Intervention Review

You have free access to this content

Progestin-only pills for contraception

  1. David A Grimes1,*,
  2. Laureen M Lopez2,
  3. Paul A O'Brien3,
  4. Elizabeth G. Raymond4

Editorial Group: Cochrane Fertility Regulation Group

Published Online: 13 NOV 2013

Assessed as up-to-date: 29 OCT 2013

DOI: 10.1002/14651858.CD007541.pub3


How to Cite

Grimes DA, Lopez LM, O'Brien PA, Raymond EG. Progestin-only pills for contraception. Cochrane Database of Systematic Reviews 2013, Issue 11. Art. No.: CD007541. DOI: 10.1002/14651858.CD007541.pub3.

Author Information

  1. 1

    University of North Carolina, School of Medicine, Obstetrics and Gynecology, Chapel Hill, North Carolina, USA

  2. 2

    FHI 360, Clinical Sciences, Research Triangle Park, North Carolina, USA

  3. 3

    London Community Healthcare, Westside Contraceptive Services, London, UK

  4. 4

    Gynuity Health projects, New York, New York, USA

*David A Grimes, Obstetrics and Gynecology, University of North Carolina, School of Medicine, CB#7570, Chapel Hill, North Carolina, 27599-7570, USA. david_grimes@med.unc.edu.

Publication History

  1. Publication Status: Edited (no change to conclusions)
  2. Published Online: 13 NOV 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]
Collaborative 1998

MethodsStratified randomized controlled trial; double blind. Stratification: breastfeeding, switchers (use of oral contraceptives within past 2 months), or starters (not a switcher or breastfeeding). Randomized 3:1 desogestrel:levonorgestrel. Allocation concealment provided by pharmacy distribution of identical pills in identical packages.


Participants1320 healthy sexually active women aged 18 to 45 years with normal cycles; 405 women were breast feeding. Recruited at 44 centers in Europe. Mean cycle length between 24 and 35 days and intraindividual variation +/- 3 days. Willing to fill in diary card on bleeding and pill intake. Body weight between 80% and 130% of ideal. Willing to give informed consent. Exclusion criteria included contraindications to steroids, prior ectopic pregnancy, pelvic inflammatory disease, or symptomatic functional ovarian cysts.


InterventionsDesogestrel 75 μg/day versus levonorgestrel 30 μg/day.


OutcomesPregnancy, adverse events, and bleeding patterns assessed after 3, 7, and 13 treatment cycles of 28 days.


NotesBoth pills were identical in appearance. Sample size chosen to allow detection of a difference in incidence of 7% between the two groups for the primary bleeding pattern variables with a power of 80% and alpha of 0.05. Intended sample size was 1200, in part to meet regulatory requirement for cycles of exposure. As a result of 10% over-recruitment, 1320 women were enrolled.


Risk of bias

BiasAuthors' judgementSupport for judgement

Allocation concealment (selection bias)Low riskNo mention of allocation concealment. Intentional exclusion of participants after randomization.

Blinding (performance bias and detection bias)
All outcomes
Low riskTablets described as "identical-looking."

Incomplete outcome data (attrition bias)
All outcomes
Low riskSmall numbers of participants excluded from analysis because of missing information or protocol violations.

Selective reporting (reporting bias)Low riskUniform determination of outcomes for both groups.

Lakha 2007

MethodsRandomized controlled trial in four centers; blocked computer-generated randomization scheme developed in Edinburgh and stratified by center. Block size and whether size varied was not stated.


Participants97 healthy women in Nigeria, South Africa, Hong Kong, and Edinburgh. Aged 18 to 40 years, regular menstrual cycles (21 to 42 days). Negative pregnancy test upon admission. Excluded women with hormonal contraception use in past 3 months except in Edinburgh, where 9 women using progestin-only pill and having regular cycles were included. Randomized in 3:1 ratio of mifepristone to levonorgestrel.


InterventionsMifepristone 5 mg versus levonorgestrel 30 μg per day. Participants were studied for one pretreatment cycle, six cycles of therapy, and one cycle after stopping therapy.


OutcomesFrequency of amenorrhea, bleeding patterns, side effects, and efficacy.


NotesProvided baseline characteristics as recommended but performed statistical testing on these, which is improper (Moher 2001).


Risk of bias

BiasAuthors' judgementSupport for judgement

Allocation concealment (selection bias)Low riskParticipants were given coded opaque pill bottles.

Blinding (performance bias and detection bias)
All outcomes
Low riskDouble dummy approach; participants received either mifepristone (half tablet) with placebo or levonorgestrel with a placebo (half tablet). Mifepristone and placebo tablets both white but of different size.

Incomplete outcome data (attrition bias)
All outcomes
Low riskSix participants excluded from analysis because of protocol violations.

Selective reporting (reporting bias)Low riskSimilar determination of outcomes for both groups.

Paulsen 1974

MethodsRandomized controlled trial in one clinic; double blinding attempted. Randomization sequence developed by study sponsor (Searle).


Participants86 healthy women in a Planned Parenthood clinic in San Jose, California. Aged 18 to 39 years and without medical contraindications to oral contraception.


InterventionsProgestin-only pill containing ethynodiol diacetate 0.25 mg versus combined oral contraceptive containing ethynodiol diacetate 1.0 mg plus mestranol 0.1 mg; followed for 12 months.


OutcomesBleeding patterns, discontinuation from study.


NotesPart of a larger trial sponsored by G.D. Searle and Co.


Risk of bias

BiasAuthors' judgementSupport for judgement

Allocation concealment (selection bias)Unclear riskNo mention of allocation concealment.

Blinding (performance bias and detection bias)
All outcomes
Unclear riskPills were prepared by the manufacturer so that participants could not distinguish between the two types of pills. Whether packaging was identical was not stated.

Incomplete outcome data (attrition bias)
All outcomes
High riskDifferential discontinuation rates.

Selective reporting (reporting bias)High riskDifferential exclusion of participants from bleeding pattern analysis.

Sheth 1982

MethodsRandomized controlled trial in two centers as part of a World Health Organization study; double-blind.


Participants518 women who were healthy and had no contraindications to oral contraception, aged 18 to 38 years, if post-partum greater than 28 and days and menstruating, if lactating to have breast fed at least 165 days and be menstruating, no oral contraceptive use in 28 days or injectable hormonal contraceptive use in 90 days, regular menstrual cycles of 21 to 35 days. Participants came from Bombay and Ljubljana, Yugoslavia.


InterventionsFour oral contraceptives, two progestin-only and two combination pills: norethisterone (norethindrone) 350 μg, levonorgestrel 30 μg, norethisterone 1 mg plus mestranol 50 μg, and levonorgestrel 150 μg plus ethinyl estradiol 30 μg.


OutcomesPregnancy, discontinuation, and bleeding disturbances.


NotesTwo of the planned four international sites were dropped from the study because of recruitment problems. The remaining sites provided about half the intended sample size (1200 women).


Risk of bias

BiasAuthors' judgementSupport for judgement

Allocation concealment (selection bias)Unclear riskRandomization technique not specified.

Blinding (performance bias and detection bias)
All outcomes
High riskPills produced by manufacturers as usual. All pills were repackaged in 28-day packs. Pill packs were identified only by a random number. Although uniform packaging was provided, pills could be identified.

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInvestigators excluded participants from analysis for noncompliance.

Selective reporting (reporting bias)Low riskLow losses to follow up.

Vessey 1972

MethodsRandomized controlled trial in one center in Ljubljana, Yugoslavia; double-blind.


Participants360 women aged 18 to 44 years, of proven fertility but with no more spontaneous abortions than live born children, no hormone preparation in past two months, normal gynecological examination, and healthy.


InterventionsFour treatment groups randomized to progestin-only pills: megestrol acetate 700 μg, norethisterone acetate 300 μg, chlormadinone acetate 500 μg, or norgestrel 75 μg. Participants were followed at 4, 7, 10, and 13 months after starting the trial. Observation planned for a year.


OutcomesDiscontinuation because of pregnancy, menstrual disturbances, other side effects, other reasons.


NotesA fifth concurrent group of 90 women was assigned in non-random fashion to a combination contraceptive, ethinyl estradiol 100 μg plus megestrol acetate 2 mg.


Risk of bias

BiasAuthors' judgementSupport for judgement

Allocation concealment (selection bias)Low riskPill cartons were "arranged in random order" and then given a serial number. Each women enrolled was given the next carton as her assignment.

Blinding (performance bias and detection bias)
All outcomes
Low riskAll progestin-only pills were manufactured to be identical in appearance and consistency. These were packaged in bubble packs of 42 tablets; nine packs were placed in a carton.

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll participants successfully followed for one year or until discontinuation.

Selective reporting (reporting bias)High riskSixteen percent of participants excluded after randomization for noncompliance.

Were 1997

MethodsRandomized controlled trial in one academic center in Eldoret, Kenya; open label.


Participants200 women aged 18 to 35 years, had delivered a health term infant with plans to breast feed for at least six months, planned to be sexually active, agreed to randomization, agreed to rely exclusively on the assigned contraceptive, and planned to remain available for follow up for 18 months. Exclusion criteria included separation from the infant for more than six hours per day, diastolic blood pressure > 90 Torr, or other contraindications to progestin-only pills.


InterventionsParticipants were randomized to begin norgestrel 75 μg daily either at six weeks postpartum (group 1) versus at six months postpartum or after return of menses (group 2), whichever occurred first.


OutcomesPregnancy and discontinuation rates by reason.


NotesThis report is from one center of a planned multicenter trial, designed to have 2000 participants (Family Health International 1991). The outcome of the overall trial is unknown. Although the trial planned for 18 months of follow up, some data beyond that time interval are reported.


Risk of bias

BiasAuthors' judgementSupport for judgement

Allocation concealment (selection bias)Low riskMethod indicator cards inside sequentially numbered, opaque, sealed envelopes.

Blinding (performance bias and detection bias)
All outcomes
High riskNot feasible.

Incomplete outcome data (attrition bias)
All outcomes
Low riskLarge number of participants lost after randomization due to ethnic strife in region.

Selective reporting (reporting bias)High riskSmall minority of participants in both groups completed trial as planned.

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Bhattacharya 2012All three study arms received combination oral contraceptives.

Korver 2005Although this was a randomized controlled trial, the outcome of interest was ovulation, not pregnancy.

Rice 1999Although this was a randomized controlled trial, the outcome of interest was ovulation, not pregnancy.

 
Comparison 1. Desogestrel 75 μg versus levonorgestrel 30 μg daily

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Pregnancy11320Rate Ratio (Fixed, 95% CI)0.27 [0.06, 1.19]

 2 Discontinuation because of adverse events11320Rate Ratio (Fixed, 95% CI)1.22 [0.81, 1.84]

 3 Discontinuation because of irregular bleeding11320Rate Ratio (Fixed, 95% CI)1.32 [0.99, 1.78]

 4 Discontinuation for all reasons11320Rate Ratio (Fixed, 95% CI)1.21 [0.99, 1.47]

 
Comparison 2. Mifepristone 5 mg versus levonorgestrel 30 μg daily

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Pregnancy1441Odds Ratio (Peto, Fixed, 95% CI)0.71 [0.07, 6.95]

 2 Amenorrhea196Odds Ratio (Peto, Fixed, 95% CI)0.02 [0.00, 0.37]

 3 Mean endometrial thickness at 24 weeks173Mean Difference (IV, Fixed, 95% CI)6.30 [4.28, 8.32]

 4 Endometrial biopsy to evaluate endometrial cavity >12 mm on ultrasound196Odds Ratio (Peto, Fixed, 95% CI)7.74 [0.98, 61.41]

 
Comparison 3. Ethynodiol diacetate 0.25 mg versus ethynodiol diacetate 1.0 mg plus mestranol 0.1 mg

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Irregular cycles166Odds Ratio (Peto, Fixed, 95% CI)135.96 [7.63, 2421.02]

 2 Intercycle bleeding166Odds Ratio (Peto, Fixed, 95% CI)6.20 [2.11, 18.22]

 3 Discontinuation from trial186Odds Ratio (Peto, Fixed, 95% CI)1.78 [0.75, 4.24]

 
Comparison 4. Norethisterone acetate 0.3 mg versus norgestrel 0.075 mg

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Pregnancy1150Rate Ratio (Fixed, 95% CI)0.78 [0.05, 12.35]

 2 Continuation at one year1Risk Difference (Fixed, 95% CI)28.2 [12.32, 44.08]

 3 Discontinuation because of menstrual disturbance1150Rate Ratio (Fixed, 95% CI)0.30 [0.15, 0.62]

 4 Discontinuation because of other side effects1150Rate Ratio (Fixed, 95% CI)0.52 [0.15, 1.87]

 5 Discontinuation for reasons unconnected with treatment1150Rate Ratio (Fixed, 95% CI)1.06 [0.57, 1.99]

 
Table 1. Cumulative net life-table discontinuation rates at 360 daysa

OutcomeNorethisterone 350 μg

N = 130
Levonorgestrel 30 μg

N = 128
Norethisterone 1 mg/mestranol 50 μg

N = 123
Levonorgestrel 150 μg/ethinyl estradiol 30 μg

N = 137
P value

Discontinuation at 360 days

Accidental pregnancy13.29.58.32.70.077

Bleeding disturbances24.226.023.29.70.052

All gastrointestinal2.55.72.511.10.011

All central nervous system5.92.713.90.90.000

All causes57.760.961.052.60.805

 aFrom Sheth 1982, Table II.
 
Table 2. Life-table continuation rates over 18 months by treatment groupb

Cumulative proportion continuing method

Interval (months)Start at 6 weeksStart at 6 months
or with first menses



PercentS.E.PercentS.E.

0 to 685.73.885.04.0

7 to 1253.67.056.77.0

13 to 1846.18.943.68.3

 bFrom Were 1997, Table 2. Although the trial was designed for 18 months of follow up, the report provided data beyond that time interval.