If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Advancing New Standards in Reproductive Health (ANSIRH), Bixby Center for Global Reproductive Health, Department of Obstetrics, Gynecology & Reproductive Sciences, University of California, San Francisco, Oakland, CA, United States
Advancing New Standards in Reproductive Health (ANSIRH), Bixby Center for Global Reproductive Health, Department of Obstetrics, Gynecology & Reproductive Sciences, University of California, San Francisco, Oakland, CA, United States
Advancing New Standards in Reproductive Health (ANSIRH), Bixby Center for Global Reproductive Health, Department of Obstetrics, Gynecology & Reproductive Sciences, University of California, San Francisco, Oakland, CA, United States
Department of Medicine, Division of General Internal Medicine, University of California, Richard Fine People's Clinic, San Francisco, CA, United States
Advancing New Standards in Reproductive Health (ANSIRH), Bixby Center for Global Reproductive Health, Department of Obstetrics, Gynecology & Reproductive Sciences, University of California, San Francisco, Oakland, CA, United States
Department of Medicine, Division of General Internal Medicine, Center for Healthcare Policy and Research, University of California, Davis, Sacramento, CA, United States
To estimate the effectiveness, acceptability, and feasibility of medication abortion with mifepristone dispensed by a mail-order pharmacy after in-person clinical assessment.
Study Design
This is an interim analysis of an ongoing prospective cohort study conducted at five sites. Clinicians assessed patients in clinic and, if they were eligible for medication abortion and ≤ 63 days’ gestation, electronically sent prescriptions for mifepristone 200 mg orally and misoprostol 800 mcg buccally to a mail-order pharmacy, which shipped medications for next-day delivery. Participants completed surveys three and 14 days after enrollment, and we abstracted medical chart data.
Results
Between January 2020 and April 2021 we enrolled 240 participants and obtained clinical outcome information for 227 (94.6%); 3 reported not taking either medication. Of those with abortion outcome information (N = 224), 216 (96.4%) completed day-3 and 212 (94.6%) day-14 surveys. Of the 224 that took medications, none reported taking past 70 days’ gestation, and complete medication abortion occurred for 217 participants (96.9%, 95% CI 93.7%-98.7%). Most received medications within three days (82.1%, 95% CI 76.5%-86.9%). In the day-3 survey, 95.4% (95% CI 91.7%-97.8%) reported being very (88.4%) or somewhat (6.9%) satisfied with receiving medications by mail. In the day-14 survey, 89.6% (95% CI 84.7%-93.4%) said they would use the mail-order service again if needed. Eleven (4.9%, 95% CI 2.5%-8.6%) experienced adverse events; two were serious (one blood transfusion, one hospitalization), and none were related to mail-order pharmacy dispensing.
Conclusions
Medication abortion with mail-order pharmacy dispensing of mifepristone appears effective, feasible, and acceptable to patients.
Implications
The in-person dispensing requirement for mifepristone, codified in the drug's Risk Evaluation and Mitigation Strategy, should be removed.
For over 20 years, the U.S. Food and Drug Administration (FDA) has required that mifepristone (200 mg) used for medication abortion be dispensed only in clinics, medical offices, and hospitals, by or under the supervision of a certified prescriber; this restriction is codified in the drug's Risk Evaluation and Mitigation Strategy (REMS) [
]. In addition, a recent study found that dispensing mifepristone at brick-and-mortar pharmacies after in-person assessment by a clinician was effective and acceptable to patients seeking abortion, with a low prevalence of adverse events (AEs) [
]. Mail-order pharmacy dispensing of mifepristone could further allow medication abortion patients to bypass geographical obstacles to the service and receive abortion care earlier in pregnancy [
], it could also help increase the number of clinicians willing and able to provide medication abortion by enabling them to avoid cost and logistical challenges of stocking and dispensing mifepristone in their facilities, thereby improving access to care [
Under an Investigational New Drug Application (IND) from the FDA, we began a prospective cohort study of the acceptability, feasibility, and effectiveness of dispensing medications for abortion using a mail-order pharmacy following an in-person assessment for eligibility. In April 2021, the FDA announced it would suspend the in-person dispensing requirement for mifepristone during the COVID-19 public health emergency. We conducted this interim analysis at the midpoint of planned recruitment to provide additional information to the FDA as it considers the future of the mifepristone REMS after the emergency ends.
2. Material and Methods
From January 2020 to April 2021, we enrolled patients at five clinics in four states seeking medication abortion who consented to receive their medications from a mail-order pharmacy; two clinics had not previously provided abortion care. The Institutional Review Boards (IRBs) of the University of California San Francisco (UCSF), Albert Einstein College of Medicine, Kent Hospital, and Alameda Health System approved the study. Site investigators completed the mifepristone Prescriber Agreement Form. Clinicians evaluated patients in person for medical eligibility according to the mifepristone label after confirming interest in medication abortion. Patients were eligible for the study if they were aged 15 years or older, spoke English or Spanish, were willing to administer misoprostol buccally as mandated by the FDA, and had a gestational duration of ≤ 63 days (to reduce the chance that shipping delays might result in a participant taking mifepristone past 70 days’ gestation). Clinicians at study sites determined gestational duration based on patients’ history, confirmed by physical examination or ultrasound.
Those interested and eligible provided written informed consent. Patients signed the mifepristone Patient Agreement Form and received instructions on medication use. All participants were scheduled for clinical follow-up, which varied at sites and included: (1) in-person ultrasonography one to two weeks after mifepristone, (2) serum human chorionic gonadotropin (hCG) measurements on the day of enrollment and one to two weeks after mifepristone, or (3) a telephone call one week after mifepristone with a home urine pregnancy test four weeks after mifepristone, sometimes accompanied by a telehealth visit. If clinical follow-up did not occur, we relied on patients’ self-report of abortion outcome in their survey responses, as done in prior research [
Treating clinicians electronically prescribed each participant mifepristone 200 mg and misoprostol 800 mcg, as well as analgesics, antibiotics, antiemetics, or contraceptives at their discretion. Clinicians instructed participants to take misoprostol buccally 24 to 48 hours after mifepristone. The mail-order pharmacy processed the prescriptions according to their standard “next-day delivery” workflow and shipped the medications to each participant's preferred address.
Three days later, we emailed participants a Qualtrics survey link to collect information about whether their medication package delivery time was “reasonable” or “too long,” whether the package had been damaged or opened, and whether their confidentiality might have been compromised in any way. We also collected information on participants’ sociodemographic characteristics, satisfaction with receiving the medications by mail, if and/or when they took or planned to take the medications, whether they thought they had “passed the pregnancy,” and whether they had any new medical problems related to their abortion. We asked those who reported a prior medication abortion, “How would you compare your experience of getting the pills this time by mail compared with last time in the clinic?”.
Two weeks later, we emailed participants a second survey, which included similar questions about taking the medications, medical problems for which they sought further care, clinic follow-up, and abortion completion. Participants who believed their abortion was complete were asked how they knew that, including whether they had an ultrasound, blood tests, a negative urine pregnancy test, a normal period, whether their pregnancy symptoms went away, or for some other reason (with an open response). We also asked about their overall medication abortion experience and whether they would recommend mail-order dispensing of mifepristone to a friend. We also asked, “If you were to have an undesired pregnancy in the future and chose to have a medication abortion, would you want to get the pills by mail again?”.
Six or more weeks after enrollment, personnel from each participating site entered de-identified data about all visits or contact with the patient into an electronic REDCap form. We used these data abstracted from medical records, in combination with survey data, to identify abortion outcomes and AEs occurring within six weeks of enrollment, including any emergency department visits at least possibly related to the medication abortion. We considered serious AEs to be death, hospitalization, blood transfusion, or surgery, according to FDA criteria [
Participants received electronic gift cards at enrollment ($15) and after completing each survey ($25 each). The study paid for the medication abortion, including the cost of clinical visits, mifepristone, misoprostol, pharmacy dispensing fees, and shipping.
We evaluated the effectiveness of medication abortion as the proportion of participants who had a complete abortion with medications alone. We used data abstracted from clinic records for this outcome; if the patient did not have follow-up with the clinic, we used self-reported survey data to document abortion outcome, similar to prior research [
]. We evaluated safety as the proportion who had any AE that was at least possibly related to the medication abortion. For acceptability, we assessed the proportion of participants reporting in the day-14 survey they would use mail-order dispensing again if needed and the proportion reporting they were “satisfied” or “very satisfied” with the medication abortion. For feasibility, we measured the proportion who received packages within three days, reported the packages they received remained intact, and reported their confidentiality was maintained. We calculated binomial 95% confidence intervals for main outcomes in Stata 15 (College Station, TX).
The planned sample size for the ongoing study is a minimum of 440 participants. This will give us approximately 80% power to estimate acceptability of the care model and over 90% power to estimate effectiveness and demonstrate if these estimates are within a 5% non-inferiority margin of published estimates for facility-dispensed medication abortion. Prior research with patients obtaining medication abortion with in-person mifepristone dispensing has estimated the proportion who would use medication abortion again as 89.7% [
Clinic staff assessed 395 patients for eligibility and enrolled 240 participants; 155 were ineligible or declined to participate (Fig. 1). All enrolled participants received the study medications by mail. We obtained clinical outcomes for 227 participants (94.6%); 13 (5.4%) participants were lost to follow-up. Three participants reported not taking mifepristone or misoprostol; two returned the package, and one reported losing it. We excluded them from the analysis, leaving a final sample of 224 participants.
Fig. 1Study flow of patients seeking medication abortion who were evaluated in person and then received mifepristone from a mail-order pharmacy in the U.S. in 2020 to 2021
Table 1 shows the characteristics of study participants; of note, as of the time of this interim analysis, we had not recruited any minors. Table 2 shows the delivery time for the medication package; the details of when and how participants reported using the medications, including gestational duration on date of taking mifepristone, initial misoprostol dosage and route of administration, and interval between medications; and abortion outcome, including how abortion outcome was assessed. Overall, 217 of 224 participants (96.9%, 95% CI 93.7%-98.7%) had a complete abortion with medications alone, including eight (3.6%) who received a second dose of misoprostol 800 mcg. Five participants (2.2%) had ongoing pregnancies, and two (0.9%) had incomplete abortions treated with uterine aspiration. One of the participants with an ongoing pregnancy elected to continue the pregnancy; we contacted her last at 31 weeks’ gestation, and she reported her pregnancy was uncomplicated. No participant was found to have an ectopic pregnancy. We present outcomes by gestational duration in Supplemental Table 1.
Table 1Characteristics of study participants having medication abortion who received mifepristone from a mail-order pharmacy in the U.S. in 2020-21 (N = 224)
Table 2Dosing details and clinical outcomes for study participants having medication abortion and receiving mifepristone from a mail-order pharmacy in the U.S. in 2020-21 (N = 224)
4 participants underwent procedural abortion and 1 continued the pregnancy.
5
2.2
Method of confirmation of abortion outcome among participants who followed up with study clinic
N = 163
Clinical history and home urine pregnancy test (telephone visit)
78
47.9
Ultrasound
41
25.2
Clinical history alone (telephone visit)
37
22.7
Serial serum hCG testing
6
3.7
Negative urine pregnancy test (in clinic)
1
0.6
Method of confirmation of abortion outcome among participants who did not follow up with study clinic & indicated abortion completion in survey responses
Eleven participants (4.9%, 95% CI 2.5%-8.6%) had an AE that was determined by the site and principal investigators to be at least possibly related to the medication abortion; however, the investigators determined that none of these AEs would have been avoided by dispensing medications in person rather than with the mail-order pharmacy. All participants with an AE reported taking the medications at ≤ 70 days’ gestation as instructed. Among those with an AE, 10 had an emergency department visit (one reported two visits for the same problem), and two had a possible serious AE (one received a blood transfusion, and one reported being hospitalized overnight but received no additional treatment).
Packages arrived within three days for 184 of 224 participants (82.1%, 95% CI 76.5%-86.9%); most reported packages were in good condition (97.2%, 95% CI 94.1%-98.9%) and that confidentiality was uncompromised (96.8%, 95% CI 93.4%-98.7%) (Tables 2 and 3). Of those who reported compromised confidentiality, three said someone found the medications, three reported other privacy concerns, and one reported that both someone found the medications and they had other privacy concerns. Medications were not affected in any cases where the package was reported damaged. All who reported the date they took mifepristone had a gestational duration of ≤ 70 days on that date.
Table 3Acceptability and satisfaction at day-3 survey among study participants having medication abortion and receiving mifepristone from a mail-order pharmacy in the U.S. in 2020-21 (N = 216)
Medications were not affected in any cases where package was damaged.
2.3
Missing data
1
0.5
Confidentiality maintained during delivery
Yes
209
96.8
No, confidentiality was compromised
7
3.2
Adequate information received from clinic
Yes
215
99.5
No, I would have liked more information
1
0.5
Satisfaction with receiving medications by mail
Very satisfied
191
88.4
Somewhat satisfied
15
6.9
Neither satisfied nor dissatisfied
5
2.3
Somewhat dissatisfied
2
0.9
Very dissatisfied
3
1.4
Among those who had previous medication abortion, how they would compare experience of getting pills this time by mail compared with last time in clinic
N = 78
This time was better
43
55.1
Same
27
34.6
Last time was better
3
3.8
Not sure
4
5.1
Missing
1
1.3
Among participants who had abortion outcome data available and completed day-3 survey.
† Medications were not affected in any cases where package was damaged.
For the day-3 survey, we excluded eight of 224 (3.6%) participants missing data on satisfaction with receiving medications by mail. In the survey, 206 (95.4%, 95% CI 91.7%-97.8%) reported being very (88.4%) or somewhat (6.9%) satisfied with receiving medications by mail. Among those who reported a prior medication abortion (N = 78), 89.7% reported their experience of getting the pills this time by mail was better (55.1%) or the same (34.6%) compared with last time in the clinic (Table 3).
For the day-14 survey, we excluded 12 of 224 (5.4%) participants missing data on overall satisfaction with medication abortion. In this survey, 193 (91.0%, 95% CI 86.4%-94.5%) reported satisfaction with their overall medication abortion experience, and 190 (89.6%, 95% CI 84.7%-93.4%) said that they would use the mail-order service again if needed (Table 4). Most said they would recommend medication abortion (91.0%) and mail-order dispensing (94.3%) to a friend.
Table 4Acceptability and satisfaction at day-14 survey among study participants having medication abortion and receiving mifepristone from a mail-order pharmacy in the U.S. in 2020-21 (N = 212)
Preliminary findings from this study suggest that medication abortion with mail-order pharmacy dispensing of mifepristone was effective, feasible, and acceptable to patients seeking early abortion. Outcomes were similar to those reported when mifepristone has been dispensed in a clinic [
Acceptability and feasibility of early pregnancy termination by mifepristone-misoprostol. Results of a large multicenter trial in the United States. Mifepristone Clinical Trials Group.
While most participants received their medications within three days, 18% experienced a longer wait. We conservatively restricted enrollment to participants at ≤ 63 days’ gestation to account for potential delays in delivery; however, given evidence supporting mifepristone use up to 77 days’ gestation [
A repeat dose of misoprostol 800 mcg following mifepristone for outpatient medical abortion at 64-70 and 71-77 days of gestation: A retrospective chart review.
The study may have limited generalizability given that we recruited only in four states with few restrictions on abortion care. This model is not possible in the 19 states that require physicians to be in the same room with the patient when mifepristone is dispensed [
]. This interim analysis is also limited by the sample size. The full analysis will include at least four more sites, including in two new states, and up to 385 additional participants. A strength of the study is its low loss to follow-up.
We look forward to completing recruitment for the study in the coming year, which will provide more precise estimates of the effectiveness, acceptability, and feasibility of medication abortion with mail-order pharmacy dispensing of mifepristone. However, these interim results, together with studies of other care models that include mailing pills for medication abortion [
], support the removal of the FDA's in-person dispensing requirement for mifepristone.
Clinical Trial Registration
Clinicaltrials.gov, NCT03913104.
Acknowledgments
We thank the clinic and research staff at each of the study sites for assisting with recruitment and data collection, as well as the staff at American Mail Order Pharmacy.
Acceptability and feasibility of early pregnancy termination by mifepristone-misoprostol. Results of a large multicenter trial in the United States. Mifepristone Clinical Trials Group.
A repeat dose of misoprostol 800 mcg following mifepristone for outpatient medical abortion at 64-70 and 71-77 days of gestation: A retrospective chart review.
Declaration of Competing Interest: The author(s) declares(declare) no conflict of interest.
Funding: This work was supported by the Society of Family Planning Research Fund (SFPRF12-MA8) and the Taylor Fund. The views and opinions are those of the authors and do not necessarily represent the views and opinions of the funders.
Among participants who took medications and had abortion outcome data available.
00384-X&id=gr1.jpg){kind=link}