Article Text
PDF
Abstract
Objective To determine whether increasing the dose of ulipristal acetate (UPA)-containing emergency contraception (EC) improves pharmacodynamic outcomes in individuals with obesity.
Study design We enrolled healthy, regularly-cycling, confirmed ovulatory, reproductive-age individuals with body mass index (BMI) >30 kg/m2 and weight >80 kg in a randomised crossover study. We monitored participants with transvaginal ultrasound and blood sampling for progesterone, luteinising hormone (LH), and estradiol every other day until a dominant follicle measuring >15 mm was visualised. At that point, participants received either oral UPA EC 30 mg or 60 mg and returned for daily monitoring up to 7 days. After a no treatment washout cycle, participants returned for a second monitored cycle and received the other UPA dose. Our primary outcome was the proportion of subjects with no follicle rupture 5 days post-dosing (yes/no). For reference, we also enrolled a control group with BMI <25 kg/m2 and weight <80 kg who received UPA EC 30 mg during a single cycle. We also obtained blood samples for pharmacokinetic parameters for UPA and its active metabolite, N-monodemethyl-UPA (NDM-UPA) as an optional substudy.
Results We enrolled a total of 52 participants with BMI >30 kg/m2 and 12 controls, with the following cycles completed: 12 controls, 49 UPA 30 mg, and 46 UPA 60 mg. The entire cohort demographics were a mean (SD) age of 29.8 (3.4) years and BMI by group: controls 22.5 (1.4) kg/m2, group 1 37.9 (6.7) kg/m2, and group 2 39.3 (5.4) kg/m2. All 12 (100%) of controls had a delay of at least 5 days for follicle rupture. Among the high BMI group, dosing groups (UPA EC 30 mg vs 60 mg) were similar in the proportion of cycles without follicle rupture over 5 days post-UPA dosing (UPA 30 mg: 47/49 (96%), UPA 60 mg: 42/46 (91%), Fisher’s exact test p=0.43). However, after excluding cycles where dosing occurred too late (after LH surge), a delay of at least 5 days occurred in all participants at both doses. The 60 mg UPA dose resulted in a twofold increase in maximum observed concentration and the area under the curve of both UPA and NDM-UPA levels compared with 30 mg.
Conclusion A standard 30 mg dose of UPA is sufficient to delay ovulation regardless of BMI or weight. Results of our study do not support dose adjustment for body size.
- emergency contraception
- Obesity
Data availability statement
Data are available upon reasonable request. Investigators involved in this proposal are aware of and agree to abide by the principles for sharing research resources as described by NIH in 'Principles and Guidelines for Recipients of NIH Research Grants and Contracts on Obtaining and Disseminating Biomedical Research Resources'.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
A prior study identified that the effectiveness of ulipristal acetate (UPA)-containing emergency contraception (EC) might be negatively impacted by a higher weight and/or body mass index (BMI). Pharmacokinetic parameters of UPA in individuals of low and high BMI do not identify a concern for impaired effectiveness, but no prior studies of UPA’s active metabolite have been performed.
WHAT THIS STUDY ADDS
We provide critically missing pharmacodynamic data on UPA EC in individuals of varying weight/BMI. This study demonstrates that a standard dose of UPA EC (30 mg mg) has a very consistent therapeutic effect (at least a 5-day delay in ovulation) no matterregardless of an individual’s weight (and/or BMI). Additionally, the key pharmacokinetic parameters of UPA and its active metabolite appear unaffected by weight or BMI.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Pharmacokinetic and pharmacodynamic studies are indirect measures of the main outcome of interest for EC, —pregnancy. However, ovulation is necessary for pregnancy to occur and thus, the results of our study are further evidence that 30 mg mg UPA EC is effective in individuals of varying sizes.
Introduction
Emergency contraception (EC) is an essential medication for individuals at risk for, and not intending, pregnancy who experience an act of unprotected intercourse. Both orally-dosed ECs, levonorgestrel (LNG) and ulipristal acetate (UPA), work by inhibiting or delaying ovulation, reducing the risk of pregnancy for a single act of unprotected intercourse by up to 90%.1–4 Prior research has identified obesity as a known risk factor for LNG EC failure, but it remains unclear if UPA EC is similarly affected.5
Glasier et al performed a secondary analysis of two large randomised controlled trials to identify risk factors for EC failure.5 An individual with a body mass index (BMI) >30 mg/kg2 using LNG EC had more than a fourfold greater risk of pregnancy as compared with an individual with a normal BMI (OR 4.41, 95% CI 2.05 to 9.44). This re-analysis also demonstrated a higher risk of failure with UPA EC in users, but the difference in failure rate was not statistically significant (OR 2.62, 95% 0.89 to 7.00). Further research has yet to be performed to confirm or refute these clinical outcomes for UPA EC in individuals with a higher BMI or weight over 80 kg (176 lbs).
The pharmacokinetic (PK) parameters for UPA do not appear to support a difference in effectiveness. Unlike LNG EC, PK studies of UPA EC demonstrate very similar PK parameters in individuals of varying BMI.6–8 However, UPA is converted to an active metabolite, N-monodemethyl-UPA (NDM-UPA). NDM-UPA levels have not been compared across individuals of differing BMI.8 UPA is also known to bind distinctly to α-acid glycoproteins and high-density lipoproteins (HDL). Obesity is known to alter serum/plasma profiles of both drug-binding proteins: α-acid glycoprotein levels are usually doubled, and dyslipidaemia of obesity is often characterised by low HDL.9 10 These changes may translate into lower amounts of active drug at the end organ that could have an impact on EC effectiveness.
In this study, we took a similar approach to testing UPA EC in individuals of higher BMI as we and others have for testing LNG EC effectiveness.2 6 7 We designed this study to compare the pharmacodynamic effects of UPA 30 mg versus 60 mg.10–12 We hypothesised that UPA 60 mg would increase the proportion of subjects with no follicle rupture 5 days post-dosing as compared with 30 mg in individuals with BMI >30 kg/m2 and weight >80 kg. We also performed PK testing in a subset of participants to evaluate potential differences in MDM-UPA.
Methods
We conducted this randomised cross-over study at Oregon Health & Science University (OHSU) in Portland, OR, USA and Eastern Virginia Medical School (EVMS) Norfolk, VA, USA, from June 2017 to November 2021. The Institutional Review Board at OHSU and EVMS approved the study protocol. We recruited healthy women 18–35 years old with regular menstrual cycles (21–35 days not at risk for pregnancy, eg, abstinent, non-hormonal method of birth control, or non-sperm producing partner). Major exclusion criteria included: sensitivity or allergy to UPA; treatment for infertility; metabolic disorders including uncontrolled thyroid dysfunction or polycystic ovarian syndrome or clinical evidence of androgen excess; a screening serum progesterone level <3 ng/mL; impaired liver or renal function; actively seeking or involved in a weight loss programme (weight stable) or prior bariatric surgery; pregnancy or seeking pregnancy; breastfeeding; recent (8 weeks) use of hormonal contraception; smoking, vaping, or chronic marijuana use. We enrolled women with BMI ≥30 kg/m2 and weight ≥80 kg as the primary comparison group in a randomised crossover design, with each woman serving as her own control. We also enrolled a group of women with BMI <25 kg/m2 as normal BMI reference controls; controls were not randomised as they only underwent one treatment cycle.
We conducted a wide variety of recruitment activities aimed at enrolling eligible individuals from the catchment areas of both research sites. After an initial telephone screening, participants completed an in-person screening visit to collect baseline demographic and health information and a serum progesterone level during the luteal phase to confirm ovulatory status (progesterone level ≥3 ng/mL), that was an inclusion criterion for participation. All participants completed written informed consent before any study procedures. The study was conducted over three menstrual cycles; cycles 1 and 3 were treatment cycles interspersed by cycle 2, a washout cycle, in order to assure resumption of normal menstrual function after the potential for a prolonged cycle induced by UPA in cycle 1. The normal BMI control group participants only underwent one treatment cycle and thus were not part of the randomisation schema. We did not have participants undergo a baseline cycle with ultrasound and hormone monitoring other than a serum progesterone level confirming ovulation. Participants could request to space cycles 1 and 3 longer than one cycle for personal scheduling conflicts, or if a delay in menses occurred in cycle 2, suggesting persistent impact of cycle 1 treatment, but if spacing was longer than 3 months, rescreening was required (n=0). Additionally, our study procedures overlapped with the first 6 months of the COVID-19 pandemic resulting in suspended procedures for several months as directed by state and institutional mandates. Study subjects were compensated for their time.
All participants underwent cycle monitoring using transvaginal ultrasound and blood sampling for progesterone (P4), estradiol (E2), and luteinising hormone (LH). We started participant monitoring during treatment cycles 1 and 3 on day 6–8 with every other day visits until visualisation of a dominant follicle measuring ≥15 mm in at least one dimension was observed and then daily until evidence of follicle rupture (>50% reduction of mean size or complete disappearance of follicle) or for up to 7 days .4 6 7 13–15 A follicle size >15 mm triggered immediate UPA dosing. Study staff directly observed participants taking the study drug.
Participants in the high BMI cohort underwent randomisation (1:1) to UPA 30 mg or 60 mg for cycle 1 and then the other dose for cycle 3 using a computer-generated randomisation scheme maintained by the OHSU research pharmacy. Participants and investigators were not masked to allocation. The normal BMI group received a single dose of UPA (30 mg) during cycle 1 only.4 7 13
We offered participation in an optional PK substudy to all participants. For those who agreed, we obtained serum samples through an indwelling catheter before (0 hour) and 0.5, 1, 1.5, 2, 3, 4, 24, 48, 72, 96, and 120 hours following dosing. We generated PK parameters by non-compartmental methods using WinNonlin v6.3 (Certara L.P. (Pharsight), St Louis, MO, USA). Maximum drug concentration (Cmax) and time to maximum Cmax (Tmax) are observed values. The area under the curve (AUCinf) was calculated from time 0 to 120 hours using the linear trapezoidal rule and then extrapolated to infinity. Drug half-life (T1/2), apparent oral clearance (CL/F), and apparent volume of distribution (Vz/F) were generated using standard PK calculations (T1/2 =0.693/λz where λz is the terminal elimination rate constant; CL/F=dose/AUCinf; Vz/F=CL/λz). PK parameters for UPA and NDM-UPA were analysed using descriptive statistics and generated for all of the doses and their respective BMI groups.
The total number of subjects in the control group was chosen as a convenience sample (target of 10 with an additional 20% added in case of drop out), but this number of subjects is adequate to ensure that pharmacodynamic and PK results are similar to published values in individuals with a normal BMI. The sample size for the pairwise comparison in the higher BMI cohort was exploratory. We chose a difference of 20% as this would be clinically significant; a sample size of 73 would achieve 80% power to detect a 20% difference in the proportion of cycles that demonstrate a delay in follicular rupture (yes/no) using the McNemar test with a Bonferroni-adjusted 2.5% significant level. We summarised and compared demographics and baseline clinical characteristics using descriptive statistics as mean (SD) and count (%). Our primary outcome was the difference in the proportion of cycles with no follicle rupture 5 days post-dosing (yes/no) between dosing groups (30 mg vs 60 mg) in the BMI ≥30 kg/m2 cohort, and it was compared using a Fisher’s exact test. Participant data were included even if they only completed one of the two treatment cycles for this analysis. Given the crossover design, we also performed paired statistics to identify any discordant pairs (McNemar’s test). The secondary outcome, which is the timing (day) of follicle rupture between dosing groups in the BMI ≥30 kg/m2 cohort, was assessed using a Kaplan-Meier survival curve and a long-rank test. Timing of follicle rupture censored at day 5. We also calculated and descriptively compared outcomes with our normal BMI control group. If the date of follicle rupture was unclear by ultrasound imaging (eg, collapse was seen, but reduction of size was <50%), we utilised serum hormone levels to adjudicate the day of rupture. Two investigators independently reviewed these cycles while being masked to dosing, and if a disagreement occurred, a third investigator was engaged.7
Samples were stored at −80°C and batch analysed by the Endocrine Technologies Core (ETC) at the Oregon National Primate Research Centre (ONPRC, Beaverton, Oregon (https://www.ohsu.edu/onprc/endocrine-technologies-core). Serum UPA and NDM-UPA were analysed by liquid chromatography with tandem mass spectrometry (LC-MS/MS) and serum E2, P4, and LH were analysed by a Roche Cobas e411 chemiluminescence-based automated immunoassay platform (Roche Diagnostics, Indianapolis, IN, USA). The sensitivities of the E2, P4, and LH assays for the Roche e411 are 5 pg/mL, 0.050 ng/mL, and 0.1 mIU/ml, respectively. The intra- and inter-assay variation with the Roche e411 in the ETC is consistently <7% for all assays. Quality control sample analyses were repeated before each assay run.
UPA and NDM-UPA concentrations were simultaneously determined by ultra-high performance liquid chromatography-heated electrospray ionization-tandem triple quadrupole mass spectrometry (LC-MS/MS) on a Shimadzu Nexera-LCMS-8050 instrument (Shimadzu Scientific, Kyoto, Japan) (see online supplemental appendix 1). The UPA and NDM-UPA assays were developed and validated largely following US Food and Drug Administration (FDA) guidelines for bioanalytical method validation (FDA Bioanalytical Method Validation Guidance for Industry, 2018) by assessing specificity, stability, precision, accuracy, extraction efficiency (recovery), calibration curve, sensitivity, and reproducibility. The lower limit of quantification for both UPA and NDM-UPA was 0.19 ng/mL. Samples with concentrations above 200 ng/mL were re-analysed after 1:5 dilution in 0 standard. Data processing and analysis were performed using LabSolutions Software, V5.72 (Shimadzu). Intra-assay coefficient of variation (CV) for UPA ranged from 3.2–14.3% with an inter-assay CV of 6.3% (n=5 assays). Intra-assay CV for NDM-UPA ranged from 2.9–7.4% with an inter-assay CV of 4.5% (n=5 assays). Accuracy was 104.8% for UPA and 106.5% for NDM-UPA.
Supplemental material
Results
Due to the COVID-19 pandemic, we stopped enrolment after enrolling a total of 64 women (12 BMI <25 kg/m2, 52 BMI >30 kg/m2) (figure 1), which is nine fewer than our estimated sample size. The total number of individuals who completed an intervention cycle for those with BMI >30 kg/m2 was 49 who received UPA 30 mg and 46 who received UPA 60 mg. In other words, nine individuals did not complete the third study cycle (their second dosing cycle) for a variety of reasons (see figure 1). Pairwise comparison revealed no pattern of discordant pairs (McNemar p value 0.38). Twelve controls received a UPA dose of 30 mg (figure 1).
CONSORT (Consolidated Standards Reporting Trials) flow diagram for the study. UPA, ulipristal acetate.
Enrolled participants were similar in baseline demographics except for differences in BMI and weight between the control and the >30 kg/m2 BMI cohort (table 1); mean (SD) BMI by group was: controls 22.5 (1.4) kg/m2, group 1 37.9 (6.7) kg/m2, and group 2 39.3 (5.4) kg/m2. No study-related adverse events were experienced.
Characteristics of study participants
The intent-to-treat analysis of our primary outcome—the proportion of cycles that achieved at least 5 days without evidence of follicle rupture post-UPA dosing—was similar between UPA dosing groups in the BMI >30 kg/m2 cohort (Fisher’s exact test p=0.43). Almost everyone in the entire study achieved a delay of 5 days or more before follicle rupture post UPA dosing with either dose (UPA 30 mg: 47/49 (96%), UPA 60 mg: 42/46 (91%), controls 12/12 (100%)). Of note, we identified no cycles that needed adjudication to determine the date of follicle rupture utilised for this analysis.
Additionally, we performed a per-protocol analysis for the >30 kg/m2 BMI cohort participants and identified six cycles (UPA 30 mg n=2, UPA 60 mg n=4) where drug dosing clearly occurred too late, after LH rise. When excluding these cycles, all study participants in either dosing cycle (100%) achieved a delay of 5 days or more before follicle rupture. The secondary outcome of timing (day) of follicle rupture between dosing groups was also not significantly different between UPA dosing groups (log-rank test p=0.35) (figure 2).
Kaplan-Meier curve for time to rupture following UPA 30 mg (blue line) and UPA 60 mg (red line) at time zero (intent to treat analysis). The study groups were similar in their time to follicle rupture (log-rank test p = 0.35). Time to rupture was censored at day 5 . EC, emergency contraception.
PK parameters were obtained for all 12 controls. For the higher BMI cohort, 18 participants underwent PK sampling with 30 mg dosing and 17 participants with 60 mg dosing. The concentration versus time curves for UPA and its active metabolite, NDM-UPA, are shown in figure 3 for all three cohorts (normal BMI control, UPA EC 30 mg; BMI >30 kg/m2, UPA EC 30 mg; BMI >30 kg/m2, UPA EC 60 mg). In subjects dosed with 30 mg, the exposures (maximum serum concentration and area under the curve) of UPA were ~twofold higher in the BMI >30 kg/m2 cohort compared with the control BMI cohort. The exposures of UPA in the BMI >30 kg/m2 cohort were proportionally higher in the 60 mg group compared with 30 mg group. The apparent oral clearance (CL/F) of UPA was observed to be ~twofold lower in the BMI >30 kg/m2 cohort, compared with the normal BMI cohort. No differences were noted in apparent volume of distribution (Vz/F) or time to maximum concentration for UPA (approximately 1 hour). Tracking with the observed changes in CL/F, half-life (T1/2) of UPA was twofold higher in the BMI >30 kg/m2 cohort compared with the normal BMI cohort. For the active metabolite, NDM-UPA, findings similar to the parent drug, UPA, were noted.
Pharmacokinetic profiles of parent drug (UPA) (panel A) and metabolite (NDM-UPA) (panel B). Corresponding pharmacokinetic parameters for parent drug and metabolite are shown in the table below panels A and B. Mean and standard deviation are shown for each of the groups. AUCinf, area under the curve; BMI, body mass index; CL/F, apparent oral clearance; Cmax, maximum drug concentration; NDM-UPA, N-monodemethyl-ulipristal acetate; PK, pharmacokinetic; T1/2, drug half-life; Tmax, time to maximum Cmax; UPA, ulipristal acetate; Vz/F, apparent volume of distribution.
Discussion
We designed this study to gain a further understanding of UPA EC’s capacity to maintain a therapeutic effect in individuals of higher weight and BMI. A prior population-based study identified that the effectiveness of UPA might be affected by weight.5 Our results demonstrate that UPA EC has a very consistent ability to delay ovulation when dosed before the LH surge in individuals of varying BMI, even at the standard dose of 30 mg. In fact, 100% of participants dosed before the LH surge experienced a 5-day delay. We also demonstrated that the UPA serum levels, as well as its active metabolite, do not appear to be adversely affected by weight or BMI. Our findings have clinical importance as individuals of a higher weight and BMI are now a majority of the population in the USA,16 and there is a growing prevalence worldwide. It is critically important for us to know how contraceptive methods, especially emergency therapies, work in a diverse population.
Few studies exist regarding the effectiveness of UPA in individuals of varying weight and BMI. In fact, the work by Glasier et al is the only one with a substantial population of individuals with BMI >30 kg/m2 and that study was unable to definitively discern whether BMI adversely affects UPA effectiveness.5 Their results showed that it is just as likely that there is no impact of weight or BMI on UPA given the lack of statistical significance. A subsequent systematic review found a twofold increase in UPA failure in individuals of higher BMI as compared with lower BMI, but confidence intervals were wide (OR 2.6, 95% CI 0.9 to 7.0).17
It is understandable, given the gap in evidence, our indirect evidence of effect, and the proven impact of weight on oral LNG EC, that the scientific and medical community may remain concerned. Nonetheless, UPA and LNG have very different mechanisms of action and clinical pharmacology, but obesity is a wildcard in how it impacts the overall pharmacology of drugs. UPA is metabolised by CYP3A, an enzyme whose expression and activity is well documented to be down regulated in higher BMI. Our PK results are consistent with an environment where down regulation is likely occurring and thus causing lower drug clearance; we found greater exposures (AUC) in our cohort of higher BMI and appropriate increases with a higher dose. Our PK results are in contrast to a prior PK study of UPA in individuals of varying BMI which showed that the AUCs were very similar between cohorts of normal and obese BMI.8 We did employ a longer PK sampling schema which provides a more accurate depiction of AUC and calculated PK parameters (168 hours vs 48 hours), but results of both studies consistently demonstrate no adverse effect of weight or BMI on the PK parameters of UPA. Additionally, we also confirmed that the active metabolite of UPA, NDM-UPA, is not adversely affected by weight or BMI. To our knowledge, this is the first report of analysing NDM-UPA in a cohort of individuals with a higher BMI and weight which, given the known effect of obesity on drug metabolism, was critical to measure in order to have further assurance regarding our pharmacodynamic findings.
Our study has several limitations. We utilised pharmacodynamic and PK outcomes as surrogates for EC effectiveness instead of pregnancy. Pregnancy does not occur without ovulation, so this surrogate biomarker is directly related. We utilised standard, proven methodology by confirming ovulatory status in participants before study enrolment with an elevated P4.4 7 13–15 We did not have participants undergo a baseline cycle with follicular or additional hormonal monitoring beyond the single P4 level. It is possible that we may still have inadvertently enrolled individuals with ovulatory dysfunction. However, we did follow subjects through their intervention cycles to watch and identify the optimal time for dosing or, in a few cases, to stop cycle monitoring and not dose study drug because we did not see a dominant follicle developing. We also did not evaluate the entirety of the luteal phase in this study because this is not an outcome of interest given UPA’s known mechanism of action, which is to delay or stall ovulation for at least 5 days. Most of our participants had yet to ovulate post-dosing at the end of the maximum time for monitoring (7 days); thus, monitoring the entire luteal phase would have added unnecessary cost to the project and additional invasive study procedures for participants. We considered but chose not to mask the allocation to UPA dose as encapsulation of the drug might have affected the PK parameters and we wanted to be able to compare our findings to Praditpan et al.8 Finally, we did not achieve our a priori sample size. Unfortunately, the COVID-19 pandemic had an impact on our efforts to achieve our original sample size and could no longer fiscally support continuing study procedures. Our original sample size was based on what we felt would be an important clinical difference if one existed (20%), but given we found no difference between groups, we would need to have enrolled thousands of individuals to identify whether a very small difference exists, and this was not feasible or clinically relevant. While we may have missed a very small difference between groups, we have clearly demonstrated that UPA causes a very consistent effect of delaying ovulation when dosed in the optimal cycle window before the LH surge—100% of participants, regardless of BMI or UPA dose.
Our findings support the continued use of the standard dose of UPA for EC, 30 mg, regardless of a person’s weight or BMI. We recognise that access to UPA is limited as it is available only by prescription. We need to be vigilant in our practice of advanced provision and timely access, especially as the evidence base is growing regarding the adverse impact of weight on LNG EC.5–7
Data availability statement
Data are available upon reasonable request. Investigators involved in this proposal are aware of and agree to abide by the principles for sharing research resources as described by NIH in 'Principles and Guidelines for Recipients of NIH Research Grants and Contracts on Obtaining and Disseminating Biomedical Research Resources'.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by Oregon Health & Science University IRB #16291. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
Grant support for this research is from NIH R01 HD089957 (PI Edelman), support for JDH, JTJ, and the ONPRC Endocrine Technologies Core is from NIH Grant P51OD011092, and the Oregon Clinical and Translational Research Institute (UL1TR002369) for access and use of REDCap electronic data capture system.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors AE is the guarantor for this work. All of the authors were involved in the reporting of the work. AE, JH, GC, JJ, and KB also designed and conducted the work. JL designed and supported the analysis plan and lead the analyses of the data. David Archer critically reviewed the study protocol and helped to conduct the study and served as a site principal investigator. DE, SB, and SK helped design, test, and conduct the novel assays for ulipristal acetate and its metabolite.
Funding This study was funded by Oregon Clinical and Translational Research Institute (UL1TR002369)NIH (P51OD011092, R01 HD089957)
Competing interests A Edelman: AE reports travel reimbursement from ACOG, WHO, CDC. Honoraria from Gynuity for committee activities. AE receives royalties from Up to Date, Inc. Oregon Health & Science University receives research funding from OHSU Foundation, Gates, Merck, HRA Pharma, and NIH where AE is the principal investigator. J Jensen has received payments for consulting from Abbvie, Cooper Surgical, Bayer Health care, Merck, Sebela, and TherapeuticsMD. OHSU has received research support from Abbvie, Bayer Health care, Daré, Estetra SPRL, Medicines360, Merck, and Sebela. These companies and organisations may have a commercial or financial interest in the results of this research and technology. These potential conflicts of interest for A Edelman and J Jensen have been reviewed and managed by OHSU. None of these have direct conflict with this manuscript. D Archer: DFA grant support to Eastern Virginia Medical School from AbbVie, Bayer Healthcare, Dare, Mithra, Myovant Sciences, and ObsEva. Consulting fees from Agile Therapeutics, Mithra, ObsEva and TherapeuticsMD. Stock options: Agile Therapeutics and InnovaGyn, Inc. No other authors have any potential conflicts of interest. K Bond reports travel reimbursement from ACOG.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.