Contraception 91 (2015) 178 – 183
Original research article
Acceptability and feasibility of phone follow-up with a semiquantitative urine pregnancy test after medical abortion in Moldova and Uzbekistan☆,☆☆,★ Ingrida Platais a,⁎, Tamar Tsereteli b , Rodica Comendant c , Dilfuza Kurbanbekova d , Beverly Winikoff a a
Gynuity Health Projects, New York, NY 10010, USA b Gynuity Health Projects, Tbilisi, Georgia c Moldova State Medical University, Reproductive Health Training Center, Chisinau, Moldova d Women’s Wellness Center, Tashkent, Uzbekistan Received 2 June 2014; revised 31 October 2014; accepted 4 November 2014
Abstract Objective: To evaluate the feasibility and acceptability of phone follow-up with a home semiquantitative pregnancy test and standardized checklist, and compare the alternative method of follow-up with in-clinic follow-up after medical abortion. Study Design: Two thousand four hundred women undergoing medical abortion with mifepristone and misoprostol in Moldova and Uzbekistan were randomized to phone or clinic follow-up. All women in the clinic group returned to the clinic 2 weeks later. Women randomized to phone follow-up used a semiquantitative pregnancy test at the initial visit and repeated the test at home 2 weeks later when they also filled out a symptom checklist. Women were called at 2 weeks to review the test results and checklist. Participants who screened “positive” were referred to clinic to verify abortion completion. Results: Most women in the phone group were successfully contacted on the phone (97.6%). Staff were unable to contact one woman in the phone follow-up group, and all women in clinic group returned to the clinic. The ongoing pregnancy rate was similar in both groups (0.4–0.6%), and the semiquantitative pregnancy test identified all ongoing pregnancies in the phone follow-up group. Women in the phone group found the test and checklist easy to use, and most (76.1%) preferred phone follow-up in the future. Overall, 92.8% of women in the phone group did not undergo inclinic follow-up. Conclusion: Phone follow-up with a semiquantitative urine pregnancy test and symptom checklist is a feasible and a highly effective approach in identifying ongoing pregnancy after medical abortion. Implications: The semiquantitative pregnancy test can make home follow-up after medical abortion possible for many women and provide reassurance that ongoing pregnancies will be detected. © 2015 Elsevier Inc. All rights reserved. Keywords: Mifepristone; Misoprostol; Human chorionic gonadotropin; Symptoms; Checklist
1. Introduction
☆
Funding: Financial support for this study was provided by an anonymous donor. ☆☆ Disclosure of interests: The authors have no potential conflicts of interest to disclose. ★ Clinical trials registration number: ClinicalTrials.gov, www. clinicaltrials.gov, NCT01410266. ⁎ Corresponding author. Gynuity Health Projects, 15 East 26th Street, Suite 801, New York, NY 10010, USA. Tel.: +1 212 448 1230; fax: +1 212 448 1260. E-mail address: [email protected] (I. Platais). http://dx.doi.org/10.1016/j.contraception.2014.11.004 0010-7824/© 2015 Elsevier Inc. All rights reserved.
Medical abortion with mifepristone and misoprostol is highly effective and safe [1], but protocols generally require a follow-up visit 1-2 weeks after mifepristone administration. The primary purpose of this visit is to confirm that the woman is no longer pregnant and determine if further care is required. Evidence suggests that the large majority of women undergoing medical abortion do not require additional treatment. Only 2-5% of women receive a vacuum aspiration for any reason, including ongoing pregnancy and retained
I. Platais et al. / Contraception 91 (2015) 178–183
products of conception, and even fewer women receive treatment for other reasons, such as infection or excessive or prolonged bleeding [2–4]. Serious complications other than ongoing pregnancy can be self-identified, and women undergoing medical abortion are routinely counseled on how to recognize complications and when to seek help. Despite medical abortion’s efficacy and safety, routine inclinic follow-up remains the standard of care in many guidelines. However, the follow-up visit may restrict the acceptability and accessibility of the method, as some women may be discouraged from choosing medical abortion because it is too difficult or expensive to return to the clinic. In clinical studies, women frequently cite the number of clinic visits as a reason for not choosing medical abortion [5–8]. If medical abortion is chosen, frequent non-compliance with the follow-up visit schedule can be expected. Studies have reported that 20–50% of women did not return for the follow-up visit [5,9–11]. The semiquantitative urine pregnancy test is a promising new tool in identifying ongoing pregnancy after medical abortion. Studies conducted in the United States and Vietnam [12–14] demonstrate that the semiquantitative pregnancy test effectively detects ongoing pregnancy as early as one week after medical abortion. In a large randomized clinical trial conducted by Ngoc et al. [14] in Vietnam, fewer women in the home semiquantitative pregnancy test group were lost to follow-up when compared to women in the clinic follow-up group. In all studies, women reported that the semiquantitative pregnancy test is easy to use [12–14]. Medical abortion with mifepristone and misoprostol has been available in Moldova and Uzbekistan since the mid2000s. Routine follow-up two weeks after mifepristone administration is the standard clinical practice in both countries. The goal of this study was to test the acceptability and feasibility of phone follow-up using a semiquantitative pregnancy test and a standardized symptom checklist in Moldova and Uzbekistan. To do so, this study assessed the response rate obtained with the standard in-clinic and phone follow-up protocols, documented the need for further treatment among women in the two study arms, and documented women’s experience using the semiquantitative pregnancy test and the symptom checklist.
2. Materials and methods The study was a prospective, non-blinded randomized controlled trial conducted at four clinics in Moldova (Municipal Clinical Hospital and the National Center for Reproductive Health and Medical Genetics in Chisinau; the Perinatalogy Center in Balti; and the Center of Women’s Health “Ana” in Drochia), and three clinics in Uzbekistan (Clinic No. 2 of Tashkent Medical Academy, Municipal Maternity House No. 10, and Women’s Wellness Center in Tashkent). The medical abortion regimen at all study sites was 200 mg mifepristone orally, followed by 400 micrograms
179
sublingual misoprostol administered in 24–48 h at home. The protocol was approved by the institutional review board at each facility in Moldova and by the ethical review board at the Ministry of Health of the Republic of Uzbekistan for all facilities in Uzbekistan. All participants gave written informed consent. Women with pregnancies ≤ 63 days’ gestation who desired medical abortion were eligible to participate in this study. Additional inclusion criteria were: eligible for medical abortion according to clinician and clinical standards; no known contraindications to mifepristone and/or misoprostol; general good health; ability to provide consent; and willingness to be followed up by phone or in clinic. Gestational age was determined using standard assessment practice at each site: menstrual history; clinical exam; and/or ultrasound. After assessing eligibility, participants were randomly assigned to one of two study groups: 1) clinic follow-up (clinic group); or 2) phone follow-up (phone group). Providers informed women of their computer-generated randomization assignment after opening sealed opaque envelopes in sequential order. Randomization was stratified by study site, using blocks of eight, and staff at Gynuity Health Projects in New York prepared and sealed the randomization envelopes. Participants randomized to the clinic group were asked to return to the clinic to determine their abortion status two weeks after mifepristone administration per existing standard of care. Providers assessed abortion status by clinical examination, women’s report of symptoms, and ultrasound, if needed. Participants with ongoing pregnancies or excessive bleeding underwent surgical intervention. Women with retained products of conception could choose between surgery and waiting one week with or without an additional dose of misoprostol. If the abortion was still incomplete after one week, surgical intervention was provided. Participants in the phone group were asked to complete the semiquantitative pregnancy test (dBest hCG Panel Test; Ameritek, Seattle, WA, USA) at the clinic to determine their baseline human chorionic gonadotropin (hCG) range. The test assigns hCG levels to one of the following five hCG ranges: 25–99 mIU/mL; 100–499 mIU/mL; 500–1999 mIU/mL; 2000–9999 mIU/mL; and greater than 10 000 mIU/mL. Providers recorded the baseline hCG range on study forms and scheduled a telephone call two weeks after mifepristone administration. Each participant was given a second semiquantitative pregnancy test and a symptom checklist to use at home prior to the phone call. The following checklist questions were derived from a previous study [15]: 1) “Did you experience no or only one day of heavy bleeding (bleeding greater than your normal menses)?”; 2) “Did you feel any of the following things today: breast tenderness, nausea or ‘morning sickness’, need to urinate frequently, exhaustion or tiredness?”; and 3) “Do you still ‘feel pregnant’ at this moment?” On the day of scheduled follow-up, providers called women in the phone group and asked them about the results of the semiquantitative pregnancy test and the symptom checklist, and their experiences in using the two modalities. Women were asked to return to the clinic for further
180
I. Platais et al. / Contraception 91 (2015) 178–183
evaluation if they responded “yes” to any of the checklist questions or if the semiquantitative pregnancy test indicated no change or an increase in the hCG range. Participants could also choose to return to the clinic at any time during the study. After abortion was determined to be complete and prior to discharge from the study, participants in both groups were asked about the acceptability of the follow-up procedure and their satisfaction with it, as well as their future preferences for method of follow-up. We powered this study to determine if phone follow-up with the semiquantitative pregnancy test and checklist is at least 80% as effective at detecting ongoing pregnancies as inclinic follow-up, which would diagnose 100% of ongoing pregnancies for women who return. We estimated (with α= 0.05, 1-β=0.8, with a one-tailed test) that at least 14 ongoing pregnancies should occur per group. Based on previous data from these facilities, we estimated that ongoing pregnancy would occur in about 1.2% cases of medical abortion [16,17]; therefore, we would be required to enroll approximately 1167 women in each group. Finally, to account for loss to follow-up, we rounded up to 1200 women in each group, with a total sample size of 2400 women. Data were entered using an automatic data capture system (TeleForm Desktop v10.2, Cardiff Software, Vista, CA, USA) and analyzed using IBM SPSS Statistics (version 19, Chicago, IL, USA). Fisher’s exact test and Mann–Whitney U test were used to compare characteristics of the two groups. The main study outcomes were compared using relative risk (RR) and 95% confidence intervals (CIs), and p-values. Twotailed pb.05 was considered statistically significant. We used intent-to-treat analysis for the primary outcome, or the detection of ongoing pregnancy. For cases with a completed phone follow-up, sensitivity, specificity, and negative and positive predictive values of the semiquantitative pregnancy test and checklist in detecting ongoing pregnancy were calculated with VassarStat online statistics software, using the Wilson method with correction for continuity [18].
3. Results From July 2010 to November 2012, 2400 women were enrolled in the study: 1200 were randomized to the clinic group, and 1200 were randomized to the phone group (Fig. 1). In the phone group, 4 (0.3%) participants underwent follow-up under the clinic follow-up protocol and returned to the clinic as instructed, and in the clinic group, 3 (0.3%) women underwent follow-up under the phone follow-up protocol. There was no statistically significant difference in the background characteristics between the two study groups (Table 1). The median age of participants was 27 years old, and most had completed nine or more years of education. The median gestational age was 43 days, and 23% of participants had at least one previous medical abortion.
Most women completed follow-up as scheduled, and very few women in both groups returned to the clinic outside of their scheduled follow-up (Fig. 1). In the phone group, 24 (2.0%) participants came to the clinic unprompted prior to their scheduled phone follow-up, and in the clinic group, 27 (2.3%) participants came to the clinic unprompted prior to their scheduled follow-up (p=.78, RR: 0.89, 95% CI=0.52– 1.53). Overall, more women in the phone group returned to the clinic before or after their scheduled follow-up (phone: 42 or 3.5%, clinic: 27 or 2.3%, p=.09, RR: 1.56, 95% CI= 0.97–2.51). However, in total, only 7.2% of all women (86/ 1199) in the phone group were instructed or returned to the clinic on their own, and 99.8% of women (1197/1200) in the clinic group underwent in-clinic follow-up. In the phone group, 1171 participants (97.6%) were contacted by phone on their scheduled follow-up day, and the majority (96.6% [1131/1171]) were told that they did not need to return to the clinic (Fig. 1). Forty (3.4%) participants screened positive by the semiquantitative pregnancy test or the symptom checklist, of whom 32 women returned to the clinic. Lastly, the status of one woman was unknown, as she could not be reached by phone. Of those who screened positive and returned, three women received uterine aspiration for ongoing pregnancy, and seven women received aspiration for heavy bleeding. Among 1131 women who did not screen positive by the semiquantitative pregnancy test or the symptom checklist, 18 (1.6%) returned to the clinic with concerns about their abortion outcome. Of these, two received uterine aspiration due to prolonged bleeding, and 2 women requested uterine aspiration. There were no statistically significant differences in the rates of complete abortion without surgical evacuation (phone group: 96.9% [1162/1199]; clinic group: 97.6% [1171/1200], Table 2). All eight women who were screened “positive” on the phone but did not return to the clinic had a negative semiquantitative pregnancy test but answered “yes” to one of the questions on the checklist. It is unknown if the women found follow-up unnecessary, and there are no records of these women seeking additional care for ongoing pregnancy. The ongoing pregnancy rate did not differ between the two groups in an adjusted analysis that excludes the eight cases (phone: 0.4%, clinic: 0.6%, p=.78, RR: 0.72, 95% CI=0.23–2.26). Per intent-to-treat analysis, the ongoing pregnancy rate was also similar between the two groups (1.1% vs 0.6%, p=.26, RR: 1.86, 95% CI=0.74–4.64). All women contacted by phone completed the checklist and the semiquantitative pregnancy test. The majority of participants in the phone group reported that the semiquantitative pregnancy test and the symptom checklist were very easy or easy to use (93.5% for pregnancy test, 95.2% for checklist, Table 3). During the phone follow-up, one woman reported a decrease in the hCG range but was unable to interpret the pregnancy test results, and study staff instructed her to return to the clinic. Another woman could not understand instructions for completion of the checklist, and
I. Platais et al. / Contraception 91 (2015) 178–183
181
Fig. 1. Patient flowchart.
provider guided the woman through the checklist during the phone call. When asked about their preference for future follow-up, most women in the phone group (76.1%) preferred phone follow-up, and a majority of women in the clinic group (73.4%) thought that they would be able to manage medical abortion at home if a checklist and a pregnancy test were provided to them. Among the participants with a completed phone followup, the sensitivity of the phone follow-up with the semiquantitative pregnancy test and checklist to detect ongoing pregnancy was 100% (95% CI=31.0–100%), and specificity was 97.5% (95% CI=96.4–98.3%). The negative predictive value was 100% (95% CI=99.6–100%), and positive predictive value was 9.4% (95% CI=2.5–26.2%).
4. Discussion This randomized clinical trial showed that phone follow-up using a semiquantitative urine pregnancy test and a standardized symptom checklist is a feasible alternative to clinic-based follow-up two weeks after medical abortion. The semiquantitative pregnancy test and the checklist identified women with ongoing pregnancy, and women found the semiquantitative pregnancy test and the checklist easy to use. Our study showed that 95.0% (1113/1171) of participants in the phone follow-up group could forgo routine in-clinic follow-up and were able to assess their abortion outcome at home with assurance that they do not need treatment for ongoing pregnancy. Participants reported that the semiquantitative pregnancy test and the
Table 1 Participant characteristics Phone follow-up (n= 1200) Age, years: median (range) Level of education: % (n) Between 1 and 8 years Between 9 and 12 years More than 12 years Owns cell phone or landline: % (n) Gravidity: median (range) Parity: median (range) Gestational age, days: median (range) Previous abortion: % (n) At least 1 previous surgical abortion: % (n) At least 1 previous medical abortion: % (n)
Clinic follow-up (n= 1200)
27 (16–49)
27 (16–47)
2.5 (30) 51.8 (622) 45.7 (548) 99.9 (1199) 3.0 (1–15) 1.0 (0–5) 43.0 (30–63) 56.5 (678) 43.0 (516) 24.1 (289)
2.5 (30) 52.1 (625) 45.4 (545) 99.8 (1198) 3.0 (1–20) 1.0 (0–5) 43.0 (28–63) 55.8 (670) 44.0 (528) 21.6 (259)
p-value 0.938 0.993
1.000 0.910 0.715 0.615 0.742 0.651 0.158
Women in the phone follow-up group were called 2 weeks after the initial visit to review the results of the semiquantitative pregnancy test and pregnancy symptom checklist. Women screened positive by the test or the checklist were asked to return to clinic. All women in the clinic follow-up group returned to clinic 2 weeks after the initial visit.
182
I. Platais et al. / Contraception 91 (2015) 178–183
Table 2 Medical abortion outcome, % (n)
Complete abortion without surgical evacuation Reason for surgical evacuation Ongoing pregnancy Detected during early unscheduled clinic visit Detected at scheduled follow-up Retained products of conception Heavy or prolonged bleeding Woman’s request
Phone follow-up (n=1191 a)
Clinic follow-up (n=1200)
Relative risk (95% CI)
p-value
97.6 (1162)
97.6 (1171)
1.00 (0.99–1.01)
N0.999
0.2 (2) 0.3 (3) 0.9 (11) 0.9 (11) 0.2 (2)
0.3 (3) 0.3 (4) 1.1 (13) 0.6 (7) 0.2 (2)
0.67 (0.11–4.01) 0.76 (0.17–3.37) 0.85 (0.38–1.90) 1.58 (0.62–4.07) 1.01 (0.14–7.14)
N0.999 N0.999 0.853 0.469 N0.999
Women in the phone follow-up group were called 2 weeks after the initial visit to review the results of the semiquantitative pregnancy test and pregnancy symptom checklist. Women screened positive by the test or the checklist were asked to return to clinic. All women in the clinic follow-up group returned to clinic 2 weeks after the initial visit. a Women who were lost to follow-up (n= 1) or did not return to clinic after phone call (n=8) were excluded from analysis.
checklist were easy to use, and the majority of the women in the phone group preferred phone follow-up in the future. In this study, the study staff were unable to reach one woman by telephone. However, eight women with a negative semiquantitative pregnancy test had an affirmative response to one of the symptom checklist items but did not present for further care. All eight responded “yes” to the following question: “Did you feel any of the following things today: breast tenderness, nausea or “morning sickness”, need to urinate frequently, exhaustion or
tiredness?” Previous studies show that the semiquantitative pregnancy test accurately detects ongoing pregnancy in 100% of cases, produces no false negatives [12,13], and that a symptom checklist is not highly predictive of abortion failure [8,14,19]. A study conducted by Ngoc et al. [14] concluded that when the semiquantitative pregnancy test is used, the symptom checklist did not offer additional value in identifying ongoing pregnancy. In our study, the symptom checklist signaled a potential ongoing pregnancy for almost three times as many women as the semiquantitative pregnancy test, but only a small
Table 3 Feasibility of phone follow-up and women’s preference for method of future follow-up, % (n)
Woman completed checklist Ease of completion of checklist Very easy or easy Neither easy or difficult Woman completed pregnancy test Ease of completion of pregnancy test Very easy or easy Neither easy or difficult Difficult Do you think that you could have managed your medical abortion on your own at home, if your provider gave you a simple list of questions to answer and a pregnancy test to take at home? Yes No Do not know Preference for future medical abortion follow up At the clinic By phone No preference
Phone follow-up
Clinic follow-up
n=1171
n=1188 a
100 (1171)
n/a
95.2 (1115) 4.8 (56) 100 (1171)
n/a n/a n/a
93.5 (1095) 6.3 (74) 0.2 (2)
n/a n/a n/a
n/a n/a n/a n=1199⁎ 9.6 (115) 76.1 (913) 14.3 (171)
73.4 (872) 7.7 (92) 18.9 (224) n=1199 b 29.1 (349) 48.1 (577) 22.8 (273)
Women in the phone follow-up group were called 2 weeks after the initial visit to review the results of the semiquantitative pregnancy test and pregnancy symptom checklist. Women screened positive by the test or the checklist were asked to return to clinic. All women in the clinic follow-up group returned to clinic 2 weeks after the initial visit. After abortion was determined to be complete, women in phone follow-up and clinic follow-up groups were asked about acceptability of and their satisfaction with their follow-up procedure. ⁎ 289 women in the phone group had a prior medical abortion. For these women, 80.6% (233) preferred phone follow-up, 10.4% (30) preferred clinic follow-up, and 9.0% (26) had no preference/did not know. a Did not respond to question (n=9). b Did not respond to question (n= 1).
I. Platais et al. / Contraception 91 (2015) 178–183
percentage of women received further treatment. Current research is exploring whether it is possible to provide the semiquantitative pregnancy test to women without any checklist and instruct them to contact the clinic if the results of the semiquantitative pregnancy test are positive or if they have bothersome symptoms. The 2012 World Health Organization (WHO) Safe Abortion Guidance recommends that routine follow-up should not be required after uncomplicated medical abortion using mifepristone and misoprostol if the woman is provided counseling about the procedure and given information on where to seek additional care [20]. Since the semiquantitative urine pregnancy test is highly effective in detecting ongoing pregnancies, it may provide additional reassurance to providers and women in identifying women who need to return to the clinic. The present study has several limitations. Our baseline assumptions for sample size included an ongoing pregnancy rate of 1.2% based on previous studies [16,17]; however, the ongoing pregnancy rate in this study was lower among women assigned to both phone and clinic follow-up, and this fact may have reduced our statistical power to detect any differences. In addition, our study did not require all women to return to the clinic to verify their abortion outcome. Nevertheless, both study arms had similar outcomes, including the number of ongoing pregnancies, providing reassurance that both methods identified ongoing pregnancies. Lastly, the majority of women in our sample had completed secondary school, and results may not be the same among women with lower levels of education. Phone follow-up with the semiquantitative urine pregnancy test is a feasible and acceptable method of follow-up after early medical abortion. This method could reduce clinic visits, thus reducing the burden on providers who would then have more time for other services. It is also a convenient and confidential method for women, enabling them to confirm abortion success at home. For women who have to travel long distances to clinics, arrange childcare, or take time off from household or work duties, the semiquantitative urine pregnancy test may be a less costly alternative to an in-clinic follow-up visit. Acknowledgment The authors thank all the women who participated in this study and the local investigators who provided care: Stelian Hodorogea, Dilbar Jymagylova, Feruza Kasimova, Petru Nedelciuc, Veacheslav Moshin, Svetlana Nicov, Irina Sagaidac, and Dilmurod Yusupov. We are grateful to Ms. Tatyana Lotarevich for her assistance with data cleaning and analysis. References [1] Raymond EG, Shannon C, Weaver MA, Winikoff B. First-trimester medical abortion with mifepristone 200 mg and misoprostol: a systematic review. Contraception 2013;87(1):26–37.
183
[2] Henderson JT, Hwang AC, Harper CC, Stewart FH. Safety of mifepristone abortions in clinical use. Contraception 2005;72(3):175–8. [3] Fjerstad M, Trussell J, Sivin I, Lichtenberg ES, Cullins V. Rates of serious infection after changes in regimens for medical abortion. N Engl J Med 2009;361(2):145–51. [4] Cleland K, Creinin MD, Nucatola D, Nshom M, Trussell J. Significant adverse events and outcomes after medical abortion. Obstet Gynecol 2013;121(1):166–71. [5] Clark WH, Gold M, Grossman D, Winikoff B. Can mifepristone medical abortion be simplified? A review of the evidence and questions for future research. Contraception 2007;75(4):245–50. [6] Winikoff B, Sivin I, Coyaji KJ, Cabezas E, Bilian X, Sujuan G, et al. The acceptability of medical abortion in China, Cuba and India. Int Fam Plan Perspect 1997;23(2):73–8 [89]. [7] Hamoda H, Templeton A. Medical and surgical options for induced abortion in first trimester. Best Pract Res Clin Obstet Gynaecol 2010;24(4):503–16. [8] Jackson AV, Dayananda I, Fortin JM, Fitzmaurice G, Goldberg AB. Can women accurately assess the outcome of medical abortion based on symptoms alone? Contraception 2012;85(2):192–7. [9] Astle H, Cameron ST, Johnstone A. Comparison of unscheduled reattendance and contraception at discharge, among women having the final stage of early medical abortion at home and those remaining in hospital. J Fam Plann Reprod Health Care 2012;38(1):35–40. [10] Haimov-Kochman R, Arbel R, Sciaky-Tamir Y, Brzezinski A, Laufer N, Yagel S. Risk factors for unsuccessful medical abortion with mifepristone and misoprostol. Acta Obstet Gynecol Scand 2007;86(4):462–6. [11] Horning EL, Chen BA, Meyn LA, Creinin MD. Comparison of medical abortion follow-up with serum human chorionic gonadotropin testing and in-office assessment. Contraception 2012;85(4):402–7. [12] Blum J, Shochet T, Lynd K, Lichtenberg ES, Fischer D, Arnesen M, et al. Can at-home semi-quantitative pregnancy tests serve as a replacement for clinical follow-up of medical abortion? A US study. Contraception 2012;86(6):757–62. [13] Lynd K, Blum J, Ngoc NT, Shochet T, Blumenthal PD, Winikoff B. Simplified medical abortion using a semi-quantitative pregnancy test for home-based follow-up. Int J Gynecol Obstet 2013;121 (2):144–8. [14] Ngoc NT, Bracken H, Blum J, Nga NT, Minh NH, van Nhang N, et al. Acceptability and feasibility of phone follow-up after early medical abortion in Vietnam: a randomized controlled trial. Obstet Gynecol 2014;123(1):88–95. [15] Clark W, Bracken H, Tanenhaus J, Schweikert S, Lichtenberg ES, Winikoff B. Alternatives to a routine follow-up visit for early medical abortion. Obstet Gynecol 2010;115(2 Pt 1):264–72. [16] Raghavan S, Comendant R, Digol I, Ungureanu S, Dondiuc I, Turcanu S, et al. Comparison of 400 mcg buccal and 400 mcg sublingual misoprostol after mifepristone medical abortion through 63 days’ LMP: a randomized controlled trial. Contraception 2010;82 (6):513–9. [17] Raghavan S, Tsereteli T, Kamilov A, Kurbanbekova D, Yusupov D, Kasimova F, et al. Acceptability and feasibility of the use of 400 mcg of sublingual misoprostol after mifepristone for medical abortion up to 63 days since the last menstrual period: evidence from Uzbekistan. Eur J Contracept Reprod Health Care 2013;18(2):104–11. [18] VassarStats: Website for Statistical Computation [Internet]. Clinical Calculator 1. c1998-2014 [cited May 2014]. Available from: http://www.vassarstats.net/clin1.html. [19] Bracken H, Lohr PA, Taylor J, Morroni C, Winikoff B. RU OK? The acceptability and feasibility of remote technologies for follow-up after early medical abortion. Contraception 2014;90(1):29–35. [20] Safe abortion: technical and policy guidance for health systems. Geneva (Switzerland): Word Health Organization; 2012.
Original research article
Acceptability and feasibility of phone follow-up with a semiquantitative urine pregnancy test after medical abortion in Moldova and Uzbekistan☆,☆☆,★ Ingrida Platais a,⁎, Tamar Tsereteli b , Rodica Comendant c , Dilfuza Kurbanbekova d , Beverly Winikoff a a
Gynuity Health Projects, New York, NY 10010, USA b Gynuity Health Projects, Tbilisi, Georgia c Moldova State Medical University, Reproductive Health Training Center, Chisinau, Moldova d Women’s Wellness Center, Tashkent, Uzbekistan Received 2 June 2014; revised 31 October 2014; accepted 4 November 2014
Abstract Objective: To evaluate the feasibility and acceptability of phone follow-up with a home semiquantitative pregnancy test and standardized checklist, and compare the alternative method of follow-up with in-clinic follow-up after medical abortion. Study Design: Two thousand four hundred women undergoing medical abortion with mifepristone and misoprostol in Moldova and Uzbekistan were randomized to phone or clinic follow-up. All women in the clinic group returned to the clinic 2 weeks later. Women randomized to phone follow-up used a semiquantitative pregnancy test at the initial visit and repeated the test at home 2 weeks later when they also filled out a symptom checklist. Women were called at 2 weeks to review the test results and checklist. Participants who screened “positive” were referred to clinic to verify abortion completion. Results: Most women in the phone group were successfully contacted on the phone (97.6%). Staff were unable to contact one woman in the phone follow-up group, and all women in clinic group returned to the clinic. The ongoing pregnancy rate was similar in both groups (0.4–0.6%), and the semiquantitative pregnancy test identified all ongoing pregnancies in the phone follow-up group. Women in the phone group found the test and checklist easy to use, and most (76.1%) preferred phone follow-up in the future. Overall, 92.8% of women in the phone group did not undergo inclinic follow-up. Conclusion: Phone follow-up with a semiquantitative urine pregnancy test and symptom checklist is a feasible and a highly effective approach in identifying ongoing pregnancy after medical abortion. Implications: The semiquantitative pregnancy test can make home follow-up after medical abortion possible for many women and provide reassurance that ongoing pregnancies will be detected. © 2015 Elsevier Inc. All rights reserved. Keywords: Mifepristone; Misoprostol; Human chorionic gonadotropin; Symptoms; Checklist
1. Introduction
☆
Funding: Financial support for this study was provided by an anonymous donor. ☆☆ Disclosure of interests: The authors have no potential conflicts of interest to disclose. ★ Clinical trials registration number: ClinicalTrials.gov, www. clinicaltrials.gov, NCT01410266. ⁎ Corresponding author. Gynuity Health Projects, 15 East 26th Street, Suite 801, New York, NY 10010, USA. Tel.: +1 212 448 1230; fax: +1 212 448 1260. E-mail address: [email protected] (I. Platais). http://dx.doi.org/10.1016/j.contraception.2014.11.004 0010-7824/© 2015 Elsevier Inc. All rights reserved.
Medical abortion with mifepristone and misoprostol is highly effective and safe [1], but protocols generally require a follow-up visit 1-2 weeks after mifepristone administration. The primary purpose of this visit is to confirm that the woman is no longer pregnant and determine if further care is required. Evidence suggests that the large majority of women undergoing medical abortion do not require additional treatment. Only 2-5% of women receive a vacuum aspiration for any reason, including ongoing pregnancy and retained
I. Platais et al. / Contraception 91 (2015) 178–183
products of conception, and even fewer women receive treatment for other reasons, such as infection or excessive or prolonged bleeding [2–4]. Serious complications other than ongoing pregnancy can be self-identified, and women undergoing medical abortion are routinely counseled on how to recognize complications and when to seek help. Despite medical abortion’s efficacy and safety, routine inclinic follow-up remains the standard of care in many guidelines. However, the follow-up visit may restrict the acceptability and accessibility of the method, as some women may be discouraged from choosing medical abortion because it is too difficult or expensive to return to the clinic. In clinical studies, women frequently cite the number of clinic visits as a reason for not choosing medical abortion [5–8]. If medical abortion is chosen, frequent non-compliance with the follow-up visit schedule can be expected. Studies have reported that 20–50% of women did not return for the follow-up visit [5,9–11]. The semiquantitative urine pregnancy test is a promising new tool in identifying ongoing pregnancy after medical abortion. Studies conducted in the United States and Vietnam [12–14] demonstrate that the semiquantitative pregnancy test effectively detects ongoing pregnancy as early as one week after medical abortion. In a large randomized clinical trial conducted by Ngoc et al. [14] in Vietnam, fewer women in the home semiquantitative pregnancy test group were lost to follow-up when compared to women in the clinic follow-up group. In all studies, women reported that the semiquantitative pregnancy test is easy to use [12–14]. Medical abortion with mifepristone and misoprostol has been available in Moldova and Uzbekistan since the mid2000s. Routine follow-up two weeks after mifepristone administration is the standard clinical practice in both countries. The goal of this study was to test the acceptability and feasibility of phone follow-up using a semiquantitative pregnancy test and a standardized symptom checklist in Moldova and Uzbekistan. To do so, this study assessed the response rate obtained with the standard in-clinic and phone follow-up protocols, documented the need for further treatment among women in the two study arms, and documented women’s experience using the semiquantitative pregnancy test and the symptom checklist.
2. Materials and methods The study was a prospective, non-blinded randomized controlled trial conducted at four clinics in Moldova (Municipal Clinical Hospital and the National Center for Reproductive Health and Medical Genetics in Chisinau; the Perinatalogy Center in Balti; and the Center of Women’s Health “Ana” in Drochia), and three clinics in Uzbekistan (Clinic No. 2 of Tashkent Medical Academy, Municipal Maternity House No. 10, and Women’s Wellness Center in Tashkent). The medical abortion regimen at all study sites was 200 mg mifepristone orally, followed by 400 micrograms
179
sublingual misoprostol administered in 24–48 h at home. The protocol was approved by the institutional review board at each facility in Moldova and by the ethical review board at the Ministry of Health of the Republic of Uzbekistan for all facilities in Uzbekistan. All participants gave written informed consent. Women with pregnancies ≤ 63 days’ gestation who desired medical abortion were eligible to participate in this study. Additional inclusion criteria were: eligible for medical abortion according to clinician and clinical standards; no known contraindications to mifepristone and/or misoprostol; general good health; ability to provide consent; and willingness to be followed up by phone or in clinic. Gestational age was determined using standard assessment practice at each site: menstrual history; clinical exam; and/or ultrasound. After assessing eligibility, participants were randomly assigned to one of two study groups: 1) clinic follow-up (clinic group); or 2) phone follow-up (phone group). Providers informed women of their computer-generated randomization assignment after opening sealed opaque envelopes in sequential order. Randomization was stratified by study site, using blocks of eight, and staff at Gynuity Health Projects in New York prepared and sealed the randomization envelopes. Participants randomized to the clinic group were asked to return to the clinic to determine their abortion status two weeks after mifepristone administration per existing standard of care. Providers assessed abortion status by clinical examination, women’s report of symptoms, and ultrasound, if needed. Participants with ongoing pregnancies or excessive bleeding underwent surgical intervention. Women with retained products of conception could choose between surgery and waiting one week with or without an additional dose of misoprostol. If the abortion was still incomplete after one week, surgical intervention was provided. Participants in the phone group were asked to complete the semiquantitative pregnancy test (dBest hCG Panel Test; Ameritek, Seattle, WA, USA) at the clinic to determine their baseline human chorionic gonadotropin (hCG) range. The test assigns hCG levels to one of the following five hCG ranges: 25–99 mIU/mL; 100–499 mIU/mL; 500–1999 mIU/mL; 2000–9999 mIU/mL; and greater than 10 000 mIU/mL. Providers recorded the baseline hCG range on study forms and scheduled a telephone call two weeks after mifepristone administration. Each participant was given a second semiquantitative pregnancy test and a symptom checklist to use at home prior to the phone call. The following checklist questions were derived from a previous study [15]: 1) “Did you experience no or only one day of heavy bleeding (bleeding greater than your normal menses)?”; 2) “Did you feel any of the following things today: breast tenderness, nausea or ‘morning sickness’, need to urinate frequently, exhaustion or tiredness?”; and 3) “Do you still ‘feel pregnant’ at this moment?” On the day of scheduled follow-up, providers called women in the phone group and asked them about the results of the semiquantitative pregnancy test and the symptom checklist, and their experiences in using the two modalities. Women were asked to return to the clinic for further
180
I. Platais et al. / Contraception 91 (2015) 178–183
evaluation if they responded “yes” to any of the checklist questions or if the semiquantitative pregnancy test indicated no change or an increase in the hCG range. Participants could also choose to return to the clinic at any time during the study. After abortion was determined to be complete and prior to discharge from the study, participants in both groups were asked about the acceptability of the follow-up procedure and their satisfaction with it, as well as their future preferences for method of follow-up. We powered this study to determine if phone follow-up with the semiquantitative pregnancy test and checklist is at least 80% as effective at detecting ongoing pregnancies as inclinic follow-up, which would diagnose 100% of ongoing pregnancies for women who return. We estimated (with α= 0.05, 1-β=0.8, with a one-tailed test) that at least 14 ongoing pregnancies should occur per group. Based on previous data from these facilities, we estimated that ongoing pregnancy would occur in about 1.2% cases of medical abortion [16,17]; therefore, we would be required to enroll approximately 1167 women in each group. Finally, to account for loss to follow-up, we rounded up to 1200 women in each group, with a total sample size of 2400 women. Data were entered using an automatic data capture system (TeleForm Desktop v10.2, Cardiff Software, Vista, CA, USA) and analyzed using IBM SPSS Statistics (version 19, Chicago, IL, USA). Fisher’s exact test and Mann–Whitney U test were used to compare characteristics of the two groups. The main study outcomes were compared using relative risk (RR) and 95% confidence intervals (CIs), and p-values. Twotailed pb.05 was considered statistically significant. We used intent-to-treat analysis for the primary outcome, or the detection of ongoing pregnancy. For cases with a completed phone follow-up, sensitivity, specificity, and negative and positive predictive values of the semiquantitative pregnancy test and checklist in detecting ongoing pregnancy were calculated with VassarStat online statistics software, using the Wilson method with correction for continuity [18].
3. Results From July 2010 to November 2012, 2400 women were enrolled in the study: 1200 were randomized to the clinic group, and 1200 were randomized to the phone group (Fig. 1). In the phone group, 4 (0.3%) participants underwent follow-up under the clinic follow-up protocol and returned to the clinic as instructed, and in the clinic group, 3 (0.3%) women underwent follow-up under the phone follow-up protocol. There was no statistically significant difference in the background characteristics between the two study groups (Table 1). The median age of participants was 27 years old, and most had completed nine or more years of education. The median gestational age was 43 days, and 23% of participants had at least one previous medical abortion.
Most women completed follow-up as scheduled, and very few women in both groups returned to the clinic outside of their scheduled follow-up (Fig. 1). In the phone group, 24 (2.0%) participants came to the clinic unprompted prior to their scheduled phone follow-up, and in the clinic group, 27 (2.3%) participants came to the clinic unprompted prior to their scheduled follow-up (p=.78, RR: 0.89, 95% CI=0.52– 1.53). Overall, more women in the phone group returned to the clinic before or after their scheduled follow-up (phone: 42 or 3.5%, clinic: 27 or 2.3%, p=.09, RR: 1.56, 95% CI= 0.97–2.51). However, in total, only 7.2% of all women (86/ 1199) in the phone group were instructed or returned to the clinic on their own, and 99.8% of women (1197/1200) in the clinic group underwent in-clinic follow-up. In the phone group, 1171 participants (97.6%) were contacted by phone on their scheduled follow-up day, and the majority (96.6% [1131/1171]) were told that they did not need to return to the clinic (Fig. 1). Forty (3.4%) participants screened positive by the semiquantitative pregnancy test or the symptom checklist, of whom 32 women returned to the clinic. Lastly, the status of one woman was unknown, as she could not be reached by phone. Of those who screened positive and returned, three women received uterine aspiration for ongoing pregnancy, and seven women received aspiration for heavy bleeding. Among 1131 women who did not screen positive by the semiquantitative pregnancy test or the symptom checklist, 18 (1.6%) returned to the clinic with concerns about their abortion outcome. Of these, two received uterine aspiration due to prolonged bleeding, and 2 women requested uterine aspiration. There were no statistically significant differences in the rates of complete abortion without surgical evacuation (phone group: 96.9% [1162/1199]; clinic group: 97.6% [1171/1200], Table 2). All eight women who were screened “positive” on the phone but did not return to the clinic had a negative semiquantitative pregnancy test but answered “yes” to one of the questions on the checklist. It is unknown if the women found follow-up unnecessary, and there are no records of these women seeking additional care for ongoing pregnancy. The ongoing pregnancy rate did not differ between the two groups in an adjusted analysis that excludes the eight cases (phone: 0.4%, clinic: 0.6%, p=.78, RR: 0.72, 95% CI=0.23–2.26). Per intent-to-treat analysis, the ongoing pregnancy rate was also similar between the two groups (1.1% vs 0.6%, p=.26, RR: 1.86, 95% CI=0.74–4.64). All women contacted by phone completed the checklist and the semiquantitative pregnancy test. The majority of participants in the phone group reported that the semiquantitative pregnancy test and the symptom checklist were very easy or easy to use (93.5% for pregnancy test, 95.2% for checklist, Table 3). During the phone follow-up, one woman reported a decrease in the hCG range but was unable to interpret the pregnancy test results, and study staff instructed her to return to the clinic. Another woman could not understand instructions for completion of the checklist, and
I. Platais et al. / Contraception 91 (2015) 178–183
181
Fig. 1. Patient flowchart.
provider guided the woman through the checklist during the phone call. When asked about their preference for future follow-up, most women in the phone group (76.1%) preferred phone follow-up, and a majority of women in the clinic group (73.4%) thought that they would be able to manage medical abortion at home if a checklist and a pregnancy test were provided to them. Among the participants with a completed phone followup, the sensitivity of the phone follow-up with the semiquantitative pregnancy test and checklist to detect ongoing pregnancy was 100% (95% CI=31.0–100%), and specificity was 97.5% (95% CI=96.4–98.3%). The negative predictive value was 100% (95% CI=99.6–100%), and positive predictive value was 9.4% (95% CI=2.5–26.2%).
4. Discussion This randomized clinical trial showed that phone follow-up using a semiquantitative urine pregnancy test and a standardized symptom checklist is a feasible alternative to clinic-based follow-up two weeks after medical abortion. The semiquantitative pregnancy test and the checklist identified women with ongoing pregnancy, and women found the semiquantitative pregnancy test and the checklist easy to use. Our study showed that 95.0% (1113/1171) of participants in the phone follow-up group could forgo routine in-clinic follow-up and were able to assess their abortion outcome at home with assurance that they do not need treatment for ongoing pregnancy. Participants reported that the semiquantitative pregnancy test and the
Table 1 Participant characteristics Phone follow-up (n= 1200) Age, years: median (range) Level of education: % (n) Between 1 and 8 years Between 9 and 12 years More than 12 years Owns cell phone or landline: % (n) Gravidity: median (range) Parity: median (range) Gestational age, days: median (range) Previous abortion: % (n) At least 1 previous surgical abortion: % (n) At least 1 previous medical abortion: % (n)
Clinic follow-up (n= 1200)
27 (16–49)
27 (16–47)
2.5 (30) 51.8 (622) 45.7 (548) 99.9 (1199) 3.0 (1–15) 1.0 (0–5) 43.0 (30–63) 56.5 (678) 43.0 (516) 24.1 (289)
2.5 (30) 52.1 (625) 45.4 (545) 99.8 (1198) 3.0 (1–20) 1.0 (0–5) 43.0 (28–63) 55.8 (670) 44.0 (528) 21.6 (259)
p-value 0.938 0.993
1.000 0.910 0.715 0.615 0.742 0.651 0.158
Women in the phone follow-up group were called 2 weeks after the initial visit to review the results of the semiquantitative pregnancy test and pregnancy symptom checklist. Women screened positive by the test or the checklist were asked to return to clinic. All women in the clinic follow-up group returned to clinic 2 weeks after the initial visit.
182
I. Platais et al. / Contraception 91 (2015) 178–183
Table 2 Medical abortion outcome, % (n)
Complete abortion without surgical evacuation Reason for surgical evacuation Ongoing pregnancy Detected during early unscheduled clinic visit Detected at scheduled follow-up Retained products of conception Heavy or prolonged bleeding Woman’s request
Phone follow-up (n=1191 a)
Clinic follow-up (n=1200)
Relative risk (95% CI)
p-value
97.6 (1162)
97.6 (1171)
1.00 (0.99–1.01)
N0.999
0.2 (2) 0.3 (3) 0.9 (11) 0.9 (11) 0.2 (2)
0.3 (3) 0.3 (4) 1.1 (13) 0.6 (7) 0.2 (2)
0.67 (0.11–4.01) 0.76 (0.17–3.37) 0.85 (0.38–1.90) 1.58 (0.62–4.07) 1.01 (0.14–7.14)
N0.999 N0.999 0.853 0.469 N0.999
Women in the phone follow-up group were called 2 weeks after the initial visit to review the results of the semiquantitative pregnancy test and pregnancy symptom checklist. Women screened positive by the test or the checklist were asked to return to clinic. All women in the clinic follow-up group returned to clinic 2 weeks after the initial visit. a Women who were lost to follow-up (n= 1) or did not return to clinic after phone call (n=8) were excluded from analysis.
checklist were easy to use, and the majority of the women in the phone group preferred phone follow-up in the future. In this study, the study staff were unable to reach one woman by telephone. However, eight women with a negative semiquantitative pregnancy test had an affirmative response to one of the symptom checklist items but did not present for further care. All eight responded “yes” to the following question: “Did you feel any of the following things today: breast tenderness, nausea or “morning sickness”, need to urinate frequently, exhaustion or
tiredness?” Previous studies show that the semiquantitative pregnancy test accurately detects ongoing pregnancy in 100% of cases, produces no false negatives [12,13], and that a symptom checklist is not highly predictive of abortion failure [8,14,19]. A study conducted by Ngoc et al. [14] concluded that when the semiquantitative pregnancy test is used, the symptom checklist did not offer additional value in identifying ongoing pregnancy. In our study, the symptom checklist signaled a potential ongoing pregnancy for almost three times as many women as the semiquantitative pregnancy test, but only a small
Table 3 Feasibility of phone follow-up and women’s preference for method of future follow-up, % (n)
Woman completed checklist Ease of completion of checklist Very easy or easy Neither easy or difficult Woman completed pregnancy test Ease of completion of pregnancy test Very easy or easy Neither easy or difficult Difficult Do you think that you could have managed your medical abortion on your own at home, if your provider gave you a simple list of questions to answer and a pregnancy test to take at home? Yes No Do not know Preference for future medical abortion follow up At the clinic By phone No preference
Phone follow-up
Clinic follow-up
n=1171
n=1188 a
100 (1171)
n/a
95.2 (1115) 4.8 (56) 100 (1171)
n/a n/a n/a
93.5 (1095) 6.3 (74) 0.2 (2)
n/a n/a n/a
n/a n/a n/a n=1199⁎ 9.6 (115) 76.1 (913) 14.3 (171)
73.4 (872) 7.7 (92) 18.9 (224) n=1199 b 29.1 (349) 48.1 (577) 22.8 (273)
Women in the phone follow-up group were called 2 weeks after the initial visit to review the results of the semiquantitative pregnancy test and pregnancy symptom checklist. Women screened positive by the test or the checklist were asked to return to clinic. All women in the clinic follow-up group returned to clinic 2 weeks after the initial visit. After abortion was determined to be complete, women in phone follow-up and clinic follow-up groups were asked about acceptability of and their satisfaction with their follow-up procedure. ⁎ 289 women in the phone group had a prior medical abortion. For these women, 80.6% (233) preferred phone follow-up, 10.4% (30) preferred clinic follow-up, and 9.0% (26) had no preference/did not know. a Did not respond to question (n=9). b Did not respond to question (n= 1).
I. Platais et al. / Contraception 91 (2015) 178–183
percentage of women received further treatment. Current research is exploring whether it is possible to provide the semiquantitative pregnancy test to women without any checklist and instruct them to contact the clinic if the results of the semiquantitative pregnancy test are positive or if they have bothersome symptoms. The 2012 World Health Organization (WHO) Safe Abortion Guidance recommends that routine follow-up should not be required after uncomplicated medical abortion using mifepristone and misoprostol if the woman is provided counseling about the procedure and given information on where to seek additional care [20]. Since the semiquantitative urine pregnancy test is highly effective in detecting ongoing pregnancies, it may provide additional reassurance to providers and women in identifying women who need to return to the clinic. The present study has several limitations. Our baseline assumptions for sample size included an ongoing pregnancy rate of 1.2% based on previous studies [16,17]; however, the ongoing pregnancy rate in this study was lower among women assigned to both phone and clinic follow-up, and this fact may have reduced our statistical power to detect any differences. In addition, our study did not require all women to return to the clinic to verify their abortion outcome. Nevertheless, both study arms had similar outcomes, including the number of ongoing pregnancies, providing reassurance that both methods identified ongoing pregnancies. Lastly, the majority of women in our sample had completed secondary school, and results may not be the same among women with lower levels of education. Phone follow-up with the semiquantitative urine pregnancy test is a feasible and acceptable method of follow-up after early medical abortion. This method could reduce clinic visits, thus reducing the burden on providers who would then have more time for other services. It is also a convenient and confidential method for women, enabling them to confirm abortion success at home. For women who have to travel long distances to clinics, arrange childcare, or take time off from household or work duties, the semiquantitative urine pregnancy test may be a less costly alternative to an in-clinic follow-up visit. Acknowledgment The authors thank all the women who participated in this study and the local investigators who provided care: Stelian Hodorogea, Dilbar Jymagylova, Feruza Kasimova, Petru Nedelciuc, Veacheslav Moshin, Svetlana Nicov, Irina Sagaidac, and Dilmurod Yusupov. We are grateful to Ms. Tatyana Lotarevich for her assistance with data cleaning and analysis. References [1] Raymond EG, Shannon C, Weaver MA, Winikoff B. First-trimester medical abortion with mifepristone 200 mg and misoprostol: a systematic review. Contraception 2013;87(1):26–37.
183
[2] Henderson JT, Hwang AC, Harper CC, Stewart FH. Safety of mifepristone abortions in clinical use. Contraception 2005;72(3):175–8. [3] Fjerstad M, Trussell J, Sivin I, Lichtenberg ES, Cullins V. Rates of serious infection after changes in regimens for medical abortion. N Engl J Med 2009;361(2):145–51. [4] Cleland K, Creinin MD, Nucatola D, Nshom M, Trussell J. Significant adverse events and outcomes after medical abortion. Obstet Gynecol 2013;121(1):166–71. [5] Clark WH, Gold M, Grossman D, Winikoff B. Can mifepristone medical abortion be simplified? A review of the evidence and questions for future research. Contraception 2007;75(4):245–50. [6] Winikoff B, Sivin I, Coyaji KJ, Cabezas E, Bilian X, Sujuan G, et al. The acceptability of medical abortion in China, Cuba and India. Int Fam Plan Perspect 1997;23(2):73–8 [89]. [7] Hamoda H, Templeton A. Medical and surgical options for induced abortion in first trimester. Best Pract Res Clin Obstet Gynaecol 2010;24(4):503–16. [8] Jackson AV, Dayananda I, Fortin JM, Fitzmaurice G, Goldberg AB. Can women accurately assess the outcome of medical abortion based on symptoms alone? Contraception 2012;85(2):192–7. [9] Astle H, Cameron ST, Johnstone A. Comparison of unscheduled reattendance and contraception at discharge, among women having the final stage of early medical abortion at home and those remaining in hospital. J Fam Plann Reprod Health Care 2012;38(1):35–40. [10] Haimov-Kochman R, Arbel R, Sciaky-Tamir Y, Brzezinski A, Laufer N, Yagel S. Risk factors for unsuccessful medical abortion with mifepristone and misoprostol. Acta Obstet Gynecol Scand 2007;86(4):462–6. [11] Horning EL, Chen BA, Meyn LA, Creinin MD. Comparison of medical abortion follow-up with serum human chorionic gonadotropin testing and in-office assessment. Contraception 2012;85(4):402–7. [12] Blum J, Shochet T, Lynd K, Lichtenberg ES, Fischer D, Arnesen M, et al. Can at-home semi-quantitative pregnancy tests serve as a replacement for clinical follow-up of medical abortion? A US study. Contraception 2012;86(6):757–62. [13] Lynd K, Blum J, Ngoc NT, Shochet T, Blumenthal PD, Winikoff B. Simplified medical abortion using a semi-quantitative pregnancy test for home-based follow-up. Int J Gynecol Obstet 2013;121 (2):144–8. [14] Ngoc NT, Bracken H, Blum J, Nga NT, Minh NH, van Nhang N, et al. Acceptability and feasibility of phone follow-up after early medical abortion in Vietnam: a randomized controlled trial. Obstet Gynecol 2014;123(1):88–95. [15] Clark W, Bracken H, Tanenhaus J, Schweikert S, Lichtenberg ES, Winikoff B. Alternatives to a routine follow-up visit for early medical abortion. Obstet Gynecol 2010;115(2 Pt 1):264–72. [16] Raghavan S, Comendant R, Digol I, Ungureanu S, Dondiuc I, Turcanu S, et al. Comparison of 400 mcg buccal and 400 mcg sublingual misoprostol after mifepristone medical abortion through 63 days’ LMP: a randomized controlled trial. Contraception 2010;82 (6):513–9. [17] Raghavan S, Tsereteli T, Kamilov A, Kurbanbekova D, Yusupov D, Kasimova F, et al. Acceptability and feasibility of the use of 400 mcg of sublingual misoprostol after mifepristone for medical abortion up to 63 days since the last menstrual period: evidence from Uzbekistan. Eur J Contracept Reprod Health Care 2013;18(2):104–11. [18] VassarStats: Website for Statistical Computation [Internet]. Clinical Calculator 1. c1998-2014 [cited May 2014]. Available from: http://www.vassarstats.net/clin1.html. [19] Bracken H, Lohr PA, Taylor J, Morroni C, Winikoff B. RU OK? The acceptability and feasibility of remote technologies for follow-up after early medical abortion. Contraception 2014;90(1):29–35. [20] Safe abortion: technical and policy guidance for health systems. Geneva (Switzerland): Word Health Organization; 2012.
