Oral contraceptive pills (OCPs) are the most popular form of reversible contraception in the United States. In the United States, more than 10 million women—28% of women using contraception—currently use OCPs,1 and 82% of women of reproductive age have used OCPs, often for several years.2 Despite their popularity, multiple studies show OCP discontinuation rates of 25–85% during the first 6–12 months of use.3,4 Many women discontinue OCPs while remaining sexually active, when they are still at risk of an undesired pregnancy, and adopt a less effective method or do not substitute another contraceptive at all.5
Reasons for discontinuation include changes in need, side effects, and logistic reasons. Running out of pills, along with difficulty in access, is a leading reason for OCP discontinuation.6,7 A prevalent approach to OCP initiation or refill is to supply pills for 3 months or less, whether in actual packages given (as in publicly funded family-planning clinics) or through a prescription for refills (as for patients with public or private insurance).4 Historically, the initial supply was restricted as a result of concerns about OCP-induced hypertension and the lack of information about possible effects of the OCP on an early pregnancy. Despite the development of high-sensitivity pregnancy tests, nd the low incidence of OCP-induced hypertension,8 these restrictive prescribing practices remain. In addition, scarce resources may prevent publicly funded clinics from providing a greater supply of OCPs. A recent analysis of Medi-Cal prescribing data showed that women who received a greater supply were more likely to continue OCP use,9 but this record linkage study could not assess whether specific patient characteristics led clinicians to provide a larger OCP supply. The objective of this randomized trial was to estimate the effect of an increased initial OCP supply on 6-month continuation rates. Planned analyses included evaluation of continuation in subgroups defined by age (younger than or older than 18 years of age) as well as continuation among Medicaid-insured participants who received their pill supply as packs compared with those who received a prescription.
We conducted a randomized clinical trial at the Family Planning Clinic of the New York Presbyterian Hospital–Columbia University Mailman School of Public Health. The Family Planning Clinic is community-based and provides more than 20,000 annual visits. The clinic receives public funding to care for poor women; 96% of patients report incomes below 100% of the Federal poverty level. Standard care in the clinic is for all women seeking OCPs to receive one pack of pills free of charge for immediate initiation of the method. Women aged 18 years and older with insurance receive a prescription for refills. Women younger than age 18 years or women without insurance receive two additional packs (for a total of three) free of charge; clinic resources preclude a greater routine supply of pills. Clinic health educators or nurses referred women aged 35 years or younger seeking the OCP to the research coordinator, who informed them about the study while they were waiting to see the clinician. They were not admitted to the study if any of the following were present: 1) contraindications to hormonal contraceptives10; 2) current use of hormonal contraceptives (continuation defined as ongoing use within the last 7 days); 3) desiring pregnancy within the next 6 months; 4) not sexually active; 5) not English- or Spanish-speaking; or 6) planning on leaving the area within 6 months. The study was approved by the Columbia University institutional review board and was registered with ClinicalTrials.gov (identifier NCT00677742).
A bilingual in English and Spanish research coordinator checked eligibility criteria and obtained informed consent; she then administered a structured interview regarding factors related to method discontinuation, including sociodemographic and health-related information and menstrual and reproductive history.
We assigned participants to three strata based on age and insurance status: 1) age younger than 18 years; 2) adult participants (age 18 years and older) without insurance; and 3) adults with Medicaid. Participants were considered insured if they possessed a valid insurance or Medicaid card that permitted the filling of prescriptions at an outside pharmacy. All participants were randomly assigned within strata to receive either a usual three-cycle (three packs) or an enhanced seven-cycle (seven packs) supply of OCPs. The randomization schedule was created by an investigator who had no contact with participants using a random number table. Randomization assignments within strata were placed in sequentially numbered, sealed opaque envelopes, which were opened by the research coordinator. The allocation ratio was 1:1 in blocks of four.
Participants in strata 1 and 2 received their entire pill supply as packs. Participants in the third stratum (age 18 years or older with insurance) had a secondary randomization to receive packs or a prescription. This additional randomization (with an allocation ratio of 2:3 in blocks of 10) allowed us to compare OCP continuation between those who receive an OCP supply and those who must visit a pharmacy for refills. There was no blinding in this study.
After enrollment and randomization, participants saw a clinician and received routine care per clinic protocols. All received emergency contraception during the visit if indicated by sexual history. The health care provider chose the OCP brand, limited to the two combination pills available in the clinic during the study period (35 micrograms ethinyl estradiol and 0.18/0.215/0.25 mg norgestimate; 25 micrograms ethinyl estradiol and 0.18/0.215/0.25 mg norgestimate). The clinician provided participants with uniform information about risks and benefits of OCP use as part of the routine informed consent process for contraception separate from the research consent process. All participants received their first package of pills free of charge, regardless of insurance status, and were encouraged to swallow the first pill during the visit per clinic protocol.4 The clinician then gave participants the allotted supply of remaining pills as either packs (free of charge) or prescription (no copay). All participants received a supply of condoms for backup contraception and for ongoing use for sexually transmitted infection protection; they also received a $10 public transportation voucher. Participants in the three-pack group were scheduled for a routine 3-month follow-up visit to allow for pill refills; participants in the seven-pack group did not have a revisit scheduled.
We began follow-up phone calls 24 weeks after enrollment. The follow-up interview focused on contraceptive use. We asked participants whether they had started the pill and if they had taken a pill in the past 7 days, which we defined as OCP continuation. For participants whose follow-up interview took place more than 6 months after enrollment, we also asked the number of packs that they had started and completed to evaluate OCP continuation. For instance, if the follow-up interview took place 8 months after enrollment, and the woman had not used the OCP in the 7 days preceding the interview but she reported completing seven packs since enrollment, then we considered her to have been continuing use at the 6-month date. This best estimate was made without knowledge of treatment assignment.
The main outcome measure was 6-month OCP continuation within each stratum. We anticipated that 25% of participants would be lost to follow-up. To have 80% power to identify a 20% difference in continuation rates (60% compared with 40%), with a two-sided α level of 0.05, we needed to enroll 125 participants per arm per stratum. We therefore needed to enroll 750 total participants. We planned to examine continuation differences within and between strata as risk differences.
The secondary outcome was the effect of the type of pill supply (packs compared with prescription) on OCP continuation. We also examined adverse events (emergency room visits and hospitalizations) and pregnancies in each group and used the electronic medical record to validate reported follow-up visits. Data were analyzed using Stata according to this pre-established analysis plan. We used χ2 tests and Fisher's exact test as appropriate. We conducted multivariable analyses with logistic regression. All significance levels quoted (P values) are two-sided.
Participants were enrolled from July 2007 through March 2009. The flow of participants through the trial is detailed in Figure 1. Among 770 screened women who did not meet inclusion criteria, the main reasons were current use of OCPs or selection of another method of contraception. Among eligible women, reasons for not enrolling included no interest in participation (n=142), wanted a specific OCP not available through the study (n=15), or already had an OCP supply at home (n=11). We randomized 359 women were randomized to the three-cycle arm, and 341 women to the seven-cycle arm. Fourteen participants were excluded after randomization because the medical provider identified a contraindication to combined oral contraception, and 25 women chose another contraceptive method while seeing the medical provider; we did not attempt to follow up these 39 women. Of the 661 participants, 19% were enrolled in stratum 1 (age younger than 18 years), 64% in stratum 2 (age 18 years or older without insurance), and 17% in stratum 3 (age 18 years or older with insurance, primarily Medicaid). We completed follow-up interviews with 504 of the 661 participants who received OCPs (approximately 76% in both the three-pack and seven-pack arms and in each of the three strata). Follow-up ended in October 2009; the median duration of follow-up was 184 days, or 6.0 months. For 66%, we completed follow-up interviews within 196 days, when a seventh pack of pills would be ending; 90% of participants had follow-up within 8 months. We attempted to contact participants at least 15 times before considering them lost to follow-up.
Table 1 shows baseline demographic variables of the study population. During the time available, the clinic saw somewhat fewer women aged younger than 18 years and fewer adults with Medicaid than expected; thus, these strata were smaller than planned. Participants in the seven-pack group were more likely to have given birth and to have two or more children. Participants were mainly Hispanic; almost half of the participants came from the Dominican Republic, and less than 10% were from Mexico or Puerto Rico. The racial and ethnic mix of women who enrolled mirrors the composition of the clinic population as a whole; we did not collect information on immigration status. Two thirds of participants spoke primarily Spanish at home. Ten percent had given birth in the prior 6 months. Participants with and without follow-up were similar with respect to baseline characteristics except that those who were lost to follow-up were less likely to have education beyond high school (P=.03).
The primary analysis included all 504 participants who received the assigned pill supply and for whom we have follow-up information (260 in the three-pack group and 244 in the seven-pack group). Using our best estimate of continuation, participants in the seven-pack group reported a higher 6-month continuation rate than participants in the three-pack group (51% compared with 35%, P<.001; Table 2). The benefit of receiving seven packs was considerably greater among participants younger than age 18 (36.9% compared with 11.8%, P=.014; Tables 2 and 3). This difference by age was not affected by adjusting for smoking, previous OCP use, being parous, and wanting OCPs for more than 6 months (Table 3); this table expresses the results in odds ratio terms to permit the simultaneous adjustment for these other factors by logistic regression analysis. Evaluation of continuation solely on OCP use in the 7 days preceding the interview gave similar results.
Comparison of participants who were randomized to receive their pill supply as either packs or a prescription was underpowered as a result of the small number of women enrolled in this stratum (Table 4). Participants who received a prescription were less likely to continue OCP use than those who received a supply of packs (P=.027).
In univariable analyses, previous OCP use, being parous, and planning to use OCPs for at least 6 months were associated with greater OCP continuation. Smokers had less OCP continuation (data not shown). As discussed previously, adjusting for these variables did not modify the main effect (Table 3). We also examined body mass index, education, history of abortion, insurance status, change in insurance status during the study, time since last pregnancy, and plans for future children; these variables were not associated with OCP continuation.
During the follow-up phone call, we asked participants who discontinued what was the main reason that they stopped taking the pill. The most common reasons given were side effects (83 participants), ran out of pills (58), no longer sexually active (54), missed pills (49), and seeking pregnancy (10). The only given reason that differed between groups was running out of pills, reported by 42 of the 168 discontinuers in the three-pack group and 16 of the 119 discontinuers in the seven-pack group (P=.02).
Based on the number of packs dispensed at study enrollment and the date of OCP discontinuation, many participants would have had some remaining OCP packs. There were 168 packs unused by participants in the three-pack group and 601 packs unused by those in the seven-pack group. Similar proportions returned to the Family Planning Clinic or another clinic to switch methods during the study (10.4% of the three-pack group, 9.4% of the seven-pack group; P=.60).
We asked participants about follow-up visits to this or any other clinic to obtain pill refills and used the electronic medical records to assess whether the reported visit actually took place (to assess for possible social desirability answering). We were able to confirm 90% of reported clinic visits. In the three-pack group, only 65% of revisits took place in time to allow uninterrupted OCP use. We also reviewed the electronic health records of all 157 women who were lost to follow-up in this study. These women returned to a clinic for pill refills at the same rate (55%) as women who continued in the study (53%).
There were six adverse events in the three-pack group, including an ectopic pregnancy, syncopal episode, upper respiratory infection, asthma exacerbation, gastroenteritis, and ulcerative colitis flare. There were three adverse events in the seven-pack group, including cholecystectomy (2 months after study enrollment), hemiplegic migraine (4 months after enrollment), and gastroenteritis. No adverse events appear to be related to receiving more packs at the time of OCP initiation or to lack of a 3-month follow-up visit.
Based on follow-up interviews, 504 study participants experienced 46 pregnancies during the study; this included 25 of 260 women (9.9%) in the three-pack group and 22 of 244 women (9.0%) in the seven-pack group (P=.88). Pregnancy rates were similar between participants younger than age 18 years who received three (six of 42 [14%]) or seven packs (five of 45 [11%]; P=.75); however, the study had exceedingly limited statistical power for these comparisons.
Increasing the supply of OCPs at the time of initiation increases 6-month continuation rates. The benefit of receiving seven packs was substantially greater among participants younger than 18 years old. In addition, receiving packs in hand led to greater continuation than receiving a prescription, although small numbers of insured women gave limited power for this comparison. Finally, the women who received the enhanced supply had less medical surveillance but did not experience more adverse health events during the study. We did not directly assess the rate of increased blood pressure in these women, but the risk appears to be low.11,12
We found few other studies that examined the effect of pack supply on OCP continuation. An historical cohort study of Medi-Cal paid claims data in California demonstrated that women who received 13 packs at their first visit had higher method continuation with fewer gaps in coverage than women who received one or three packs.13 A cluster-randomized trial among 20 Jamaican family-planning clinics found a modest improvement in OCP continuation at 4 months among women who received an initial supply of four packs compared with women who received a single pack (72% compared with 65%),14 and participants in clinics that distributed the enhanced supply were less likely to experience a gap in pill use. A cohort study of U.S. women who obtained OCPs from a family-planning clinic near the U.S.–Mexican border found that those who received only one to five packs had a 62% higher discontinuation rate than those who received six or more packs at the clinic.15 A recent record linkage study of Medi-Cal data (more than 84,000 women) showed that women who received a 1-year supply of OCPs were less likely to have a pregnancy when compared with those who received one to three cycles of pills; dispensing a 1-year supply was also associated with a 30% reduction in the odds of conceiving an unplanned pregnancy.9 We found no studies that randomized individual women to different regimens of pill supplies.
Our study population is largely Hispanic, urban, and poor, so our results may not be generalizable to all groups of women. The observed benefits may nonetheless prove to be relevant to other populations with great contraceptive needs. We tried to evaluate a prescription subgroup that would resemble the larger population of insured U.S. women who need to refill their OCP prescriptions each month. Because the insured subgroup in this study was smaller than anticipated, we had limited power to compare pack supply with prescriptions. We observed a trend toward lower continuation rates among participants who received a prescription for refills compared with women who received their OCP supplies in hand; however, this issue needs further study.
The 6-month OCP continuation rates seen here were low and are comparable to those reported from other publicly funded clinics4 as well as among privately insured women.3 Medical record validation of OCP refill visits in this study showed no evidence that the participants overreported their OCP continuation. Although we could verify the OCP refill visits, we could not verify actual pill consumption. Measuring correct OCP use remains difficult.16
Concern about the high rate of unplanned pregnancy in the United States has led to many interventions that do not lead to increased contraceptive use.17 Most recently, Kirby and Raine18 evaluated an intensive telephone follow-up intervention for 805 adolescents beginning contraceptive use and found no effect on OCP continuation. Our results indicate that a health services intervention may be more effective at improving OCP continuation rates than attempts to change individual behavior. Compared with many other prescription medications, OCP use requires minimal medical supervision, because there are no dosages to adjust, and patients can self-identify side effects and the need for continued treatment. In this study, an unforeseen advantage of a greater initial pill supply was that it gave women more time in which to return for a refill and thus more opportunity for uninterrupted continuation.
Dispensing multiple packs of pills at the time of OCP initiation would require substantial changes in the current service model in public family-planning clinics. More than 4 million women receive family-planning services each year at a Title X-supported clinic. These clinics serve 15% of all women in the United States who obtain contraceptive prescriptions or supplies or who receive a checkup for birth control each year.19 As in the rest of the U.S. fee-for-service system, clinics typically are reimbursed for the number of patient encounters they provide, with little incentive to reduce the number of “easy” pill refill visits. Patients would be better served by having fewer refill visits, which would also allow clinics more time to see more new clients. The approach presented here will work only in settings that stock OCPs on-site, and stocking and dispensing more pills may be prohibitively expensive for publicly funded clinics. The cost deterrence of providing an enhanced OCP supply is ironic when all modern contraceptive methods are cost-effective, saving more in expenditures than they cost to provide20; analysis of Medi-Cal data shows that OCPs save $4 for each dollar spent on supply.21
Discontinuation of OCPs is not just a problem among poor or uninsured women.3 Changes in insurance plan rules to permit multiple refills could increase access to and continuation of OCPs. Single-month refills are a barrier to OCP continuation, even for insured women.22 Our study shows a trend toward decreased continuation when women need to obtain pharmacy refills. Some women with private insurance can get three OCP packs at a time through mail-in programs; our study showed that three packs of OCPs at once may not be enough to enhance continuation and that seven packs—beyond the scope of any mail-in program—is significantly better. The optimum number of OCP packs to dispense at one time has not yet been determined. Although beyond the scope of our study, a 1-year supply of OCPs may be even better than the seven packs given here.
These results have implications beyond OCP continuation. Many medications need to be taken regularly to prevent disease progression and disability. Drug-dispensing limits—30-day supplies when filled at community pharmacies—are used to control prescription drug costs but may result in overall higher costs as a result of worsening disease.23 Medication continuation could be greatly enhanced by simply allowing patients to have a greater supply of pills on hand.
2. Mosher WD, Jones J. Use of contraception in the United States: 1982–2008. Vital Health Stat 2010:1–44.
3. Nelson AL, Westhoff C, Schnare SM. Real-world patterns of prescription refills for branded hormonal contraceptives: a reflection of contraceptive discontinuation. Obstet Gynecol 2008;112:782–7.
4. Westhoff C, Heartwell S, Edwards S, Zieman M, Cushman L, Robilotto C, et al.. Initiation of oral contraceptives using a quick start compared with a conventional start: a randomized controlled trial. Obstet Gynecol 2007;109:1270–6.
5. Rosenberg MJ, Burnhill MS, Waugh MS, Grimes DA, Hillard PJ. Compliance and oral contraceptives: a review. Contraception 1995;52:137–41.
6. Smith JD, Oakley D. Why do women miss oral contraceptive pills? An analysis of women's self-described reasons for missed. J Midwifery Womens Health 2005;50:380–5.
7. Kerns J, Westhoff C, Morroni C, Murphy PA. Partner influence on early discontinuation of the pill in a predominantly Hispanic population. Perspect Sex Reprod Health 2003;35:256–60.
8. Chasan-Taber L, Willett WC, Manson JE, Spiegelman D, Hunter DJ, Curhan G, et al.. Prospective study of oral contraceptives and hypertension among women in the United States. Circulation 1996;94:483–9.
9. Foster DG, Hulett D, Bradsberry M, Darney P, Policar M. Number of oral contraceptive pill packages dispensed and subsequent unintended pregnancies. Obstet Gynecol 2011;117:566–72.
10. World Health Organization. Medical eligibility criteria for contraceptive use. Geneva (Switzerland): WHO Press; 2009.
11. Curtis KM, Chrisman CE, Peterson HB. Contraception for women in selected circumstances. Obstet Gynecol 2002;99:1100–12.
12. Pollara T, Kelsberg G, Safranek S, Schrager S. Clinical inquiries. What is the risk of adverse outcomes in a woman who develops mild hypertension from OCs? J Fam Pract 2006;55:986–8.
13. Foster DG, Parvataneni R, de Bocanegra HT, Lewis C, Bradsberry M, Darney P. Number of oral contraceptive pill packages dispensed, method continuation, and costs. Obstet Gynecol 2006;108:1107–14.
14. Chin-Quee D, Otterness C, Wedderburn M, McDonald O, Janowitz B. One versus multiple packs for women starting oral contraceptive pills: a comparison of two distribution regimens. Contraception 2009;79:369–74.
15. Potter JE, McKinnon S, Hopkins K, Amastae J, Shedlin MG, Powers DA, et al.. Continuation of prescribed compared with over-the-counter oral contraceptives. Obstet Gynecol 2011;117:551–7.
16. Hall KS, White KO, Reame N, Westhoff C. Studying the use of oral contraception: a review of measurement approaches. J Womens Health (Larchmt) 2010;19:2203–10.
17. Halpern V, Grimes DA, Lopez L, Gallo MF. Strategies to improve adherence and acceptability of hormonal methods for contraception. The Cochrane Database of Systematic Reviews 2011, Issue 4. Art. No.: CD004317. DOI: 10.1002/14651858.CD004317.pub3.
18. Kirby D, Raine T, Thrush G, Yuen C, Sokoloff A, Potter SC. Impact of an intervention to improve contraceptive use through follow-up phone calls to female adolescent clinic patients. Perspect Sex Reprod Health 2010;42:251–7.
19. Gold RB. Title X: three decades of accomplishment. The Guttmacher Report on Public Policy 2001;4:5–8.
20. Trussell J, Lalla AM, Doan QV, Reyes E, Pinto L, Gricar J. Cost effectiveness of contraceptives in the United States. Contraception 2009;79:5–14.
21. Foster DG, Rostovtseva DP, Brindis CD, Biggs A, Hulett D, Darney PD. Cost savings from the provision of specific methods of contraception in a publicly funded program. Am J Public Health 2009;99:446–51.
22. Phillips KA, Stotland NE, Liang SY, Spetz J, Haas JS, Oren E. Out-of-pocket expenditures for oral contraceptives and number of packs per purchase. J Am Med Womens Assoc 2004;59:36–42.
23. Phillips KA, Liang SY, Haas JS, Stebbins M, Alldredge BK. Prescription drug dispensing limits and patterns. Manag Care Interface 2005;18:41–6.
© 2011 by The American College of Obstetricians and Gynecologists.