Medication Abortion Safety and Effectiveness With Misoprostol Alone

Key Points Question What is the effectiveness of self-managed medication abortion using misoprostol alone? Findings In this cohort study of 637 callers to safe abortion hotlines and accompaniment groups, almost all (98.1%) had a complete abortion using misoprostol without procedural intervention. Meaning These findings suggest that the effectiveness of misoprostol alone for abortion may be higher than previously thought, and exploration of expanding access via this regimen is warranted in both clinical and nonclinical contexts.


Introduction
There are 2 medication regimens recommended by the World Health Organization for abortion: mifepristone in combination with misoprostol and misoprostol alone. 1Both regimens are highly effective, with low rates of complications. 2,3Clinical evidence has suggested that misoprostol alone is less effective than the combined regimen (approximately 80% vs approximately 95%), 2,3 and many clinical guidelines recommend misoprostol alone only if mifepristone is unavailable. 4,5As a result, the combined regimen has long been the preferred method for medication abortion in countries where mifepristone is registered as a pharmaceutical product.
In settings where mifepristone is inaccessible, most medication abortions are performed using misoprostol alone. 6Misoprostol has many advantages, including low cost, a wide range of indications, availability in many settings without a prescription, ease of administration, and shelf stability. 7As a result, misoprostol for self-managed abortion (defined as when a person does something to end their own pregnancy without clinical supervision) has risen globally. 8Many models of care exist for providing information and support to people who cannot or do not want to access clinical care, such as community-based distribution, 9,10 feminist accompaniment and safe abortion hotline models, 11,12 and online telemedicine services. 13e aim of much of the clinical literature on misoprostol alone was to establish recommended dosing and administration routes; as a result, clinical data on the effectiveness of currently endorsed misoprostol-alone protocols are sparse. 2,14There are only 2 clinical trials on misoprostol-alone regimens that use the currently endorsed regimen: 800-μg misoprostol tablets administered sublingually, vaginally, or buccally every 3 hours for at least 3 doses (2400 μg total). 4,15In a 2007 study, 84% of 1025 participants with pregnancy duration less than 9 weeks randomized to an arm following this regimen had a complete abortion without procedural intervention at 2-week follow-up. 16In a 2019 study, 93% of 388 participants with pregnancy duration less than 10 weeks randomized to an arm following this regimen had a complete abortion at 1-week follow-up. 17While there are a growing number of prospective studies of self-managed misoprostol-alone use, 9,10,[18][19][20] only 2 documented the currently endorsed misoprostol-alone regimen.In a pilot study from Nigeria and Argentina of 94 participants with pregnancy durations less than 17 weeks (86% had pregnancy durations <10 weeks), 94% had a complete abortion at 3 week follow-up. 19A study of 568 participants in the US with pregnancy durations less than 10 weeks found effectiveness to be 87% at 4 week follow-up. 20Additional data with a shorter evaluation time and information on physical experience and health care seeking are needed from self-managed contexts.
Additional research is needed on the currently endorsed misoprostol-alone regimen, particularly given the lack of detailed information about the physical experience of using misoprostol alone. 6The aim of this study was to estimate the safety and effectiveness of misoprostol alone for abortion from a sample of callers to safe abortion hotlines and accompaniment groups and to document participants' physical experiences and health care-seeking experiences.Findings from this study may inform counseling and support for out-of-clinic models, as well as clinical policy and practice, to expand the range of practitioners offering medication abortion.

Methods
Data are reported from the Studying Accompaniment Feasibility and Effectiveness Study, a prospective observational cohort study of callers to safe abortion hotlines and accompaniment groups in Argentina, Nigeria, and a country in Southeast Asia (anonymized).Full protocol details for Argentina and Nigeria, 21 findings from a pilot study, 19 and main outcomes (noninferiority of selfmanaged abortion of pregnancy duration <9 weeks vs historical clinical controls) 18 have been published previously.Procedures in Southeast Asia followed the same protocol as detailed in the protocol article. 21This study was approved by the Allendale Investigational Review Board; the Fundación Huésped institutional review board approved the Argentina-specific protocol.An independent data monitoring and oversight committee reviewed the study protocols and instruments and a planned interim analysis of safety events.Informed consent was obtained verbally.

JAMA Network Open | Obstetrics and Gynecology
The study was prospectively registered with the ISRCTN Registry. 22We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines for observational studies and Medical Abortion Reporting of Efficacy (MARE) guideline for reporting efficacy from medication abortion studies.

Participant Recruitment
Accompaniment groups that served as recruitment sites all provide detailed counseling on abortion options, self-assessing medication abortion eligibility, how to take abortion medications according to currently endorsed regimens, how to determine completion, and how to navigate the formal health care system if needed.During the data collection period, abortion was legally restricted at all sites and largely unavailable in the formal health care system.
Between on what was accessible in their setting: mifepristone in combination with misoprostol or misoprostol alone.This analysis is restricted to participants who reported taking misoprostol alone.Due to lack of variation in medication regimen used within each country site, we were unable to conduct a direct comparison between misoprostol-alone users and combined regimen users.

Outcomes
The primary outcome was effectiveness, defined as abortion completion without procedural intervention, as assessed at each follow-up through the questions, "Do you think your abortion is complete?"(Argentina and Nigeria) or "Do you think you are still pregnant?"(Southeast Asia).
Participants were asked how they determined that their abortion was complete and direct questions about receipt of manual vacuum aspiration or dilation and curettage or evacuation procedures.The secondary outcome was safety, which was conceptualized as an experience of potential warning signs (bleeding that soaks more than 2 pads per hour for >2 hours, pain that does not go away with pain relievers and impedes normal activities, fever >38 °C for >24 hours, and/or foul-smelling yellow or green vaginal discharge) and potential adverse events indicated by receipt of the following medical treatments: intravenous fluids, blood transfusion, and/or overnight stay in a hospital.Data collectors assessed safety outcomes using a checklist read to participants (eg, "Did you experience any of the following…?").Other outcomes of interest included details on regimen dosing, initiation and duration of bleeding and cramping, adverse effects, time to expulsion (measured in hours from first dose of misoprostol), and health care-seeking experiences.Participants self-reported all outcomes over the phone via a standardized interviewer-administered questionnaire at first and second follow-up (eFigure 1 in Supplement 1).

Statistical Analysis
The data analysis was performed between January 6, 2022, and September 8, 2023.We calculated proportions of outcomes, estimated 95% CIs for primary outcomes, and calculated frequencies and ranges of other outcomes (1) overall, (2) by pregnancy duration at the time of the abortion, and (3) by regimen adherence.Pregnancy duration was assessed by date of last menstrual period 23 or ultrasound.We compared ever seeking health care between participants who ever experienced potential warning signs and those without warning signs using χ 2 tests of independence.Among participants who sought health care, we compared reasons for care seeking and treatment received between those who reported potential warning signs and those without warning signs using χ 2 tests of independence.We considered a 2-sided P < .05 to be statistically significant.

JAMA Network Open | Obstetrics and Gynecology
While clinical assessment of outcomes was impossible for legal and ethical reasons, prior research has demonstrated the validity of patient self-assessment of abortion outcome. 24We additionally conducted a Monte Carlo sensitivity analysis 25 to account for (1) possible misclassification of the primary outcome due to self-report and (2) selection bias from possible differential loss to follow-up (a full description of Monte Carlo sensitivity analysis methods is provided in the eMethods in Supplement 1).
Main analyses were conducted using Stata, version 15 software (StataCorp LLC).Sensitivity analyses were conducted using R, version 4.0.2software (R Foundation for Statistical Computing).

Results
A total of 2265 callers were screened for eligibility, of whom 1594 were eligible and 1352 consented and enrolled (85%).For this analysis, we excluded 715 callers who did not use misoprostol alone Most participants reported experiencing some type of physical pain related to their abortion process (591 [92.8%]).Most participants (575 [90.3%]) did not take anything to prevent pain prior to starting

Discussion
This cohort study is among the few prospective studies of misoprostol-alone effectiveness using currently endorsed regimens.The findings suggest that misoprostol alone is highly effective across pregnancy durations, with low adverse events.Effectiveness of misoprostol alone may be similar to the combined regimen of mifepristone and misoprostol, especially in settings with limited access to or use of early procedural intervention.
The effectiveness reported in our study aligns with other studies of self-managed abortion with misoprostol alone (range, 88%-100%) 9,10,19,20,26 and is higher than the effectiveness reported by comparable clinical studies (range, 84.2%-93.7%). 16,17This difference in observed effectiveness may be due to several study design and service delivery features.For instance, clinical studies are predisposed to early procedural intervention, may use shorter timelines to evaluate completion and intervention (1-2 weeks vs 3-4 weeks in self-management study contexts), do not always allow for additional doses of misoprostol, and may not provide comprehensive on-demand counseling and support to participants about the expected experience of misoprostol alone.As most participants in this study reported completing their abortion at 1 week using only the planned doses, it may be possible that the higher effectiveness in the context of this study is due to comprehensive counseling about what to expect and how to ascertain that the abortion is complete as well as lack of ultrasonography for assessment of completion (which may lead to potential procedural overtreatment). 27,28 have provided detailed data on experiences of bleeding, cramping, and side effects.These findings may provide needed evidence that can be used to update patient information forms and counseling protocols to ensure that all individuals using misoprostol alone for abortion, regardless of setting, have accurate information about what to expect.0][31] Reducing unnecessary health care seeking is particularly important in contexts where health care seeking after an abortion places individuals at legal risk.

Implications
Misoprostol alone is a highly effective medication abortion regimen and warrants renewed attention.
Clinicians and advocates can use these data to inform expansions of their scope of practice to include misoprostol-alone regimens for abortion where they currently do not and to advocate for decriminalization and expansion of the full spectrum of options for abortion care.Clinicians, community health care practitioners, and safe abortion hotline counselors and accompaniers alike can draw from the detailed data on physical experiences to enhance the counseling and preparedness of people initiating a misoprostol-alone abortion and thereby potentially reducing unnecessary care seeking.
These findings may contribute to growing confidence in the misoprostol-alone regimen.Even in contexts where mifepristone is available, misoprostol alone could remain an option, especially for people who want to immediately begin their abortion process (vs waiting 24-48 hours after mifepristone as required by the combined regimen).This option may be particularly salient in contexts where people need to travel for abortion care and have concerns about expelling the pregnancy en route to or at home.This study provides needed guidance on the effectiveness of this

Strengths and Limitations
This study has several strengths, including a large sample size, detailed and systematic data collection on previously unreported aspects of the misoprostol-alone experience (particularly physical and health care-seeking experiences), low loss to follow-up, and strong collaborative partnership 32 in design and implementation of the study.This study also has several limitations.Due to legal restrictions on abortion access in the study sites at the time of data collection, this study used selfreported, observational outcomes rather than clinical assessment of randomly assigned study interventions.However, research has demonstrated concurrence between patient self-assessment of abortion outcome vs ultrasonography following medication abortion. 24Additionally, 45 participants (7.1%) did not complete a 3-week follow-up; while the majority of these participants (41 [91.1%]) reported a complete abortion without procedural abortion at 1-week follow-up, it is possible that some procedural interventions that occurred after 1-week follow-up were not captured.To account for this limitation, we conducted a Monte Carlo sensitivity analysis, which indicated that our measure of effectiveness was robust to the possible misclassification and selection factors described here.

Conclusions
The findings of this study indicate that misoprostol-alone regimens may open opportunities for innovative access through nontraditional practitioners, pharmacists, and other contexts.As abortion access, and specifically medication abortion, comes under increasing legal attack around the world, expanding the availability of existing, evidence-based methods for medication abortion can help to ensure that all people can access abortion when and where they need it.

Table 3 .
Bleeding, Cramping, Expulsion, and Side Effects Among Participants Using Misoprostol Alone in the Studying Accompaniment Feasibility and Effectiveness Study Most participants (584 [91.6%; 95% CI, 89.4%-93.7%])didnotreport any warning signs that would indicate the possible need for follow-up care, such as prolonged bleeding, extreme pain, high fever, or problematic vaginal discharge (Table2).Medical experiences that could indicate an adverse event (eg, receiving intravenous fluids, blood transfusion, or overnight hospital stay) were

Table 4 .
Care-Seeking Experiences Among Participants Using Misoprostol Alone Who Sought Follow-Up Care in the Studying Accompaniment Feasibility and Effectiveness Study a Fisher exact test.bParticipants could select more than 1 response.Downloaded From: https://jamanetwork.com/

on 10/27/2023 likely
to receive no medical treatment when seeking care than those with at least 1 warning sign (97 [77.0%] vs 3 [13.0%];P < .001).Almost all participants (589 [99.5%]) reported receiving all the support they needed from the hotline.When asked about preference for future abortion care, 557 participants (94.1%) reported that they would prefer to self-manage their abortion with support from the hotline, and 15 (2.5%) reported that they would prefer to go to a health facility (eTable 6 in Supplement 1).

JAMA Network Open | Obstetrics and Gynecology Medication
Abortion Safety and Effectiveness With Misoprostol Alone more widely available regimen, particularly in the face of increasingly stringent and severe restrictions on access to mifepristone in the US and internationally.

eTable 4 .
Abortion Completion Among Participants Who Used Misoprostol Alone in the SAFE Study by Misoprostol Amount and Regimen (N = 637) eTable 5. Bias-Corrected Effectiveness From a Monte Carlo Sensitivity Analysis of Data From the SAFE Study (N = 50 000 Iterations) eTable 6. Support and Preferences for Future Abortion Care Among Participants Who Used Misoprostol Alone in the SAFE Study at Last Follow-Up (n = 592) eFigure 2. Initiation of Bleeding, Cramping, and Expulsion Among Participants Using Misoprostol Alone in the SAFE Study (N = 637)