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
Departments of Psychology, Psychiatry, Pharmaceutical Sciences, and Clinical Translational Science, University of Pittsburgh, Pittsburgh, PA, United StatesPinney Associates, Pittsburgh, PA, United States
To measure adherence over six months of progestin-only pill (POP) use.
Study Design
Prospective observational cohort study measuring adherence to daily dosing and timing of dose in patients prescribed a POP, with up to six months of follow-up, conducted from January to October 2020. A pharmacy benefit manager identified potential participants with a newly prescribed POP and extended an invitation to participate. We enrolled qualified respondents by telephone, trained them to use an electronic diary to report daily whether they had taken their POP and at what time. We followed participants for up to six months. We calculated adherence to daily pill taking as the proportion of evaluable days in which a participant took a POP, and the proportion of participants reporting ≥85% adherence. We calculated adherence to same time each day as the proportion of doses taken no later than three hours after the previous dose time of day.
Results
The user population comprised 199 participants, 154 (77.4%) of whom completed six months of follow-up. The majority (n = 170, 85.4%) were taking norethindrone. Norethindrone users reported POP intake on 22,327 (96.4%) of 23,156 evaluable days, with 155 (91.2%) participants reporting ≥85% adherence; less than half (n = 73, 42.9%) reported 100% adherence. Participants reported adherence to same time each day on 21,698 of 22,157 (97.9%) evaluable days.
Conclusions
Among participants taking a prescribed POP, participants demonstrated high adherence for daily pill taking and the same time of day, though the majority were not 100% adherent.
Implications
This study reports data specific to adherence among those taking a progestin-only pill (POP) in the prescription setting. Clinicians who counsel patients about POP use should be aware that majority of patients were not 100% adherent, although most report ≥85% adherence.
Progestin-only pills (POPs) contain a low dose of progestogen which is taken daily without any hormone-free interval. Clinical guidance advises taking pills at the same time every day; a POP is considered late if taken more than three hours after it should have been. However, White et al. note that this is not based on firm evidence; no clinical data are available that correlate pregnancy rates with timeliness in taking POPs [
]. POPs prevent pregnancy through multiple potential mechanisms, including variable inhibition of ovulation and thickening of cervical mucus to render it relatively impenetrable to sperm [
]. Two small pharmacokinetic studies undertaken among users taking d-norgestrel 30 mcg or norgestrel 75 mcg suggested that peak concentrations are reached about 2 hours after ingestion, and rapid distribution and elimination result in serum steroid levels near baseline after 24 hours [
Serum levels of d-norgestrel, luteinizing hormone, follicle-stimulating hormone, estradiol, and progesterone in women during and following ingestion of combination oral contraceptives containing dl-norgestrel.
]. While there are no direct data that define the increased pregnancy risk associated with missing a dose or taking a pill late, the available data have been interpreted to mean that POPs are more sensitive to deviations from dosing instructions than combined oral contraceptives (COCs). If that is the case, then understanding adherence among POP users may support clinicians in providing optimal counseling and care.
2. Materials and Methods
We conducted a 24-week prospective observational cohort study entitled Baseline Evaluation of adhereNce among Consumers with Home Management of A Prescribed POP as Representative in Kind (BENCHMARK), to evaluate adherence to recommendations to take POPs every day, at the same time of day. The Sterling Institutional Review Board (Atlanta, GA) approved this study.
2.1 Recruitment
Two research companies (United Biosource Corporation, Blue Bell, PA and Healthcore, Wilmington, DE) with access to pharmacy benefits manager data (Express Scripts, St. Louis, MO and Humana Pharmacy Solutions, Lexington, KY) identified patients across the U.S. with a new prescription within the past two months for any POP available in the U.S., and no other within the past year. They sent invitations in four waves from January 16, 2020, to April 7, 2020, during which norethindrone and drospirenone were available by prescription only. Respondents either called in for telephone screening or completed an on-line questionnaire. We enrolled participants able to read, speak and understand English, assigned female at birth, aged 18 years or older, who had internet access and a smartphone or tablet, who did not work in healthcare, and who had started taking a POP within the previous two months.
2.2 Study procedures
In a standardized enrollment telephone interview, participants provided explicit verbal consent and reported brief medical history, current medication use, and demographic information. At the end of the enrollment interview we trained participants to use an app-based electronic diary (e-diary). We asked participants to record each day whether, on the day before, they had taken their POP and at what time, whether they had sexual intercourse, and whether they used any additional form(s) of contraception. When participants recorded that they missed a POP dose, the e-diary application prompted them to indicate from a list of options the reason(s) why. To avoid influencing adherence, we sent reminders to complete the e-diary every four days. The e-diary allowed participants to enter data up to 10 days in arrears and recorded the date and time of each entry.
Trained nurse interviewers collected data via up to seven telephone interviews at weeks two, four, eight, 12, 16, 20, and 24 documenting if and when participants took the POP, any adverse events (AEs), concomitant medications, and other actions related to the use of the POP. To further assess the possible impact of missing diary entries, we inquired about POP use on the missing days during scheduled telephone interviews and collected clarifications to be analyzed separately. Participants could discontinue the POP at their discretion or in consultation with their provider, which we considered distinct from nonadherence to dosing directions. If a participant elected to discontinue use of the POP, we conducted the scheduled interview and end-of-study interview and ended their participation. At the end of the study, participants performed a self-administered urine pregnancy test. The first participant was enrolled on January 23, 2020, with enrollment continuing until April 13, 2020. The final participant concluded participation on October 29, 2020.
2.3 Measures
We defined “adherence evaluation days” as the days between each participant's first reported use of the POP through their discontinuation of the POP or study completion. We designated the primary measure of adherence as the proportion of adherence evaluation days where a diary report was available on which participants reported taking the POP. Additional analyses took into consideration whether participants took actions to mitigate pregnancy risk associated with missing a pill by using a barrier method of contraception or not having intercourse during the following 48 hours. We calculated adherence to daily use at the participant level as the proportion of participants who took their POP on at least 85% of their adherence evaluation days, representing no more than one missed pill per week on average, consistent with adherence to COCs under typical use conditions based on literature reports [
We measured participants’ use of their POP at the same time of day among all adherence evaluation days in which participants reported a dosing time. We considered dosing within three hours after the immediately previous pill's dosing time as correct for “same-time” dosing.
For all measures we calculated simple proportions with 95% confidence interval (Exact method), excluding missing data for the primary analyses. To assess the impact of missing data, we performed best-case and worst-case sensitivity analyses in which each day with missing diary data the POP was either assumed to have been taken (best-case) or not been taken (worst-case).
We determined the target sample size of approximately 200 based on the number needed to obtain a margin of error of 5% at a proportion of 85% with a confidence level of 95%.
3. Results
3.1 Participant Characteristics
We extended 8,830 invitations, to which 354 respondents initiated screening, of whom 146 (41.2%) did not qualify, most (n = 115, 78.8%) because they had started POP use more than 2 months prior. The study flow chart (Fig. 1) documents participant screening, disposition, and participation throughout the study. Of 203 enrollees, 199 (98.0%) reported use of their POP at least once (user population), of whom 85.4% (n = 170) were taking norethindrone and comprised the adherence analysis population. About two-thirds of all enrollees (n = 141, 69.5%; 70.9% of users) completed all follow-up assessments.
Fig. 1Study flow diagram: Prospective cohort study of pill adherence to progestin-only pills (POP) in the U.S. from January to October 2020
Serum levels of d-norgestrel, luteinizing hormone, follicle-stimulating hormone, estradiol, and progesterone in women during and following ingestion of combination oral contraceptives containing dl-norgestrel.
Table 1 presents characteristics of the user population. Table 2 outlines the distribution of health conditions of interest, many of which included possible contraindications to estrogen [
]. Over three-quarters of participants reported at least one of the conditions of interest (n = 155, 77.9%); almost half (n = 92, 46.2%) reported current breastfeeding. Over a quarter (n = 54, 27.1%) reported having migraine headaches, more than half of whom (n = 33, 16.6%) reported migraine with aura. There were 59 participants (29.6%) with a contraindication for estrogen and 145 (72.9%) who were breastfeeding and/or had a contraindication for estrogen. Based on BMI, 52 (26.1%) were obese.
Table 1Demographics of new users of progestin-only pills (POP)
Table 2Health conditions of interest among new users of progestin-only pills participating in a prospective cohort study on pill adherence in the U.S. from January to October 2020
All users
(N = 199)
Self-reported health conditions of interest
Ever been diagnosed with cancer
1 (0.5%)
Liver problems
1 (0.5%)
High blood pressure
20 (10.1%)
Gallbladder disease
0 (0.0%)
High cholesterol
7 (3.5%)
Diabetes
0 (0.0%)
Migraines with aura
33 (16.6%)
Migraines without aura
21 (10.6%)
Blood clot history
0 (0.0%)
Heart disease
2 (1.0%)
History of heart attack or stroke
0 (0.0%)
Current smoker
11 (5.5%)
Currently breastfeeding
92 (46.2%)
Total Number of Participants Who Report Having at Least one of the Health Conditions of Interest
155 (77.9%)
Total Number of Participants Contraindicated for Use of an estrogen-containing contraceptive
Participants who reported at least one of the following: migraine with aura, smokers age 35 and older, breast cancer, hypertension, heart disease, liver problems, or history of blood clots, stroke, or heart attack.
59 (29.6%)
Total Number of Participants Contraindicated for Use of an estrogen-containing contraceptive
Participants who reported at least one of the following: migraine with aura, smokers age 35 and older, breast cancer, hypertension, heart disease, liver problems, or history of blood clots, stroke, or heart attack.
and/or Breastfeeding
145 (72.9%)
a Participants who reported at least one of the following: migraine with aura, smokers age 35 and older, breast cancer, hypertension, heart disease, liver problems, or history of blood clots, stroke, or heart attack.
While we did not design the study to evaluate efficacy, no pregnancies were reported.
3.2 Adherence
Among the user population, participants completed their e-diary entries within four days in 90% of instances (n = 24,979 of 27,552 evaluable e-diary entries); most (n = 20,837; 75.6%) reported on the first (n = 16,018; 58.1%) or second (n = 4,819; 17.5%) day. Participants taking norethindrone reported taking the POP on 22,327 (93.7%) of 23,838 adherence evaluation days; participants reported not taking the POP on 829 (3.5%) days and e-diary data were missing for 682 (2.9%) days, yielding 96.4% (95% CI 96.3–96.7, n = 22,327 of 23,156) adherence to daily dosing. Sensitivity analyses showed worst-case adherence to be 93.7% (95% CI 93.3–94.0; n = 22,327 of 23,838) and best-case adherence to be 96.5% (95% CI 96.3–96.8; n = 23,009 of 23,838). Among 368 missing days that we were able to clarify during telephone interviews, participants reported having taken their POP on most such days (n = 350, 95.1%). Some missing days were not clarified because of missing interim interviews or time elapsed since the missing entry.
For the 829 days without POP intake, 189 (22.8%) participants reported having sex at least once that day or the following two days without using a barrier method, 71 (8.6%) reported sex with a barrier method, and 550 (66.3%) reported no sex.
When evaluated by participant, 155 (91.2%, 95% CI 85.9–95.0) were adherent to daily use of their POP on at least 85% of adherence evaluation days. Sensitivity analyses for missing days yielded estimates of 87.1% (95% CI 81.1–91.7; n = 148 of 170) under worst-case assumptions, and 91.8% (95% CI 86.6–95.4; n = 156 of 170) under best-case assumptions. As shown in Fig. 2, individual adherence ranged from 58.9% to 100%, with less than half (n = 73, 42.9%) reporting 100% adherence.
Fig. 2Participant level adherence among new users of progestin-only pills participating in a prospective cohort study on pill adherence in the U.S. from January to October 2020.
Of the 22,157 (92.9%) adherence evaluation days evaluable for time of dose, reported dosing was no more than three hours after the time of day of the previous dose on 21,698 (97.9%, 95% CI 97.7–98.1) days, with 20,695 (93.4%, 95% CI 93.1-93.7) within no more than one hour.
We performed adherence analyses for participants taking drospirenone (n = 22, 11.1%) and for those taking an unknown POP (n = 7, 3.5%) and observed no clinically meaningful differences in adherence (data not shown).
Among 959 missed doses for the entire user population, the most common reason selected was “Forgot” (n = 635, 66.2%). Reasons related to access to pills were common: “Ran out of pills – but plan to continue” was cited for 116 (12.1%) missed doses, and “Didn't have pills with me” for 68 (7.1%). “Other reason” (no further information) was cited for 127 (13.2%).
4. Discussion
This study provides data on adherence to daily pill taking and timeliness of dosing among people taking a prescribed POP. Adherence with all types of daily medication is far from perfect, especially for preventive medications including oral contraceptives (OC) [
]. Published data on adherence to COCs from studies with a range of methodologies suggests that most users are not fully adherent to daily dosing and that there is significant variability in adherence behaviors between users and between cycles within the same users [
Adherence with ethinylestradiol 20 μg/drospirenone 3 mg in a flexible extended regimen supported by the use of a digital tablet dispenser with or without acoustic alarm: an open-label, randomized, multicenter study.
]. Two adherence studies for COC users with daily diaries, each with 3 months of participation, demonstrated that up to 18% of users reported missing three or more active pills per cycle [
We found that POP users are generally adherent to daily pill taking, though as with COC users, they do sometimes miss taking their pills. There is inter- and intra-participant variability in adherence behavior. The participant-level analysis demonstrates that while most participants reported high levels of adherence, the range in overall adherence level is wide.
The potential outcome of nonadherence is pregnancy. No study has directly quantified pregnancy risk or occurrence after one or more missed POPs, and while it is presumed that an extended period of nonadherence is likely to increase pregnancy risk, there are no data to suggest how episodic nonadherence relates to pregnancy risk.
According to prescribing information, norgestrel should be taken at the same time each day, within a three-hour window. The literature on OC adherence suggests that adherence to “same time” directions is limited. In two studies evaluating dose timing in COC users, the majority of pills were not taken within two hours of the time of day of the previous dose [
Adherence with ethinylestradiol 20 μg/drospirenone 3 mg in a flexible extended regimen supported by the use of a digital tablet dispenser with or without acoustic alarm: an open-label, randomized, multicenter study.
]. Participants in our study reported a high degree of adherence to the recommendation to take the pill at the same time each day, doing so on 97.9% of evaluable days. The expected impact of departures from “same-time” directions on pregnancy risk in POP users is difficult to evaluate, especially as the three-hour window recommendation is based on limited pharmacokinetic data [
Serum levels of d-norgestrel, luteinizing hormone, follicle-stimulating hormone, estradiol, and progesterone in women during and following ingestion of combination oral contraceptives containing dl-norgestrel.
There is considerable interest and support from the medical community in expanding access to oral contraceptives by making them available over-the-counter [
]. We found that some nonadherence was attributable to limitations in access to the medication; in one out of five instances of nonadherence, participants reported that they had run out of pills but planned to continue, or that they did not have their pill with them. Over-the-counter access to an oral contraceptive would reduce such barriers. Participants in our study may have been given directions on daily same-time dosing. Of course, carrying out those directions still rests with the user. Review of available studies suggests that provider counseling may not improve adherence [
Strengths of this study include a sample comprising patients who were prescribed a POP in the context of a natural interaction with a healthcare provider. We collected adherence data prospectively using a daily electronic diary on the participants’ own devices. Contemporaneous reporting of sexual activity and any use of additional methods of contraception inform pregnancy risk associated with missing doses. We collected up to 24 weeks of daily use data from participants. Missing data were minimal, and sensitivity analyses outline best- and worst-case adherence measures, and diary clarification efforts provided context for understanding missing diary entries. The large number of adherence evaluation days allows for a relatively precise estimate of overall adherence.
Our study was limited by omission of uninsured participants. While we did not collect data on the nature of insurance coverage, the pharmacy benefits managers we used included both private and public plans. The number of participants taking drospirenone, a newer POP, is small. While the dosing instructions for drospirenone advise users to take it at the same time every day, they do not include the three-hour window instruction. We began collecting data up to two months into use, which may bias overall measures of adherence by either missing the adherence in the first weeks of use, or by omitting those who stopped use within two months. Study measurement procedures may have influenced participants’ adherence.
Sample characteristics may limit generalizability to contraception-seeking women in general. Our sample was older and more educated than the general population of oral contraceptive users in the U.S.[
] In the US, POPs are typically used in those who are breastfeeding or who have a medical contraindication to estrogen, which may also be influenced by age. Oral contraceptive users that are older, have more years of education, higher levels of health literacy, and/or better socioeconomic status have comparatively greater odds of adherence and oral contraceptive continuation, so this study may represent a best-case sample for adherence, and over-estimate adherence for a more general population of OC users [
]. Nearly half of the study participants reported breastfeeding, with another quarter reporting a history of migraine headaches. This contrasts with the expected characteristics of a general oral contraceptive user population, which might be expected to be younger and healthier, with a much smaller proportion of those who are recently postpartum.
In summary, participants were largely adherent to daily pill taking, though missed pills were not uncommon. They were also highly adherent to the recommendation to take the product at the same time every day.
Acknowledgments
The authors would like to offer a very special thanks to all those participants who agreed to enroll in this study, sharing with us their personal experiences and contributing to our collective understanding of how POPs are used. We also thank all of those at PEGUS Research, Pinney Associates and HRA Pharma who contributed to the conduct of this study and this manuscript. We especially appreciate Lucy Owen's skilled editorial review.
References
White K
Potter JE
Hopkins K
Fernández L
Amastae J
Grossman D.
Contraindications to progestin-only oral contraceptive pills among reproductive-aged women.
Serum levels of d-norgestrel, luteinizing hormone, follicle-stimulating hormone, estradiol, and progesterone in women during and following ingestion of combination oral contraceptives containing dl-norgestrel.
Adherence with ethinylestradiol 20 μg/drospirenone 3 mg in a flexible extended regimen supported by the use of a digital tablet dispenser with or without acoustic alarm: an open-label, randomized, multicenter study.
☆Declaration of competing interest: RD Bradford and SJ Farnsworth are employees of PEGUS Research, Inc, the contract research organization hired to conduct this study. I Laurora and H Guillard are employees of HRA Pharma, the sponsor of this study. S Sober, A Glasier and S Shiffman serve as consultants to HRA Pharma.
☆☆Funding: This study was funded by HRA Pharma and conducted by PEGUS Research, Inc. HRA Pharma contributed to the design of the study, review of the report, and decision to submit this article for publication, but had no role in the collection or analysis of the data.
00108-1&id=gr1.jpg){kind=link}
00108-1&id=gr2.jpg){kind=link}