Contraception 76 (2007) 66 – 70
Original research article
Increasing access to safe abortion services in rural India: experiences with medical abortion in a primary health center Shuchita Mundle a , Batya Elul b , Abhijeet Anand c , Shveta Kalyanwala d,⁎, Suresh Ughade e a Department of Obstetrics and Gynaecology, Government Medical College, Nagpur 440003, India Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, NY 10032, USA c Formerly, Population Council, New Delhi 110003, India d Population Council, New Delhi 110003, India e Biostatistics, Department of Preventive and Social Medicine, Government Medical College, Nagpur 440003, India Received 5 December 2006; revised 14 February 2007; accepted 21 March 2007 b
Abstract Introduction: To increase access to safe abortion in rural India, the feasibility and acceptability of mifepristone–misoprostol abortion was assessed in a typical government run primary health center (PHC) in Nagpur district, Maharashtra State, that does not offer surgical abortion services and must refer off-site for emergency and backup services. Materials and Methods: Consenting pregnant women (n=149) with ≤56 days amenorrhea seeking terminations received 200 mg mifepristone, and returned 48 h later for 400-μg sublingual misoprostol and 12 days later for abortion confirmation. Surgical backup was conducted at a nearby community health center (CHC). Results: Nearly all women (98.6%) with known outcomes had successful medical abortions, and those who did not (1.4%) were successfully referred to the CHC for surgical backup. Women reported the method's ease and simplicity as the best features. Conclusion: Medical abortion provision is feasible and acceptable in an Indian rural PHC that does not offer surgical abortion services. This study suggests that introduction of medical abortion at lower levels of the health-care system could increase access to safe abortion in rural India. © 2007 Elsevier Inc. All rights reserved. Keywords: Medical abortion; Mifepristone; Misoprostol; Primary health centers; India
1. Introduction While abortion has been legal in India for more than 30 years following the enactment of the Medical Termination of Pregnancy (MTP) Act, unsafe abortion persists, particularly in rural areas1, where it accounts for 12% of all maternal deaths [1]. The MTP Act stipulates that all public sector facilities at the primary health center (PHC) level and higher are to provide abortion services, but such services are rarely available at the lower levels of the public health care system, which often serve as the first access point for reproductive health care, including abortion. Indeed, recent multistate abortion facility assessments estimate that only 5–58% of ⁎ Corresponding author. Population Council, India Habitat Center, New Delhi 110 003, India. Tel.: +91 11 2464 4008/4009; fax: +91 11 2464 2903. E-mail address:
[email protected] (S. Kalyanwala). 1 Rural India includes areas where 75% or more of the male population is engaged in agriculture as their main occupation. 0010-7824/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.contraception.2007.03.010
PHCs offer abortion services [2]. PHCs not offering these services commonly cite the absence of a provider trained in surgical methods or the lack of equipment and/or infrastructure required to provide surgical abortion services as barriers. In this context, mifepristone–misoprostol abortion, a noninvasive alternative for early pregnancy termination, holds great promise to increase access to safe abortion. In April 2002, the Drugs Controller of India approved 600 mg mifepristone coupled with 400 μg oral misoprostol for pregnancy termination in gestations of 49 days or less. While over 15 years of research has consistently demonstrated the safety, efficacy and acceptability of medical abortion worldwide and, in India at the facilities that provide surgical services [3–7], the present study expands on those overwhelmingly positive results to assess the feasibility of medical abortion provision in a rural PHC, which had not previously provided abortion services and did not have onsite surgical services for backup.
S. Mundle et al. / Contraception 76 (2007) 66–70
Although the efficacy of 200 mg of mifepristone for early medical abortion has been proven in numerous studies and endorsed by the World Health Organization [8–11], no consensus has been reached on the ideal dose and route of misoprostol. A pharmacokinetic analysis showed that sublingual misoprostol acts as quickly as oral misoprostol, quicker than vaginal misoprostol, and has a higher systemic bioavailability than all other routes of administration [12]. A recent observational study conducted in India using 400 μg oral misoprostol following 200 mg mifepristone in one clinic and 400 μg sublingual misoprostol following 200 mg mifepristone in another clinic reported higher efficacy for the sublingual route. Given these experiences, we tested a regimen consisting of 200 mg mifepristone followed 48 h later by 400 μg sublingual misoprostol administered in the PHC for women with gestations of 56 days or less. 2. Materials and methods The study was conducted at the government run PHC in a village in Nagpur district, Maharashtra state. The village, with a population of 35,000, is situated 10 km from a town where a community health center (CHC), a secondary care center, is located, and 50 km from the city of Nagpur where the Nagpur Government Medical College, a tertiary referral center, is located. The PHC is fairly typical for the State of Maharashtra: it has six beds, an operating theater and a labor room, where approximately seven uncomplicated vaginal deliveries occur weekly. The PHC does not have an ultrasound machine. At the time of the study, the PHC was staffed by two medical officers and three paramedical personnel. Although the PHC is mandated by law to offer abortion services and the medical officers were trained to conduct surgical terminations under the government Reproductive and Child Health Program, as the medical officers were not confident in their surgical abortion skills due to limited practice during their training, the service was offered onsite only when medical officers from the CHC visited to conduct monthly sterilization camps.2 Until the study was initiated, women who presented for abortion were asked to return during a sterilization camp or referred to the CHC or Nagpur Government Medical College for more immediate services. In preparation for the study, the two medical officers at the PHC and the chief medical officer from the CHC received a 2-day training session, which emphasized gestational age dating in early pregnancy using menstrual history and pelvic examination, abortion status confirmation using clinical parameters, clinical management of medical abortion and counselling techniques. Case studies on assessing abortion status and abnormal bleeding patterns were used during the 2 Sterilization camps refer to sterilization procedures carried out on a mass and monthly basis at a PHC for eligible, consenting men and women. Non-PHC providers travel to the PHC to assist with the sterilizations.
67
training. The medical officers were not trained to conduct surgical interventions for failed medical abortion but, rather, were instructed to refer all such cases to the CHC in the nearby town or to Nagpur Government Medical College, both of which have established surgical abortion services. While the PHC had a vehicle when the study was initiated, the study team ensured that it would be available with fuel and a driver 24 h a day to transport women referred for surgical intervention or requiring emergency backup care. This transport service was provided free of cost to women. All women seeking pregnancy termination at the PHC between August 2004 and October 2005 were screened for eligibility. Women were eligible to participate in the study if they had a positive urine pregnancy test, an intrauterine pregnancy of 56 days or less since the last menstrual period based on clinical exam and menstrual history, were in general good health, had no contraindications to mifepristone or misoprostol, lived within an hour of the PHC and were willing to return for at least two additional visits. Eligible women were informed about the availability of medical abortion at the PHC and of surgical abortion at the CHC and Nagpur Government Medical College. Those who opted for medical abortion were enrolled in the study after giving written informed consent, either by providing their signature or their thumbprint. Women were not compensated for participating in the study or reimbursed for their travel to the PHC. The study protocol was approved by the institutional review boards at the Population Council and Nagpur Government Medical College. The study protocol entailed a minimum of three clinic visits. On the first visit, women swallowed 200 mg mifepristone and were observed for about 15 min. Women returned to the clinic 48 h later for two sublingual 200-μg tablets of misoprostol. The provider ensured that the tablets were placed sublingually and also observed the woman for 4–6 h. On the day of misoprostol administration, women were told to expect bleeding and/or pain and were given an analgesic (paracetamol, 500 mg) to be used at their discretion. Women returned for follow-up 12 days later at which point a pelvic exam was done to assess abortion status. Those with complete abortion were discharged from the study. Women with incomplete abortions or ongoing pregnancies (assessed by pelvic examination and selfreported signs and symptoms) were referred to either the CHC or Nagpur Government Medical College for backup services. Women were also referred for surgical intervention if medically necessary and/or requested by the woman at any time during the study. Women were advised to return to or call the PHC at any time, either on the landline or the medical officer's mobile telephone if they had any problems or questions. At each clinic visit, medical officers collected clinical, experiential and acceptability data from women using standardized questionnaires that had been used in several medical abortion studies in India and other developing countries [3,4,6,7]. Women also completed a low-literacy
68
S. Mundle et al. / Contraception 76 (2007) 66–70
daily symptom diary card while enrolled in the study. Given the high prevalence of anemia in rural India and concerns about blood loss with medical abortion, hemoglobin was measured at study enrolment and at the follow-up visit. No posttreatment pregnancy test was given. Data were analyzed with SPSS 12.0 and STATA 8. 3. Results
Table 2 Efficacy rates (n=144) Success Failures Ongoing pregnancy at study end Incomplete abortion at study end Lost to follow-up Evidence of success No evidence either way
142 (98.6) 2 (1.4) 1 (0.7) 1 (0.7) 5 (3.3) 4 (2.7) 1 (0.7)
Data are presented as n (%).
3.1. Sample characteristics Of the 196 women who sought pregnancy termination at the PHC during the study period, 152 were eligible for participation in the study, and 150 were enrolled. One participant, who reported missed menses and had a positive pregnancy test and an enlarged uterus received both mifepristone and misoprostol but was subsequently found to have been incorrectly diagnosed as pregnant. As she did not experience any bleeding or cramping after taking either mifepristone or misoprostol, she was referred to Nagpur Government Medical College for surgical backup for suspected ongoing pregnancy. Clinical examination and ultrasound conducted there, however, indicated that she had never been pregnant. As indicated in Table 1, the remaining 149 study participants averaged 25.7 years of age and had completed 10.2 median years of schooling. Three quarters reported the present abortion to be their first pregnancy termination. The majority of women (75.8%) presented with gestational ages of 7 or 8 weeks. 3.2. Compliance and efficacy Two main components of compliance were examined: taking mifepristone and misoprostol at the proper time and returning for the final follow-up visit at the assigned time. About three quarters (72.5%) of participants complied perfectly with both aspects: they took both mifepristone and misoprostol as indicated and returned as scheduled for their follow-up visit. When women did not return as scheduled, the PHC staff made diligent efforts to trace them, calling (in the few cases where the woman had a phone) or visiting them at home. Of the 41 women who did not comply with the protocol, three took misoprostol late and
Table 1 Participants' characteristics (n=149) Age (years) [median (range)] Completed schooling (years) [median (range)] Married [n (%)] Primigravida [n (%)] Prior elective abortion [n (%)] Gestational age (weeks) [median (range)] Gestational age (weeks) [n (%)] ≤5 6 7 8
25.7 (16–41) 10.2 (0–17) 148 (99.3) 5 (3.3) 37 (24.9) 6.8 (5.8) 3 33 49 64
(2.1) (22.1) (32.9) (42.9)
33 were late for their follow-up visit after they had already taken both mifepristone and misoprostol. Most of these women indicated that they had not returned to the clinic as scheduled because they had gone out of town unexpectedly or were having household problems. The remaining five women (3.3%) were lost to follow-up. All but one had already taken misoprostol. Four of these five women had expelled the products of conception during the postmisoprostol observation period. In the remaining case, the available information was not sufficient to make a preliminary determination of the medical abortion outcome. Following established methods of efficacy classification for medical abortion [13], 98.6% of the 144 women with known outcomes had successful medical abortions (Table 2). One of the two women with failed medical abortions had an ongoing pregnancy and was referred to the CHC for surgical intervention without incident. The other woman was discharged from the study with a complete medical abortion at her follow-up visit. She subsequently returned to the PHC and reported that, after being discharged, she had heavy bleeding and underwent an evacuation procedure for incomplete abortion at a private provider facility. As the provider in question could not be contacted, it was difficult to confirm the reported reason for the intervention.
Table 3 Abortion experience (n=144)
Side effects and abortion symptoms Any bleeding Heavy bleeding Normal bleeding Spotting Nausea Vomiting Abdominal cramps Fever/chills Experiences with pain Severity of pain experienced, median (range) a Hemoglobin Median change (g/dL) over study period (range)
Prevalence [n (%)]
Median days (range)
149 (100.0) 124 (86.1) 133 (92.3) 124 (86.1) 114 (76.5) 68 (45.6) 97 (65.1) 43 (28.9)
8.6 (1–15) 2.7 (1–10) 3.4 (1–12) 3.6 (1–11) 2.6 (1–12) 1.6 (1–3) 3.0 (1–13) 1.6 (1–4) 2.4 (1–5)
0.1 (0.0–0.8)
Data are presented as n (%). a Pain was rated using a seven-point Visual Analogue Scale.
S. Mundle et al. / Contraception 76 (2007) 66–70 Table 4 Best and worst features of medical abortion (n=144) a Best features Easy, simple method Quick results, early expulsion Outpatient procedure, no anaesthesia Less painful Free service Natural experience Less bleeding Others Worst features None, no bad feature Pain and backache Prolonged bleeding, excess bleeding Weakness and giddiness Nausea and vomiting Too many visits Incomplete abortion Taste of sublingual misoprostol
41 (28.4) 31 (21.5) 31 (21.4) 24 (16.6) 10 (6.9) 3 (2.0) 2 (1.3) 2 (1.3) 109 (75.6) 9 (6.2) 7 (4.8) 7 (4.8) 6 (4.1) 2 (1.3) 1 (0.6) 1 (0.6)
Data are presented as n (%). a Participants could cite up to two best features and two worst features.
3.3. Abortion experiences No woman experienced any serious complications as a result of her medical abortion. Similarly, no blood transfusions or hospitalizations were required. As shown in Table 3, all women (100.0%) reported at least some bleeding and documented an average of 8.6 days of bleeding in the daily symptom cards. Most of this bleeding was described as similar to or less than that experienced during a normal menstrual cycle. The median change in hemoglobin over the study period was 0.1 g/dL. Other side effects mentioned included nausea (76.5%) and vomiting (45.6%) lasting for a median of 2.6 and 1.6 days, respectively. Abdominal cramping was also common (65.1%) and experienced for 3 days on average. Women classified the severity of pain experienced using a seven-point Visual Analogue Scale and, on average, reported low levels of pain, with a median score of 2.4. Nearly all (96.9%) women reported that they had received adequate pain medication. Of these, only 22.9% of participants reported using the analgesic tablets that they had received at the time of enrolment. Information on additional care required by women was collected — unscheduled visits or calls made to the PHC — to assess how they managed their side effects and abortion experiences. Only nine (6.0%) women made unscheduled visits to the PHC, mainly with questions about their abortion status and bleeding pattern and/or to confirm whether they could breast-feed during the medical abortion process. Slightly more women — 17 or 11.4% — called the PHC. Most of these calls were to confirm or postpone an appointment, but some women also called with concerns about the amount and duration of their bleeding. 3.4. Acceptability Participants were asked to rate their overall abortion experience as satisfactory, neutral or unsatisfactory. All
69
women (100.0%) with known outcomes rated their experience to be satisfactory, including those with failed medical abortions. One quarter (25.0%) of these women reported that they were “very prepared” for what happened during their abortion, while the remainder indicated that they were “somewhat prepared.” Nearly all (97.9%) reported that they would select medical abortion again if they needed another abortion, and 19.1% indicated that they would prefer home administration of misoprostol for future medical abortions. Open-ended questions were used at the follow-up visit to identify the best and worst features of women's experiences with medical abortion (Table 4). Almost half the women (49.9%) rated the ease and speed of the abortion amongst its best features. One fifth (21.4%) appreciated that the abortion was done in an outpatient setting and/or that no anesthesia was required and 16.6% of women perceived medical abortion to be a less painful method than the surgical alternative. The other best features cited by the women included that the method was available free of cost (6.9%) and that the abortion was “natural” or like their menses (2.0%). When asked about the method's worst features, the great majority (75.6%) of women reported that there were none. Even though many women reported experiencing pain and/or nausea and vomiting, few women (pain and/or backache: 6.2%, nausea and vomiting: 4.1%) reported these aspects of their abortion experience among the worst features of the method.
4. Discussion Our findings suggest that mifepristone–misoprostol abortion can be offered in a rural PHC with no surgical abortion services onsite, without increasing or changing existing staff configurations. The high success rate (98.6%) and low loss to follow-up rate (3.3%), coupled with the universal acceptability (100.0%) of the method, indicate that medical abortion should be included in the current package of services offered at PHCs in rural India. Indeed, when provided with a short training on dating early pregnancy and diagnosing complete medical abortion, medical officers with limited prior training in and no experience providing surgical abortion can offer medical abortion services in lower-level health care settings, as long as surgical services are available for backup at a referral facility. Our experience suggests that good community outreach is an essential component of successful medical abortion provision in rural settings. The medical officers and other staff at the PHC where our research was conducted, including community health outreach workers, had longstanding relationships with the community, which ultimately were used to inform women about the availability of the method and served as an entry point for home visits to trace women who were late for scheduled visits.
70
S. Mundle et al. / Contraception 76 (2007) 66–70
While our study did not aim to assess the safety and effectiveness of sublingual misoprostol following mifepristone, our high success rate (98.6%), particularly notable in a rural setting among providers with no previous medical abortion experience, suggests that further attention should be paid to this route of administration. While the prevalence of side effects reported in this study was only marginally higher than that reported in other medical abortion studies conducted in India using a similar dose of oral misoprostol [4,14], a randomized controlled trial comparing 800 μg sublingual and vaginal misoprostol following 200 mg mifepristone noted statistically significant higher rates of fever, chills and gastrointestinal side effects among women who received sublingual administration and higher rates of fever and chills than noted in our study [15]. Given this, researchers should consider including a regimen of 200 mg mifepristone followed by a reduced 400-μg dose of sublingual misoprostol in future double-blinded randomized controlled trials examining the optimal route and dose of misoprostol. The findings of our study should be interpreted in light of the following limitations. As the study was conducted in Maharashtra, which has a relatively good infrastructure when compared to other states in India, its generalizability to other PHCs in the country may be limited. PHCs in other more resource-poor states may not have backup surgical services within a reasonable distance, as was the case in the present study setting, limiting the feasibility of referral for backup services for the infrequent cases requiring immediate surgical intervention. Given these limitations, researchers should continue to examine early medical abortion provision in PHCs across India, including in states with extremely poor infrastructure where the need for safe abortion services is the most critical. Studies examining provision of medical abortion by nurses and other midlevel providers who provide the bulk of services at lower-level public health facilities in India are also needed. As medical abortion use is quickly increasing in India, however, future efforts should not be restricted to research. Carefully designed, in-service and user-friendly provider training and patient information materials — including those for low literacy groups — are urgently needed if the method is to be provided safely and acceptably through the public health care system. Acknowledgments The authors would like to thank Shireen Jejeebhoy for comments on an earlier version of this article. They would also like to thank the medical officers and the paramedical
personnel at the PHC, as well as the medical officers at the CHC for their enthusiasm and commitment to expanding access to safe abortion. They express their gratitude to Naina Kunnawar, Suvarna Joshi and Rakhi Tarapder for their inputs to the study and acknowledge all of the women who participated in the study. References [1] Government of India. Survey of causes of maternal deaths (rural): annual report. New Delhi: Office of the Registrar General of India; 1990. [2] Khan ME, Barge S, Kumar N, Almroth S. Abortion in India: current situation and future challenges. In: Pachauri S, editor. Implementing a reproductive health agenda in India: the beginning. New Delhi: Population Council; 1999. p. 507–29. [3] Coyaji K, Elul B, Krishna U, et al. Mifepristone abortion outside the urban research hospital setting in India. Lancet 2001;357:120–1. [4] Coyaji K, Elul B, Krishna U, Otiv S, et al. Mifepristone–misoprostol abortion: a trial in rural and urban Maharashtra, India. Contraception 2002;66:33–40. [5] Takkar D, Aggarwal N, Seghal R, Buckshee K. Early abortion by mifepristone (RU 486) followed by vaginal gel (metenprost) versus oral (misoprostol) prostaglandin. Advances in Contraception 1999;15:163–73. [6] Winikoff B, Sivin I, Coyaji KJ, et al. Safety, efficacy, and acceptability of medical abortion in China, Cuba, and India: a comparative trial of mifepristone–misoprostol versus surgical abortion. Am J Obstet Gynecol 1997;176:431–7. [7] Winikoff B, Sivin I, Coyaji KJ. et al. The acceptability of medical abortion in China, Cuba and India. Int Fam Plann Perspect 1997;23:73–8, 89. [8] World Health Organisation Task Force on Post-Ovulatory Methods of Fertility Regulation. Termination of pregnancy with reduced doses of mifepristone. BMJ 1993;307:532–7. [9] World Health Organisation Task Force on Post-Ovulatory Methods of Fertility Regulation. Comparison of two doses of mifepristone in combination with misoprostol for early medical abortion: a randomised trial. Br J Obstet Gynaecol 2000;107:524–30. [10] Elul B, Hajri S, Ngoc NN, et al. Can women in less-developed countries use a simplified medical abortion regimen? Lancet 2001;357:1402–5. [11] World Health Organisation. Safe abortion: technical and policy guidance for health systems. Geneva: World Health Organisation; 2003. [12] Tang OS, Schwer H, Seyberth HW, Sharon WH, Ho PC. Pharmacokinetics of different routes of administration of misoprostol. Hum Reprod 2002;17:332–6. [13] Winikoff B, Ellertson C, Clark S. Analysis of failure in medical abortion. Contraception 1996;54:323–7. [14] Elul B, Ellertson C, Winikoff B, Coyaji K. Side effects of mifepristone–misoprostol abortion versus surgical abortion. Contraception 1999;59:107–14. [15] Tang OS, Chang CCW, Ng EHY, Lee SWH, Ho PC. A prospective, randomized, placebo-controlled trial on the use of mifepristone with sublingual or vaginal misoprostol for medical abortions of less than 9 weeks gestation. Hum Reprod 2003;11:2315–8.
