because women wished to confirm success prior to their scheduled follow-up visit (60%, 21/35). In the regions, only about a quarter of women worked outside of the home and fewer than 5% were students. Very few of these women reported that the abortion interfered with their job or academic responsibilities. Eighty-six per cent of the participants had children and the majority said they had not needed extra assistance with childcare.

At the exit interview, participants were asked about their experience with and acceptability of the medical abortion (Table 6). Around 95% were satisfied or very satisfied with the method. The majority of women preferred the medical method over aspiration if they needed another abortion. In the regions, 94.8% of participants who chose home mifepristone would do the same in the future, while 90.1% of clinic users would switch to home administration. Women at all sites seemed satisfied with taking the misoprostol at home, with 96.0% preferring home use if they needed to make that choice later. Medical abortion would also be recommended to a friend by 96.1% of participants. Regional providers were overwhelmingly supportive of the home mifepristone option; they recommended home use in the future for 96.3% of participants if they were to have another abortion.

Women were also asked to report the best and worst features of the abortion (Table 6). More than three-quarters of participants felt that being able to avoid surgery and anaesthesia was the best feature, followed by having a successful abortion without complications. Among home mifepristone users, 43.9% liked being able to control when the bleeding would occur. Overall, more than half felt that the pain and cramps was the worst feature. Some women reported that there were no best (2.6%) or worst (11.1%) features of the abortion.

Assessment

Assessment activities were initiated five months to three years after the clinical study concluded. The results show that all five of the research sites have continued providing medical abortion up to 63 days LMP. About half of the women with pregnancies ≤ 63 days chose medical abortion during the case review period of interest (Table 7); this proportion did not vary much across the three time points (47.4%, 50.8%, 43.9%). However, uptake varied between the sites, ranging from 29.7% at GMH to 75.0% at CFP. The low proportion of women choosing medical abortion at GMH prevented us from reaching our target case review sample at each round of data collection.

Table 7 Overall uptake of medical abortion among eligible women and treatment outcome during the assessment case review periods by site (n = 710)^a.^b

A total of 710 medical abortion records were reviewed. Overall, the vast majority of women (96.8%) took 200 mg mifepristone and 800 μg misoprostol for their medical abortion. Almost all women (98.5%) administered the misoprostol buccally. Home administration of mifepristone was chosen by 40.9% of women, while 91.8% chose home administration of misoprostol. Four of five sites achieved overall success rates of 93% or above (Table 7). Very few women (1.3%) were lost to follow-up, and no serious adverse events were documented in the medical records. Rates of ultrasound use to determine abortion status were high in both the capital and in the regions, at 90.4% and 84.0%, respectively.

We interviewed 69 women who received surgical abortions during the weeks of data collection. The majority (81.2%) reported that they had heard of medical abortion, but only half (50.7%) were actually offered this method by their provider. Among those women, 15 reported choosing aspiration abortion because they did not want to deal with the bleeding and side effects and 11 wanted the abortion over more quickly. Less than one quarter (23.2%) of them said they would prefer medical abortion if they needed a future abortion.

A total of 121 clinical provider interviews were conducted (Round 1, n = 48; Round 2, n = 44; Round 3, n = 29). Some may have been interviewed at more than one round of the assessment. All clinicians interviewed were familiar with medical abortion, and almost all reported offering the method to some or all of their patients. Medical abortion was believed to be equally safe or safer than aspiration by almost all providers as reported during the three rounds (95.9%, 97.7%, 96.5%, respectively). The four clinicians who did not offer medical abortion believed that aspiration abortion was the safer method.

The review of facility documents revealed that, across all sites, the cost of medical abortion was similar to, if not less than, surgical abortion. Costs remained fairly stable over the three assessment periods. Of the 70 counselling sessions observed, medical abortion was almost always offered (94.2%), and most providers gave detailed and accurate information about the method. Women from more rural areas voiced more fear and scepticism of medical abortion, and therefore were rarely interested in pursuing this option.

Discussion

Findings and interpretation

Our clinical study shows that mifepristone-misoprostol medical abortion is highly acceptable and feasible to use for women and providers in Armenia. Success rates similar to those reported in the literature^4–6 were achieved in the capital city of Yerevan, as well as in the regional cities of Vanadzor and Gyumri. The majority of women were very satisfied or satisfied with their abortions, would choose the medical method and would prefer home administration of mifepristone and misoprostol if they needed to terminate an unwanted pregnancy in the future.

Women who take mifepristone at home may have better opportunities to schedule the bleeding around their personal responsibilities, and previous research has shown that this option is safe and acceptable to women and providers.^6,7 Over 60% of women given the option administered the mifepristone at home, and the majority of them chose it because they thought it was more compatible with their household duties. Home administration of misoprostol is common in many countries,⁸ and many women in our study also preferred that option, with almost all choosing to do so.

Home users of mifepristone or misoprostol did not burden the clinic staff with extra contacts. Home mifepristone users did not appear to make more calls or in-person visits than clinic users, 9.6% vs. 9.2%, respectively. Practically all women who chose home administration took the medications at the scheduled times, lending further support to the idea that women are able to take the medicines on their own as directed. Providers were also confident in women’s ability to do so without their supervision, recommending home administration for almost all of their patients for the future.

Sixteen per cent of women enrolled at one site in Yerevan made an unplanned visit to the doctor, the highest rate among all sites; the majority of these women went on to have successful abortions. Shortening the period between the initial visit and the follow-up visit may be warranted since most of the unscheduled visits were made to confirm success, not because of pain or heavy bleeding.

Alternatively, in-person follow-up may not be necessary. WHO recently recommended that following an uncomplicated medical abortion using mifepristone and misoprostol, there is no medical need for a routine follow-up visit or ultrasound.⁹ The next step to lessen the burden on women and providers in Armenia is to advise that once clinicians are comfortable and confident in their skills to provide and counsel for medical abortion, a self-administered pregnancy test^10,11 with phone follow-up¹² may be feasible in most cases, saving the time, costs and resources associated with an additional visit to the clinic and unnecessary use of ultrasound.

Before the initiation of our research, medical abortion services were effectively not available in the country. However, the assessment has shown that mifepristone–misoprostol medical abortion is currently well established, with substantial uptake of the method at all sites. More than one quarter of women who attended the regional clinics and more than half who attended the clinics in the capital chose the medical method. All clinics continue to deliver services primarily using the evidence-based regimen of 200 mg mifepristone followed by 800 μg buccal misoprostol, and all but one are achieving high rates of success. All providers continue to offer women the option of home administration of misoprostol. Additionally, women in the region are being offered home administration of mifepristone by their clinicians, as they were trained for the clinical study.

Nevertheless, we identified several areas for improvement. High staff turnover and the lack of trained, experienced providers at CFP, unlike the other sites, were likely contributing to the higher failure rate at that clinic. Additionally, women at CFP were undergoing surgical interventions at higher than usual rates, suggesting the need for providers to improve their diagnostic skills or become more comfortable with offering women the options of expectant management or medical management with an additional dose of misoprostol. The use of ultrasound at follow-up, already at high rates during the clinical study, continued at high rates at all sites. Providers should be encouraged to use ultrasound only for complicated or problematic cases. Since women residing in remote areas voiced fear, scepticism and lack of information about the medical option, increasing knowledge may change women’s attitudes in these areas.

Strengths and weaknesses of the study

We conducted our clinical study in the capital of Armenia and in two regions of the country, generating local data on the acceptability and feasibility of medical abortion. Because of the very low lost-to-follow-up rate, abortion outcomes were assessed in almost all women, which is important for the validity of our results and also for the education and training of providers new to this method.

Women living in the capital city are more likely to work outside the home than those living in other regions. However, because they did not have the opportunity to take the mifepristone at home, we were unable to assess the uptake and acceptability of this option among those who may find it more beneficial to plan the bleeding. Also, the difference in methods used to obtain information on side effects limited our ability for a direct comparison between the capital and the regions.

The follow-up assessment was intended to get a sense of how well medical abortion services continued once research activities ended. However, the study design may have limited the comparability of results between sites. Since we initiated the assessment for all sites at the same time, the time that elapsed between the conclusion of the clinical study and initiation of the assessment varied between sites from five months to three years. Hence, differences in uptake between the regions and the capital may reflect the shorter time period rather than a lower acceptability of the method in the outlying regions. In addition, due to differences in patient volume and popularity of medical abortion, the period of time covered by each retrospective review of cases covered weeks at some sites and months at others. Counselling sessions may have been biased since providers knew they were being observed. A strength of the assessment is the different sources of data used to form a more complete understanding of the situation at each clinic.

Differences in results and conclusion in relation to other studies

Ongoing pregnancies were diagnosed in 1.7% of women in the clinical study. This rate falls within the range of 1–3% ongoing pregnancy rate for pregnancies ≤ 49 days LMP, regardless of regimen, found by Kahn et al. in their meta-analysis of the efficacy of medical abortion,¹³ and is also similar to the ongoing pregnancy rate of 1.1% in a review of first trimester mifepristone medical abortion conducted by Raymond et al.¹⁴

The uptake of home use of mifepristone has varied among women in different countries. In Armenia, 62.8% of women chose this option in our clinical study and 40.9% in the follow-up assessment, while 74% did so in Azerbaijan,⁷ and 46.2% did so in the United States.⁶ The top reasons why women chose this option differed as well, reflecting the flexibility that home use gives to women so that their medical abortion can conform to their priorities and circumstances.

Relevance of the findings: implications for clinicians and policymakers

Our introductory clinical study supported the inclusion of first trimester medical abortion into reproductive health services for women in Armenia and spurred local experts and health officials to convene a working group to develop national guidelines for practice, which were published in 2010. The guidelines have been disseminated to clinicians across the country and are an important reference to help them in the delivery of high quality medical abortion services.

The follow-up assessment reveals the importance of regularly monitoring outcomes and the many avenues through which medical abortion services can be enhanced. Recommendations have been made to the Ministry of Health to incorporate specific indicators into their national health statistics so that the quality of medical abortion care can be monitored systematically.

Conclusion

The clinical study and follow-up assessment provide evidence that mifepristone-misoprostol medical abortion is a safe, acceptable and sustainable option for Armenian women and providers. Having a simplified, non-surgical method for pregnancy termination is an important step towards providing improved, comprehensive reproductive health care for women.

1st trimester medical abortion and study training in Gyumri, December 2010

Acknowledgements

We acknowledge the technical and financial support of an anonymous donor. We thank our local investigators and coordinators in Armenia, Drs. Ruzanna Abrahamyan, Armine Harutyunyan, Anna Aghajanyan, Aram Avalyan, and Gayane Grigoryan for study implementation and Tatyana Lotarevich from Gynuity Health Projects, New York, for data support.

Notes

* Vanadzor and Gyumri are the next biggest cities in Armenia after Yerevan. VH and GMH are the main providers of abortion in the Lori and Shirak regions, respectively.

a GMH (Gyumri Maternity Hospital, Gyumri)

a Two women were lost to follow up.

b Success defined as abortions completed without surgical intervention at any point during the study period.

c Incomplete abortion defined as having intrauterine debris or retained products of conception.

a Women could choose more than one answer.

a Women who took the drug ± 6 hours of the scheduled time were considered to have taken the drug on time.

b Women were instructed to take the misoprostol 24–48 hours after mifepristone. Women who took the drug ±6 hours of the scheduled time were considered to have taken the drug on time.

a Women could choose more than one answer.

a Women from the capital reported side effects on the home study card.

b Women in the region reported side effects at their exit interview.

c Other reported side effects included headache, dizziness, and joint pain.

a Success is defined as abortions completed without surgical intervention at any point during the study period.

b n = 110 because there were not enough cases in the review period to reach target sample.