Combined hormonal contraceptive methods are very popular in the United States. The most commonly used hormonal contraceptive in the United States continues to be oral contraceptive pills (OCPs). The most recent National Survey of Family Growth reported 28% of contracepting women currently use OCPs, whereas 2% of women use the contraceptive vaginal ring, and less than 1% use the contraceptive patch.1 Nearly half of the unintended pregnancies in the United States occur while using some form of contraception.2 A large proportion of unintended pregnancies occur as a result of method nonadherence and discontinuation3; thus, understanding the challenges to correct and consistent use of combined hormonal contraception remains an important question.
Although many women initiate OCPs, the patch, and the ring, a large proportion will discontinue within 1 year. In large population-based studies of OCPs, over 30% of women discontinued by 6 months,4 and nearly 50% discontinued by 12 months.5,6 Discontinuation rates were even greater in university clinic populations with 60% of new users discontinuing OCPs at 6 months.7 In clinical trials of the ring, discontinuation at 1 year ranged from 12% to 35%.8–10 Only 54% of women continued to use the ring at 12 months in a large community-based study.5 Perfect use of the patch was reported in nearly 90% of women through 13 cycles in a clinical trial.11 However, real-world use among teens enrolled in a longitudinal cohort study was much lower with only 48% of adolescents still using the patch at 1 year; those who discontinued commonly reported a prescription lapse as a result of lack of transportation or insurance.12,13 Studies from administrative claims databases also demonstrate low continuation rates for the OCP, patch, and ring: 51–64% at 3 months and 16–35% at 12 months.12,14
Studies of OCP use often cite side effects as the most common reason for discontinuation.3,4,15 Interestingly, in a large study of publicly funded clinics, the most important factor associated with discontinuation was not side effects, but difficulty obtaining and correctly using OCPs.7 There is conflicting evidence of the importance of demographic and socioeconomic factors on contraceptive discontinuation.4,6 Although adverse events were the most common reason for discontinuation of the ring in clinical acceptability trials,8,9 another cited reason for discontinuation was partner nonacceptability.10
To further examine factors associated with contraceptive discontinuation, we studied contraceptive use in a subset of women using OCPs, the patch, or ring in the Contraceptive CHOICE Project. The primary objective of this study was to examine factors associated with discontinuation of the OCP, patch, and ring.
MATERIALS AND METHODS
The Contraceptive Choice Project (CHOICE) is a longitudinal cohort study seeking to decrease unintended pregnancies in the St. Louis region and has been described previously.16 CHOICE is a convenience sample of women who are followed for 2–3 years during which they receive contraception and sexually transmitted infection screening and treatment at no cost. To be eligible for CHOICE, participants must be 14–45 years of age; sexually active with a male partner in the past 6 months or plan to be sexually active in the next 6 months; at risk of unintended pregnancy (no tubal ligation or hysterectomy) with no plans for pregnancy in the next 12 months; reside in St. Louis or seek clinical care in study-affiliated clinics; and willing to start a new method of contraception. Notably, participants could have previously used their baseline method if they were not currently using it at enrollment. At enrollment, participants provide written informed consent and respond to a staff-administered baseline questionnaire assessing demographic characteristics, reproductive history, and sexual behaviors. All enrolled participants receive standardized contraceptive counseling describing all reversible contraceptive methods. After enrollment, participants are contacted for telephone interviews at 3, 6, and 12 months and every 6 months afterward for the duration of follow-up. The study was approved by the Washington University in St. Louis School of Medicine Human Research Protection Office before the initiation of recruitment.
Participants eligible for this analysis chose the OCP, patch, or ring at baseline, initiated use by the 3-month follow-up interview, and either completed their 12-month follow-up survey or provided information on method discontinuation before 12 months. Participants selecting the OCP, patch, or ring are required to fill their prescription on a monthly basis to duplicate requirements of common insurance plans in the St. Louis area. CHOICE partnered with one local grocery pharmacy chain to dispense contraceptive methods to study participants. In addition, participants who are patients of a local family planning clinic obtain their methods from that clinic. Those who moved from the St. Louis area obtained methods through the mail from the pharmacy chain or elected to pay for the methods from another pharmacy.
Continuation was determined by asking at each follow-up survey “are you still using the method?” If she answered “yes,” she was asked “did you ever stop using the method?” To be considered a continuer, the participant must have reported using the method at each follow-up survey with no gap in use of 1 month or longer. Our method allocation log and pharmacy refill records were queried for any participant not responding to these questions to determine continuation status. We censored participants who discontinued as a result of pregnancy or to attempt pregnancy or those lost to follow-up. Additionally, participants were asked how satisfied they were with the method: “very satisfied,” “somewhat satisfied,” or “not satisfied.” All the participants who discontinued their method by 12 months were considered “not satisfied.” Participants who discontinued were additionally asked their reason for discontinuation; multiple factors could be recorded for each discontinuer.
We present continuation rates and multivariable predictors of discontinuation for women choosing the OCP, patch, or ring. To compare baseline characteristics and continuation rates among the OCP, patch, and ring users, we used χ2 or Fisher’s exact test for categorical variables and Student’s t test for continuous normally distributed variables. Normality was evaluated by inspection of the variable's distribution using a histogram chart. Kaplan-Meier survival curves were used to estimate the continuation rates. We then compared the survival distributions by the log-rank test.
Effect modification was checked by including an interaction term between the method and the covariate of interest in the model and detected if the interaction term was statistically significant at the prespecified α level. We found no evidence of effect modification. We constructed two multivariable Cox proportional hazard models to estimate hazard ratios for discontinuation. We used stepwise selection with entry significance level of 0.25 and removal of 0.05. Model 1 included baseline demographic and behavioral factors significantly associated with method discontinuation in univariable analyses. Because race, age, and marital status have been shown to be associated with method continuation, they were forced into model 1. We retained contraceptive method in the model because it was the primary exposure in our study. We considered time-dependent factors collected in the 3-, 6-, and 12-month surveys part of the causal pathway between contraceptive method choice and discontinuation. These factors mediated the association between contraceptive method choice and discontinuation and included feelings about future pregnancy, satisfaction with the method, difficulty obtaining the method, side effects of using the method, and certainty about their method continuation despite side effects. Therefore, model 2 only included the time-dependent factors assessed at each follow-up survey. All analyses were performed using Stata 11 and the P value was set at .05.
Between August 2007 and September 2011, 1,686 women selected the OCP (n=874), patch (n=166), or ring (n=646) as their baseline method. Of those women, 1,452 (86.1%) met inclusion criteria for this analysis. Demographic characteristics and reproductive history by method are shown in Table 1. Participants were young with a mean age of 24 years. Oral contraceptive pill, patch, and ring users varied in terms of race, measures of socioeconomic status, and reproductive characteristics. Nearly one-third of patch users received some form of public assistance compared with 20% of OCP and 19% of ring users (P=.003). Eighteen percent of patch users reported at least three unintended pregnancies (compared with OCP users 8%, ring users 10%, P<.001), and 44% had an abortion (compared with OCP users 29%, ring users 28%, P=.001).
At 12 months, 47% of OCP and 49% of ring users had discontinued their method (P=.91). Fifty-eight percent of patch users had discontinued, which is significantly higher than ring (P=.02) and OCP (P=.03) users (Fig. 1). Baseline demographic and behavioral factors were examined in model 1 and are shown in Table 2. Model 1 initially included the following variables: contraceptive method, age, race, education, marital status, receipt of public assistance, reported difficulty paying for basic expenses, insurance status, income level, parity, history of sexually transmitted infection, and prior use of the current method. The adjusted model did not demonstrate any significant effect of method choice on discontinuation. However, women who had previously used their chosen method were less likely to discontinue (adjusted hazard ratio [HR] 0.77, 95% confidence interval [CI] 0.66–0.91). Age and black race were significantly associated with discontinuation. Women living with a partner (adjusted HR 0.79, 95% CI 0.63–0.99) were less likely to discontinue, whereas women who were no longer married were more likely to discontinue (adjusted HR 1.83, 95% CI 1.27–2.65). Women receiving public assistance were 30% more likely to discontinue (adjusted HR 1.30, 95% CI 1.08–1.57).
We considered factors measured over time in model 2 (Table 3). We found that women who were certain about continuing the method regardless of side effects (adjusted HR 0.57, 95% CI 0.43–0.77) were significantly less likely to discontinue. Conversely, women experiencing breast tenderness (adjusted HR 1.39, 95% CI 1.06–1.82) were more likely to discontinue. Women who reported any difficulty obtaining their method were more than twice as likely to discontinue (adjusted HR 2.43, 95% CI 1.81–3.27). Other side effects, method satisfaction, partner nonacceptability, and ambivalence toward a pregnancy were not associated with discontinuation.
Fifty percent of OCP and ring users were satisfied with their method at 12 months (P=.91); only 38% of patch users were satisfied with their method (P=.02). We collected self-reported reasons for discontinuation. For patch and ring users, side effects were the most commonly cited reason (23% and 28%, respectively). However, OCP users reported difficulty remembering (21%) or obtaining their method (19%) in addition to side effects (17%). Partner dissatisfaction was rarely the reason for discontinuation, even among ring users (3%).
Within 4 weeks of discontinuation, 20% of women selected a more effective method (intrauterine device [IUD], implant, or injection) and 25% selected an equally effective method (OCP, patch, or ring). However, more than half of participants (55%) who discontinued selected a barrier or no method. Only 35 (5%) women who discontinued reported not being sexually active.
In this study in which financial barriers to contraception have been removed, almost half of OCP and ring users and the majority of patch users discontinued their method by 12 months. This rate is consistent with rates reported in the 2002 National Survey of Family Growth.6
Side effects have often been cited as the major reason for discontinuation of hormonal contraceptives.3,4,15 Westhoff and colleagues7 found that women from publicly funded clinics reported difficulty accessing or using methods correctly as a significant barrier to continuation. The most commonly reported reasons for discontinuation in our study reflect that both side effects and method-related factors lead to discontinuation. The only side effect significantly associated with discontinuation in this study was breast tenderness. However, women who were sure they would use their method despite the presence of side effects were significantly less likely to discontinue their method. Difficulty obtaining contraception was also significantly associated with discontinuation and may reflect difficulty with transportation, uncertain living conditions, and work schedules as posited by Nelson et al.17
Because discontinuation of effective contraception is affected by many factors, it is imperative to work with patients on factors that are modifiable. Making patients aware that many women experience side effects, and most side effects are mild or transient, may encourage a longer trial of a method.4 An open dialogue about what to do if dissatisfied with a method could also decrease the risk of gaps in contraception. Selection of a contraceptive method should take into account ability to obtain and correctly use a given method.
Switching contraceptive methods is a common phenomenon, and previous studies found that up to 80% of women who discontinue switch to less effective methods.4,15,18 With free access to IUDs, subdermal implants, and depot medroxyprogesterone acetate through the CHOICE Project, 20% of participants who discontinued their initial method selected a more effective method of contraception. However, over 50% of women who discontinued chose a less effective method leaving them at risk of unintended pregnancy. Only a few of these women were abstinent. Given the high discontinuation rates among OCP, patch, and ring users, clinicians should encourage methods with the highest contraceptive effectiveness such as the IUD and implant.
Our study has a number of strengths. We prospectively measured discontinuation in a large group of women using contraceptives. We were also able to prospectively capture reasons for discontinuation in OCP, patch, and ring users. Because we did not randomize participants, residual confounding may remain. Another limitation was that women could only receive 1 month of their method at a time to mimic local insurance plan coverage. This practice limits access and may lead to increased method discontinuation as demonstrated by the California Family PACT.13,19,20 Although women in our study received contraception at no cost, our findings remain informative, particularly for women with health insurance under the Affordable Care Act. Finally, we assumed that women who discontinued their method were dissatisfied. This may not be true in all cases, but we believe that most women in need of contraception who discontinue are not happy with their method.
Even when financial barriers to effective contraception are removed, discontinuation at 12 months among OCP and ring users is close to 50% and even greater among patch users. Proper counseling to identify whether a woman is a good candidate for the OCP, patch, or ring should consider her tolerance for side effects and plans for obtaining and using the method. Clinicians should consider long-acting reversible methods such as the IUD or implant as first-line methods for any woman seeking contraception given their superior contraceptive effectiveness and their higher level of continuation and satisfaction.21 Ultimately, contraceptive choice must be individualized to the patient's specific needs and desires.
1. Mosher WD, Jones J. Use of contraception in the United States: 1982–2008. Vital Health Stat 23 2010:1–44.
2. Finer LB, Henshaw SK. Disparities in rates of unintended pregnancy in the United States, 1994 and 2001. Perspect Sex Reprod Health 2006;38:90–6.
3. Rosenberg MJ, Waugh MS, Meehan TE. Use and misuse of oral contraceptives: risk indicators for poor pill taking and discontinuation. Contraception 1995;51:283–8.
4. Rosenberg MJ, Waugh MS. Oral contraceptive discontinuation: a prospective evaluation of frequency and reasons. Am J Obstet Gynecol 1998;179:577–82.
5. Peipert JF, Zhao Q, Allsworth JE, Petrosky E, Madden T, Eisenberg D, et al.. Continuation and satisfaction of reversible contraception. Obstet Gynecol 2011;117:1105–13.
6. Vaughan B, Trussell J, Kost K, Singh S, Jones R. Discontinuation and resumption of contraceptive use: results from the 2002 National Survey of Family Growth. Contraception 2008;78:271–83.
7. Westhoff CL, Heartwell S, Edwards S, Zieman M, Stuart G, Cwiak C, et al.. Oral contraceptive discontinuation: do side effects matter? Am J Obstet Gynecol 2007;196:412.e1–6.
8. Ahrendt H-J, Nisand I, Bastianelli C, Gomez MA, Gemzell-Danielsson K, Urdl W, et al.. Efficacy, acceptability and tolerability of the combined contraceptive ring, NuvaRing, compared with an oral contraceptive containing 30 microg of ethinyl estradiol and 3 mg of drospirenone. Contraception 2006;74:451–7.
9. Dieben TO, Roumen FJ, Apter D. Efficacy, cycle control, and user acceptability of a novel combined contraceptive vaginal ring. Obstet Gynecol 2002;100:585–93.
10. Sabatini R, Cagiano R. Comparison profiles of cycle control, side effects and sexual satisfaction of three hormonal contraceptives. Contraception 2006;74:220–3.
11. Archer DF, Bigrigg A, Smallwood GH, Shangold GA, Creasy GW, Fisher AC. Assessment of compliance with a weekly contraceptive patch (Ortho Evra/Evra) among North American women. Fertil Steril 2002;77:S27–31.
12. 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.
13. Thurman AR, Hammond N, Brown HE, Roddy ME. Preventing repeat teen pregnancy: postpartum depot medroxyprogesterone acetate, oral contraceptive pills, or the patch? J Pediatr Adolesc Gynecol 2007;20:61–5.
14. Murphy PA, Brixner D. Hormonal contraceptive discontinuation patterns according to formulation: investigation of associations in an administrative claims database. Contraception 2008;77:257–63.
15. Huber LRB, Hogue CJ, Stein AD, Drews C, Zieman M, King J, et al.. Contraceptive use and discontinuation: findings from the contraceptive history, initiation, and choice study. Am J Obstet Gynecol 2006;194:1290–5.
16. Secura GM, Allsworth JE, Madden T, Mullersman JL, Peipert JF. The Contraceptive CHOICE Project: reducing barriers to long-acting reversible contraception. Am J Obstet Gynecol 2010;203:115.e1–7.
17. Nelson AL, Pietersz D, Nelson LE, Aguilera L. Documented short-term continuation rates for combined hormonal contraceptives in an indigent population with ready access to contraceptive supplies. Am J Obstet Gynecol 2007;196:599.e1–5.
18. Frost JJ, Singh S, Finer LB. U.S. women's one-year contraceptive use patterns, 2004. Perspect Sex Reprod Health 2007;39:48–55.
19. World Health Organization. Selected Practice Recommendations for Contraceptive Use. 2nd ed. Geneva (Switzerland): World Health Organization; 2004.
20. 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.
21. Winner B, Peipert JF, Zhao Q, Buckel C, Madden T, Allsworth JE, et al.. Effectiveness of long-acting reversible contraception. N Engl J Med 2012;366:1998–2007.