Abstract
Objectives
To synthesize published literature on POP effectiveness and efficacy.
Study design
We searched PubMed Central, PubMed, and the Cochrane library through March 07, 2022. We included articles written in English reporting a Pearl Index or life table rate for pregnancy. We excluded articles only assessing formulations that: were never marketed globally, are only sold in combination with estrogen, are currently sold only for noncontraceptive purposes, or were not given to participants continuously. Four researchers independently extracted data and two analyzed data using Excel and R.
Results
We included 54 studies. Among studies at low or moderate risk of bias, the median Pearl Index rate (the failure rate during typical use) was 1.63 (range 0.00–14.20, IQR 4.03) and the median method failure Pearl Index rate (the failure rate during perfect use) was 0.97 (range 0.40–6.50, IQR 0.68). Excluding the newer formulations, Desogestrel and Drospirenone, which are closer to combined oral contraceptives in that they prevent pregnancy by inhibiting ovulation, the median Pearl Index rate is 2.00 (range 0.00–14.12, IQR 2.5) and the median method failure Pearl Index rate is 1.05 (range 0.00–10.90, IQR 1.38).
Conclusions
Among studies at low or moderate risk of bias, the median Pearl Index rate during typical POP use was much lower than currently estimated (7.00), while the median perfect use rate was similar to current estimates.
Implications
Future research should investigate the possibility that POPs may be much more effective during typical use than currently believed.
Keywords
1. Introduction
Moving oral contraceptive pills—combined oral contraceptives (COCs) or progestin-only pills (POPs) —to over-the-counter (OTC) status in the United States (US) could increase accessibility for individuals encountering barriers to getting a prescription [
[1]
,[2]
]. At the time of writing, the United States Food and Drug Administration (FDA) is currently reviewing the first-ever application for an OTC POP product which contains .075 mg Norgestrel [[3]
]. A large coalition of prominent clinicians, researchers, and reproductive health, rights, and justice organizations has long focused on making POPs available OTC in the US because POPs have few contraindications and would therefore be appropriate for a wide range of people [HRA Pharma. Perrigo's HRA Pharma submits application to FDA for first-ever OTC birth control pill. n.d. https://www.hra-pharma.com/articles/perrigos-hra-pharma-submits-application-to-fda-for-first-ever-otc-birth-control-pill-66 (accessed October 7, 2022).
[4]
].OCs OTC Working Group. Statement of purpose. n.d. https://ocsotc.org/statement-of-purpose/ (accessed March 1, 2022).
Although there is interest among US women in an OTC POP product [
[5]
], POP users constitute only 4% of contraceptive pill users [[6]
], and this low user rate may be due to clinicians’ hesitation to prescribe POPs based on their views of the pill's effectiveness. One study assessing how evidenced-based information influences clinicians’ thoughts about an OTC oral contraceptive found that before receiving information, 69% of clinicians did not support an OTC POP, with 17% citing “less effective pill formulation” as a reason [[7]
]. However, a 2013 systematic literature review of randomized controlled trials of progestin-only pills concluded there was insufficient evidence to compare POPs to COCs [[8]
].- Grimes DA
- Lopez LM
- O'Brien PA
- Raymond EG.
Progestin-only pills for contraception.
Cochrane Database Syst Rev. 2013; CD007541https://doi.org/10.1002/14651858.CD007541.pub3
Efficacy rates refer to failure rates only when the pill is taken as directed (perfect use), whereas effectiveness refers to failure rates during typical use (which includes perfect use as well as incorrect and inconsistent use) [
[9]
]. Although pill effectiveness rates vary based on the population using the method [[10]
], it is estimated that in the US 7% of individuals using oral contraceptives will have an unintended pregnancy during their first year of typical use [[9]
]. However, this estimation does not distinguish between COCs and POPs and may be a better reflection of the failure rate of COCs because they are more commonly used in the US [[9]
]. It is thought that this failure rate may be slightly higher for POPs because a common belief is that POPs lose effectiveness when not taken within a rigid timeframe of 24 hours after the previous pill with a strict 3-hour window [[11]
,[12]
], although little clinical data exist to support this belief [[8]
,- Grimes DA
- Lopez LM
- O'Brien PA
- Raymond EG.
Progestin-only pills for contraception.
Cochrane Database Syst Rev. 2013; CD007541https://doi.org/10.1002/14651858.CD007541.pub3
[13]
].In addition to a lack of evidence for the 3-hour window, there are many different POP formulations, including two newer formulations—Desogestrel and Drospirenone—that have been shown to inhibit ovulation even after long delays in pill intake (12-hour delays for Desogestrel and 24-hour delays for Drospirenone) [
[14]
,[15]
] so the 3-hour window recommendation is likely not applicable to them. One study has also found that a 6-hour delay or a single missed POP containing Norgestrel 0.075 mg appears to not negatively impact contraceptive efficacy [[16]
] Current estimates of the pregnancy rate for the first year of use among perfect users is based on the lowest reported pregnancy rate as well as pregnancy rates reported in recent studies—for COCs, the rate of pregnancy among perfect users is estimated to be 0.30% and although this rate is cited as the rate for all oral contraceptives, the pregnancy rate among perfect users of POPs is unknown [[11]
]. As of 2018, the lowest reported pregnancy rate for POP use was 1.1% [[11]
].Contraceptive effectiveness and efficacy rates can both be measured by the Pearl Index and the life table. The Pearl Index calculates an effectiveness rate by dividing the number of total pregnancies (contraceptive failures) by 100 person-years of exposure to the contraceptive method [
[17]
]. A Pearl Index that measures efficacy (often referred to as a method failure Pearl Index) only includes pregnancies resulting from a method failure among perfect users. Due to its ease of calculation, the Pearl Index has been reported more frequently in the literature than life table rates, but a significant limitation is that estimates can vary with study length. For longer durations of pill use, the Pearl Index tends to underestimate failure rates since pregnancies are more common at the beginning of pill use and study participants more likely to conceive become pregnant early and withdraw from the study, leaving a group of participants less likely to conceive [[18]
]. This limitation is eliminated by a life table analysis, which estimates monthly probabilities of failure and cumulative probabilities over time.Previous reviews of the effectiveness and efficacy of POPs have focused on few formulations [
[19]
,[20]
] and there is insufficient information among randomized trials to make comparisons between different POPs [[8]
]. This review aims to expand upon previous reviews by including a range of study types reporting effectiveness and efficacy rates of various POP formulations, while recognizing the limitations of nonrandomized trials. Our findings can help ensure that policymakers, reproductive health advocates, and the general public have the necessary information, backed by clinical evidence, to make decisions about an OTC POP product.- Grimes DA
- Lopez LM
- O'Brien PA
- Raymond EG.
Progestin-only pills for contraception.
Cochrane Database Syst Rev. 2013; CD007541https://doi.org/10.1002/14651858.CD007541.pub3
2. Material and methods
2.1 Literature search and study selection
We searched PubMed Central, PubMed, and the Cochrane library for articles and reports written in English on the effectiveness or efficacy of POPs through March 07, 2022. The search did not include limits by study publication type, date, or study design. We did not include overall findings from literature reviews but searched the references of reviews and included articles with original or primary research. Four researchers screened the titles and abstracts of articles for eligibility. We included randomized and nonrandomized studies (with or without a control group) with data on pregnancy rates among users of at least one POP formulation currently or previously sold in any country. We excluded articles assessing formulations that: were never marketed globally, are only sold in combination with estrogen, are currently sold only for noncontraceptive purposes, or were not given to participants continuously (except for the newer Drospirenone-only pills which is the only POP product sold with placebo pills in a pack). We also excluded articles if they did not report information on the duration of person-time used for estimations of effectiveness. See Appendix A for details of our search process and Appendix B for search terms.
2.2 Data extraction and calculations
We extracted the following data from each study: first author, title, year of publication, study location, participant characteristics, study design features (n, duration, randomization), POP formulation and dosage, loss-to-follow-up, number of total pregnancies, number of pregnancies attributed to user error and method failure, Pearl Index, method failure Pearl Index, Pearl Index rates adjusted for patient characteristics or behaviors, and life table data. We converted dosage units to milligrams and summed up dosages taken more than once a day as a single daily dosage. We also calculated Pearl Index rates if sufficient data were available and compared them to reported rates. In our analyses, we used Pearl Index rates reported by studies and only used calculated rates if no single Pearl Index rate was reported. We included aggregated Pearl Index rates that combined rates from multiple formulations or studies.
As failure rates tend to be high at the start of studies and decrease over time [
[18]
], we also extracted data on study duration to analyze effectiveness and efficacy rates by study length. Studies reported study duration in different units (cycles, months, years), which we converted into months. (since 12 months is equivalent to 13 cycles, we calculated that there are .92 months in a cycle). If study length was not reported, we estimated study duration based on the longest reported cycle or month of treatment completed. If we could not estimate study length, we used average length of participation in the study, if available. In our analysis of study duration and Pearl Index rates, we excluded one Pearl Index rate that was pooled from two studies of different durations [[21]
].Since person-time depends on both duration of use and number of participants, we also looked at Pearl Index rates by study size. We recorded the number of participants who started treatment in each study arm. For retrospective studies, we extracted the number of participants included in the analysis, if available.
Four researchers independently extracted data and placed data in an Excel spreadsheet.. Two researchers used Excel and R for data analysis and visualizations [
[22]
].R Core Team (2022). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
2.3 Assessing risk of bias
We assessed risk of bias for included studies using the Cochrane risk-of-bias tool for randomized trials (RoB 2) [
[23]
], the Cochrane Risk Of Bias In Non-Randomized Studies of Interventions (ROBINS-I) tool [- Sterne J
- Savović J
- Page M
- Elbers R
- Blencowe N
- Boutron I
- et al.
RoB 2: a revised tool for assessing risk of bias in randomised trials.
BMJ. 2019; 366l4898https://doi.org/10.1136/BMJ.L4898
[24]
] for nonrandomized studies with a comparator, and a newly developed tool for assessing bias in estimation of contraceptive failure rates in studies lacking a comparator. We included data from a single arm of a study when a POP was compared with a non-POP method or compared with a POP formulation not given continuously to participants. We only extracted pooled data when studies compared the impact of the same formulation on different groups of participants (as opposed to comparing two or more different formulations). To assess risk of bias for these single arm estimates of contraceptive failure, we created a tool adapted from existing Cochrane tools that included domains for assessing bias relevant to valid estimation of contraceptive failure rates [- Sterne JA
- Hernán MA
- Reeves BC
- Savović J
- Berkman ND
- Viswanathan M
- et al.
ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions.
BMJ. 2016; 355i4919https://doi.org/10.1136/BMJ.I4919
[25]
]. We assessed all studies using the appropriate risk of bias tool by two researchers, who judged all studies to be in one of three categories: low risk, moderate risk, or high risk of bias. Across all tools, we judged studies based on the most severely rated domain. If we assessed three or more domains to be at moderate risk of bias, then we judged the study's overall risk of bias to be high. If a domain did not have enough information for us to make an assessment, we assessed the domain as moderate. If more than one domain did not have sufficient information to make an assessment, we excluded the study from our analysis. See Appendix C for our tool to assess risk of bias in noncomparative studies and Appendix D for tables summarizing our risk of bias judgments for all studies.2.4 Data synthesis and analysis approach
Our main outcome was median effectiveness and efficacy rates reported by studies assessed to be at low or moderate risk of bias, although we also conducted a sensitivity analysis by calculating the median effectiveness and efficacy rates across all studies. We also analyzed effectiveness and efficacy rates by study duration, size, and formulation. We conducted additional sensitivity analysis by removing rates for Desogestrel and Drospirenone, as these newer formulations are different from other POP because they reliably inhibit ovulation after delayed pill intake [
[14]
,[15]
]. In addition, given the FDA's current review of an application for an OTC product containing Norgestrel 0.075 mg, we also looked at Pearl Index rates reported by studies analyzing that particular dosage and formulation. As most Pearl Index rates were not accompanied by confidence intervals, we report ranges and interquartile ranges to provide information on the spread of data. If a study was aiming to compare different groups of participants (e.g., lactating versus nonlactating) and reported separate Pearl Index rates, these rates are included in our main findings. However, if studies reported rates that took into account participant behaviors or characteristics (e.g., taking into account that their sample included some lactating participants and reporting analyses with this group excluded from calculations) in addition to overall Pearl Index rates, we did not include these adjusted or stratified rates in our overall calculations but report these separately.3. Results
3.1 Included studies
Fifty-four studies met our eligibility criteria [
Study location was not reported in the article so the country where researchers were based are listed instead. Korver T is listed as the corresponding author. The study was written by a collaborative study group.
[21]
,26
, 27
, 28
, 29
, 30
, 31
, 32
, 33
, 34
, 35
, 36
, 37
, 38
, 39
, 40
, 41
, 42
, 43
, 44
, 45
, 46
, 47
, 48
, 49
, 50
, 51
, 52
, 53
, 54
, 55
, 56
, 57
, 58
, 59
, 60
, 61
, 62
, 63
, 64
, 65
, 66
, 67
, 68
, 69
, 70
, 71
, 72
, 73
, 74
, 75
, 76
, 77
, 78
]. Table 1 displays descriptive information for each article. Included studies were published between 1966 and 2019, with almost half published in the 1970s [[26]
,[31]
,[32]
,34
, 35
, 36
,39
, 40
, 41
, 42
,[45]
,[47]
,[48]
,[52]
,55
, 56
, 57
, 58
, 59
, 60
, 61
,[64]
,[66]
,[68]
,[71]
,[72]
,[75]
,[76]
]. Forty-four studies were conducted primarily in either Europe or North America [[21]
,27
, 28
, 29
,31
, 32
, 33
, 34
, 35
, 36
, 37
, 38
, 39
, 40
,[42]
,[43]
,45
, 46
, 47
, 48
, 49
, 50
, 51
, 52
,57
, 58
, 59
, 60
, 61
,[63]
,[67]
,[68]
,[70]
,[72]
,74
, 75
, 76
, 77
, 78
], and all were prospective studies with the exception of three that were retrospective [[70]
,[74]
,[79]
]. Seven were randomized trials [[21]
,[43]
,[44]
,[49]
,[57]
,[62]
,[66]
], 15 were nonrandomized comparative studies [[34]
,[39]
,[48]
,50
, 51
, 52
, 53
, 54
,[64]
,[65]
,[67]
,69
, 70
, 71
,[74]
], and 32 were noncomparative studies [26
, 27
, 28
, 29
, 30
, 31
, 32
, 33
,35
, 36
, 37
, 38
,40
, 41
, 42
,45
, 46
, 47
,[55]
,[56]
,58
, 59
, 60
, 61
,[63]
,[68]
,[72]
,[73]
,75
, 76
, 77
, 78
]. All studies were peer-reviewed except two—one was described in a letter published in the correspondence section of a peer-reviewed journal [[29]
] and one was an abstract published in a conference proceeding [[52]
].Table 1Description of studies reporting effectiveness and/or efficacy rates of progestin-only pills
| First author | Year published | Study location (country) | Progestin(s) | Study design** | Risk of bias assessment | Participant characteristics |
|---|---|---|---|---|---|---|
| Apelo [26] | 1973 | The Philippines | Levonorgestrel | Noncomparative study | High | Age Range: 17–37 years Mean: 26 years Fertility All participants had at least one pregnancy Mean 3.5 Mean interval between last delivery and start of medication was 8 months with a range of 1–55 months Weight Range: 75–149 lbs Mean 101.6 lbs |
| Archer [37] | 2015 | Czech Republic, Germany, Hungary, Poland, and Romania | Drospirenone | Noncomparative study | Moderate | Age Range: 18–46 years Mean: 28.7 years Age group: 79.8% were 35 years old or younger Fertility 42.8% had a previous delivery Previous Contraceptive Use 63.8% had "prior treatment with sex hormones and modulators of the genital system" Race 99.6% Caucasian Weight BMI range: 16–38 Mean BMI: 23 |
| Aznar-Ramos [48] | 1971 | Mexico | Chlormadinone acetate | Nonrandomized comparative study (comparing two different divided dosages) | Moderate | Age Range: 19–35 years |
| Bernstein [59] | 1972 | United States | Chlormadinone acetate | Noncomparative study | High | Age Maximum: All participants were under 40 years |
| Bisset [70] | 1990 | United Kingdom | Ethynodiol diacetate Levonorgestrel Norethisterone Norgestrel | Nonrandomized comparative restrospective study | High | Lactating or postpartum 6% of participants on ethynoldiol diacetate lactating 26% of participants on levonorgestrel lactating 2% of participants on norethisterone lactating 23% of participants on norgestrel lactating |
| Board [75] | 1971 | United States | Norethindrone | Noncomparative study | High | Fertilitiy Each participant of proved fertility Marital Status All participants were living with their husbands Previous Contraceptive Use All had taken either combination or sequential oral contraceptives 31.8% of participants had not been using OCs for at least 2 months prior to the study Most participants started norethindrone immediately after discontinuing their previous oral contraceptive |
| Board [76] | 1976 | United States | Norethindrone | Noncomparative study | High | Fertility Each participant of proved fertility Marital Status All participants were living with their husbands Previous Contraceptive Use Most had taken either combination or sequential oral contraceptives 29% of participants did not take an oral contraceptive for 2 months prior to beginning the study |
| Broome [77] | 1990 | United Kingdom | Ethynodial diacetate Norethisterone Levonorgestrel | Noncomparative restrospective study | High | Age Majority: 59% of 358 women were 31–40 years old Lactating or postpartum Excluded from analysis (not included in the 358) |
| Butler [78] | 1969 | United Kingdom | Chlormadinone acetate | Noncomparative study | Moderate | Age Maximum: 34 years Fertility Each participant had at least one living child Marital Status All participants were married Previous Contraceptive Use No OCs used in the 2 months before the study |
| Canto [27] | 1989 | Mexico | Norgestrel | Noncomparative study | High | Age Minimum: 18 years Majority: 56% were 20–29 years Mean: 26.1 years Fertility All participants had given at least one live birth Mean: 3.5 births Lactating or postpartum All were breast feeding on admission 43.5% were <6 weeks postpartum 56.5% were 6–26 weeks postpartum 83% were still breastfeeding at end of study Previous Contraceptive Use (month before the study) 38.5% participants were not using any method 26.5% used oral contraceptives 17% used injectables 11.5% used an IUD 5% used withdrawal/rhythm 1.5% used a condom |
| Cerais [73] | 1991 | Sudan | Norgestrel | Noncomparative study | High | Age Mean: 26.3 years Fertility All participants had at least one live birth Mean: 2.3 live births Lactating or postpartum All were breastfeeding on admission 177 women were between 42 day and 26 weeks postpartum 23 women were less than 42 days postpartum Previous Contraceptive Use (month before the study) 61% of participants not using any contraception immediately prior to admission or conception 32% of those using a method were using an oral contraceptive 34.5% reported ever having used an oral contraceptive prior to the study |
| Christie [28] | 1969 | Jamaica Mexico United Kingdom United States | Chlormadinone acetate | Noncomparative study | High | – |
| Cox [29] | 1969 | United Kingdom | Norgestrel | Noncomparative study | High | Fertility Mean parity: over 2 |
| Dunson [30] | 1993 | 22 medical facilities in Africa, Latin America and the Caribbean | Norgestrel | Noncomparative study | High | Age Mean: 25.7 (±4.9) Fertility Mean number of live births: 2.5 (±1.7) Lactating or postpartum 74% entered the study when they were between 1 and 2 months postpartum 56.6% breastfeeding at admission with no supplementation: 43.4% breastfeeding at admission with supplementation Previous Contraceptive Use 62% of participants had not used any contraception in the month before the study began Of women who had used a method, oral contraception was the most common (23.5%) |
| Eckstein [31] | 1972 | United Kingdom | Norgestrel | Noncomparative study | High | Age Maximum: All participants were under 40 years Majority: 24–35 years Fertility All participants had at least one living child of the present marriage |
| Foss [32] | 1975 | United Kingdom | Norgestrel | Noncomparative study | Moderate | Age Range: 19–41 years Fertility All participants had more than one child Previous Contraceptive Use This study included participants who wished to continue using Norgestrel from Foss study, so all had previously used this pill. |
| Foss [33] | 1968 | United Kingdom | Norgestrel | Noncomparative study | High | Age: Age group: 92% of participants were between 17 and 40 years old; 8% were between 41 and 18 years old Fertility 88% of participants were of proven fertility Range of number of children: 0–9 |
| Hawkins [34] | 1977 | United Kingdom | Chlormadinone acetate Norethisterone | Nonrandomized comparative study | Moderate | Age Mean age for chlormadinone acetate: 26.1 (±6.1) Mean age for norethisterone: 24.7 (±5.1) Fertility 95% of participants parous Average parity: 1.9 (±1.5) Lactating or postpartum 71% of patients were within 3 months postpartum at start of study and a higher proportion of patients given norethisterone were more than 6 months postpartum on admission Race 76% White 16% Black 8% Asian, Latin American, mixed race |
| Heinen [35] | 1970 | Germany | Chlormadinone acetate | Noncomparative study | High | Age Average: 30 years Majority: 60% of study participants were 26 –35 years old |
| Hernandez-Torres [36] | 1970 | Puerto Rico | Norgestrel | Noncomparative study | Moderate | Age Maximum: No one was older than 36 years Fertility All had one or more previous pregnancies or abortions Lactating or postpartum All were nonlactating Previous Contraceptive Use No participants received oral contraceptive treatment for 90 days prior to the study |
| Howard [38] | 1969 | United Kingdom | Chlormadinone acetate | Noncomparative study | High | Fertility Participants not necessarily of proven fertility Lactating or postpartum 26% of patients were lactating and amenorrhoeic at the start of the study |
| Jeppsson [39] | 1970 | Sweden | Chlormadinone acetate | Nonrandomized comparative study | High | Age Majority: 83.5% were between the ages of 20–39 years Fertility 32% had never been pregnant Previous Contraceptive Use The sample included both women who had and hadn't use oral contraceptives before Other 48% of patients were upper-middle class and sought contraceptive advice at an outpatient department for private patients; 52% sought free advice on contraceptives at a public family planning facility |
| Jick [74] | 2009 | United Kingdom | Levonorgestrel Norethisterone Desogestrel | Nonrandomized comparative retrospective study | Moderate | Age Range: Minimum 13 years Fertility 71% had no prior deliveries Lactating or postpartum Evaluated the recency of delivery in users of the progestin-only pills compared to the COCs (to evaluate whether POP users were more likely to be breastfeeding) Weight BMI: (<20, 20–23, 24–27, 28+, Unknown) |
| Jubhari [40] | 1974 | United States | Quingestanol acetate | Noncomparative study | High | Age Mean: 23.1 years Fertility Most had never had a child Marital Status Most were single Race Most were White |
| Kesserü [41] | 1972 | Peru | Levonorgestrel | Noncomparative study | High | Age Range: 16–43 years Mean: 26.5 years Fertility Range number of pregnancies: 1–19 Mean number of pregnancies: 4.9 |
| Korba [42] | 1974 | Puerto Rico Unites States | Norgestrel | Noncomparative study | High | – |
| Korver [43] | 1998 | Finland Germany The Netherlands Norway Sweden United Kingdom | Desogestrel Levonorgestrel | Randomized control trial | Low | Age Range: 18–45 years Mean age: 29.6 Fertility Mean number of pregnancies: 1.6 Lactating or postpartum Desogestrel users: 30.7% breastfeeding Levonorgestrel users: 30.9% breastfeeding Previous Contraceptive Use in Previous 2 months Desogestrel users: 36.5% switched directly from another pill Levonorgestrel users: 37.9% switched directly from another pill Weight Range: All were between 80% and 130% of the ideal body weight Mean BMI: 22.8 kg/m2 |
| Lakha [44] | 2007 | China Nigeria South Africa United Kingdom | Levonorgestrel | Randomized control trial | Moderate | Age Mean age: 30.4 years Previous Contraceptive Use The majority 21 (of 23) did not use contraceptives in the previous few months Weight Mean: 58.4 kg Mean BMI: 22.4 kg/m2 |
| Laurie [45] | 1972 | Puerto Rico Unites States | Norgestrel | Noncomparative study | High | Age Mean: 23 years Fertility 87.4% of participants multigravidae Race 52.7% White |
| Lawson [46] | 1972 | Jamaica New Zealand United Kingdom | Norethisterone | Noncomparative study | High | Age Range: 16–54 years Median: 27 years Fertility 78% of participants had a previous pregnancy Lactating or postpartum 9% of participants were breastfeeding Previous Contraceptive Use 53% of participants switched directly from another oral contraceptive |
| Maqueo [47] | 1972 | Mexico | Quingestanol acetate | Noncomparative study | High | Age Mean: 29 years Fertility Mean number of previous pregnancies: 4.5 Previous Contraceptive Use 36% of participants had previously received varying doses of quingestanol acetate for other studies 61% patients had no previous oral contraceptive therapy |
| Martinez-Manatou [49] | 1967 | Mexico | Chlormadinone acetate | Randomized comparative study (comparing different doses) | High | Fertility Women of proven fertility with no more than two children Lactating or postpartum No participants were lactating |
| Martinez-Manatou [50] | 1967 | Mexico | Chlormadinone acetate | Nonrandomized comparative study (comparing lactating to nonlactating group) | High | Age Maximum: Less than 36 (at least among the nonlactating group) Lactating or postpartum In one group, all women (100) were lactating and were between 1 and 15 months postpartum; the other group consisted of nonlactating participants |
| Martinez-Manatou [51] | 1966 | Mexico | Chlormadinone acetate | Nonrandomized comparative study (comparing cyclical to continuous pill taking regimen) | High | Age Maximum: Less than 36 Fertility All participants of proven fertility Lactating or postpartum No participants were lactating |
| McQuarrie [52] | 1972 | United States | Norethindrone | Nonrandomized comparative study | High | Age Range: 16–42 years Mean: 26.4 years Fertility Parity range: 1–9 Mean: 2.5 children delivered |
| Mears [53] | 1969 | Yugoslavia | Chlormadinone Norethisterone acetate Norgestrel | Nonrandomized comparative study | Moderate | Age Range: 18–40 years Fertility All participants of proven fertility Previous Contraceptive Use All participants took no hormones or oral contraceptives during the previous 2 months |
| Moggia [54] | 1991 | Argentina | Norgestrel | Nonrandomized comparative study | Moderate | Age Range: 18–35 years Fertility All participants had given birth 2–6 times Lactating or postpartum All participants were lactating at beginning of study |
| Moggia [55] | 1972 | Argentina | Quingestanol acetate | Noncomparative study | High | Age Range: 15–44 years Mean: 26.1 (±0.2) Fertility Mean number of prior pregnancies: 2.7 (±0.1) Lactating or postpartum 80% of participants were postpartum Weight Range: 40–100 kg Mean: 60.0 kg (±0.4) |
| Moggia [56] | 1973 | Argentina | Quingestanol acetate | Noncomparative study | High | Age Range: 15–44 years Mean: 26.1 years (±0.2) Fertility Range number of prior pregnancies: 0–9 Mean number of prior pregnancies: 2.7 (±0.1) Lactating or postpartum 53.65% of patients lactating 76% postpartum Weight Range: 40–100 kg Mean 60.1 (±0.3) |
| Palacios [21] | 2019 | Austria, Czech Republic, Germany, Hungary, Poland, Romania, Slovakia and Spain | Drospirenone Desogestrel | Randomized control trial | Low | Age Drospirenone: Range: 18–45 years; Mean: 28.9 years; Age group: 79.5% were 35 years old or younger Desogestrel: Range: 18–45 years; Mean: 28.9 years; Age group: 78% were 35 years old or younger Fertility Drospirenone:46 % had a previous delivery Desogestrel: 45 % had a previous delivery Previous Contraceptive Use 54.7% of Drospirenone users and 58.7% of Desogestrel users had "prior treatment with sex hormones and modulators of the genital system" Race 99.8% of Drospirenone users and 99.7% of Desogestrel users were Caucasian Weight Drospirenone: BMI range: 16.6–41; Mean BMI: 22.96 Desogestrel: BMI range: 15.9–38; Mean BMI: 22.82 |
| Paulsen [57] | 1974 | United States | Ethynodiol diacetate | Randomized control trial | High | Age Range: 18–39 years Mean: 20.5 years (±2.7) Fertility 5% had a previous pregnancy: 2.5% had a previous live birth Previous Contraceptive Use Majority of patients did not have prior experience with oral contraceptives |
| Postlethwaite [58] | 1979 | United Kingdom | Ethynodiol diacetate | Noncomparative study | High | Age Range: 17–48 years |
| Rice-Wray [60] | 1972 | Mexico | Levonorgestrel | Noncomparative study | High | Age Range: 18–40 years Fertility All participants of proven fertility Previous Contraceptive Use None had any steroid therapy for at least 60 days prior to initiating study. |
| Scharff [61] | 1971 | Germany | Levonorgestrel | Noncomparative study | High | – |
| Sheth [62] | 1982 | India Yugoslavia | Levonorgestrel Norethisterone | Randomized control trial | Moderate | Age Range: 18–38 Levonorgestrel users mean age: 25.7 years (±4.57) Norethisterone users mean age: 25.6 (±4.68) Previous Contraceptive Use No participants had used oral contraceptives within 28 days or long acting injectable hormonal contraceptives within 90 days of starting treatment 27.4 % of levonorgestrel users had ever used oral contraceptives 26.8% of Norethisterone users had ever used oral contraceptives |
| Shroff [63] | 1987 | United Kingdom | Ethynodiol diacetate | Noncomparative study | High | Age Range: 16–45 years Age group: 72% were 16–34 years old 28% were 35–47 years old Median: 30 years Fertility 75% experienced at least one previous pregnancy Previous contraceptive use None: 8% COCs: 48% POPs: 9% OCs: (unknown) 1% IUCD: 15% Other 19% |
| Statzer [72] | 1972 | United States | Norgestrel | Noncomparative study | High | Age Range: 15–44 years Fertility All participants demonstrated fertility. Previous pregnancies ranged from 1 to 9 Previous contraceptive use Subjects have taken no oral or injectable contraceptive for 90 days or more In some cases, subjects switched directly from Oral (norgestrel 0.5 mg and ethinyl estradiol 0.05 mg) to microdose norgestrel Race 14% White 86% Black |
| Tejuja [64] | 1974 | India | Norgestrel | Nonrandomized comparative study (comparing two doses) | Moderate | Age 50 μg Norgestrel users: 79.2% between 20 and 29 years old 75 μg Norgestrel users: 80.1% between 20 and 29 years old Fertility 50 μg Norgestrel users: >99% of participants had had at least one pregnancy 75 μg Norgestrel users: >99% of participants had had at least one pregnancy Lactating or postpartum 50 μg Norgestrel users: 27.2% had lactational amenorrhea prior to commencement of the study 75 μg Norgestrel users: 29.8% had lactational amenorrhea prior to commencement of the study Weight 50 μg Norgestrel users' average weight: 43.4 kg 75 μg Norgestrel users' average weight: 44.7 kg |
| Tyler [65] | 1968 | United States | Norgestrel | Nonrandomized comparative study (comparing two doses) | High | – |
| Vessey [66] | 1972 | Yugoslavia | Chlormadinone acetate Norethisterone acetate Norgestrel | Randomized control trial | Moderate | Age Chlormadinone acetate users' mean age: 30.4 years Norethisterone acetate users' mean age: 30 years Norgestrel users' mean age: 30.1 Fertility All participants of proven fertility Chlormadinone acetate users mean number of full term births: 1.7 Norethisterone acetate users' mean number of full term births: 1.8 Norgestrel users' mean number of full term births: 1.8 Weight Chlormadinone acetate users' mean weight: 65.8 kg Norethisterone acetate users' mean weight: 65.6 Norgestrel users' mean weight: 66.1 kg |
| Vessey [67] | 1985 | United Kingdom | Norethisterone Norgestrel Levonorgestrel Ethynodial diacetate | Nonrandomized comparative study (comparing different POP formulations) | Moderate | Age Range: 25–39 years Marital Status All participants were married Race All participants were White |
| Whyte [68] | 1973 | Canada | Norethindrone | Noncomparative study | High | Age Mean: 23.3 years Fertility 98% of participants had at least one previous pregnancy Previous Contraceptive Use All participants had previously used a type of oral contraceptive Other One third of patients were either contraindicated to estrogen or found the combined pill unacceptable due to side effects. The remaining sample had never used any oral contraceptive before and had no contraindications to a combined or progestin-only pill. |
| Zañartu [69] | 1968 | Chile | Chlormadinone acetate | Nonrandomized comparative study (comparing two groups of participants of different socio-economic statuses) | High | Age Minimum: 16 years Fertility All participants had been pregnant at least once Lactating or postpartum 110 women started use after childbirth while lactating and/or experiencing amenorrhea Other 45 women were came from families with an above-average income 345 were from low-income groups Combined oral contraceptives or sequential oral contraception was either poorly tolerated or not acceptable to all women from low-income group and in the majority (40) among women from families with an above-average income |
| Zanartu [71] | 1974 | Chile | Ethynodiol diacetate Norgestrienone | Nonrandomized comparative study (comparing different formulations among two different groups of patients -continuous use with precoital use) | High | Age Mean:28.8 Range: 18–41 Fertility Mean parity:5.5 |
^ This study reports on results from two studies; the first of which are already reported in the study by Archer et al. Only results from the second study, and any pooled results, are reported here. **Studies that did not explicitly say they were randomized are categorized as nonrandomized.
Data on study participant characteristics varied widely: age (reported by 44 studies) [
[21]
,[26]
,[27]
,30
, 31
, 32
, 33
, 34
, 35
, 36
, 37
,39
, 40
, 41
,43
, 44
, 45
, 46
, 47
, 48
,50
, 51
, 52
, 53
, 54
, 55
, 56
, 57
, 58
, 59
, 60
,62
, 63
, 64
,66
, 67
, 68
, 69
,71
, 72
, 73
, 74
,[77]
,[78]
] and fertility (reported by 40 studies) [[21]
,[26]
,[27]
,29
, 30
, 31
, 32
, 33
, 34
,36
, 37
, 38
, 39
, 40
, 41
, 42
, 43
,45
, 46
, 47
,[49]
,51
, 52
, 53
, 54
, 55
,[60]
,[63]
,[64]
,[66]
,[68]
,[69]
,71
, 72
, 73
, 74
, 75
, 76
,[78]
] were most frequently reported. Although studies reported ages differently (range, average age, maximum age, or a mix of these measures), ages ranged from 13 to 54 years. Studies also measured fertility in various ways, including number of previous pregnancies, number of living children, and number of live births. Twenty-seven studies noted that all or most participants (between 75% and 99%) had proven to be fertile [[26]
,[27]
,31
, 32
, 33
, 34
,[36]
,[39]
,[41]
,[45]
,[46]
,[49]
,51
, 52
, 53
, 54
,[60]
,[63]
,[64]
,[66]
,[68]
,[69]
,[72]
,[73]
,[75]
,[76]
,[78]
]. We included other commonly reported participant characteristics in Table 1.Thirty-six studies [
[21]
,26
, 27
, 28
, 29
, 30
, 31
,34
, 35
, 36
, 37
- Archer DF
- Ahrendt HJ
- Drouin D.
Drospirenone-only oral contraceptive: results from a multicenter noncomparative trial of efficacy, safety and tolerability.