Unintended pregnancies, pregnancies that are unwanted or mistimed at conception, are a costly public health problem. U.S. taxpayers pay approximately $11 billion dollars annually in costs associated with 1 million unintended births.1 The unintended pregnancy rate in the United States is significantly higher than in other developed countries.2 In the 2006–2008 cycle of the National Survey of Family Growth, 49% of all pregnancies were reported to be unintended. Of these, 29% were mistimed, 19% were unwanted, and 43% ended in abortion.3 Approximately half of unintended pregnancies result from nonuse of contraception, and half result from inconsistent or incorrect use and contraceptive failure.4
Intrauterine devices (IUDs) and contraceptive implants, collectively referred to as long-acting reversible contraceptive (LARC) methods, are highly effective and safe and have high satisfaction and continuation rates.5 However, these methods are underutilized in the United States. In other developed countries where IUDs are used more frequently, unintended pregnancy rates are lower.6 In fact, in a study examining global rates of unintended pregnancy between 1995 and 2008, North America was the only region of the world that did not see a decrease in the rate of unintended pregnancy.2 The most commonly used reversible contraceptive method in the United States is the oral contraceptive pill (OCP), despite its “typical use” failure rate of 8–9% per year.7 However, IUDs and implants are more than 20 times more effective at preventing pregnancy than are OCPs, the contraceptive patch, and the contraceptive vaginal ring.8
The objective of the Contraceptive CHOICE Project was to promote the use of LARC methods and to provide no-cost contraception to a large number of women and adolescents in our region in an effort to reduce unintended pregnancies. The primary population-based outcomes for our study were the rate of teen births and the percentage of abortions that were repeat abortions. We also estimated abortion rates in our metropolitan area and in our cohort and compared these rates with U.S. and regional abortion rates.
PATIENTS AND METHODS
We designed a prospective cohort study with two objectives: 1) to promote the use of the most effective contraceptive methods (IUDs and implants) and 2) to provide contraception at no cost to 10,000 female participants in our region who were at risk for unintended pregnancy.9 Before initiating recruitment, we obtained approval from the Washington University Human Research Protection Office.
CHOICE participants met the following inclusion criteria: 1) age 14–45 years, 2) desired reversible contraception, 3) not currently using a method or willing to switch to a new reversible contraceptive method, 4) no desire for pregnancy for at least 12 months, 5) currently sexually active or plan to be sexually active with a male partner within the subsequent 6 months, 6) residence in the St. Louis region, and 7) English or Spanish speaking. We excluded potential participants if they were surgically sterile. CHOICE participants were recruited by provider referral, newspaper reports and advertisements, study flyers, and word of mouth. In addition, we recruited eligible participants from local clinics and the two main abortion facilities in our region. All participants received the reversible contraceptive method of their choice at no cost for 3 years (first 5,090 participants) or 2 years (remainder of cohort). After completion of the study, participants could continue their IUD or implant, because these methods last 3 years (implant) to 10 years (copper IUD), but could no longer obtain contraception at no cost or change methods as part of the project.
All participants were read a brief script informing them of the effectiveness and safety of LARC methods at initial contact and completed an in-depth, evidence-based contraceptive counseling session at enrollment.10 Participants were offered all U.S. Food and Drug Administration (FDA)–approved contraceptive methods and could choose any method. Each participant provided written informed consent and was monitored prospectively for the duration of follow-up.
Our primary population-based outcomes included teenage births and repeat abortions as proxies for unintended pregnancies. Because participant recruitment sites included two regional abortion facilities, we believed our greatest population effect would be on repeat abortions, or the percentage of abortions that are performed in adolescents and women with a previous abortion. Thus, one of our primary outcomes of interest was the percentage of abortions that were repeat abortions. We compared repeat abortion data in the St. Louis region with that in Kansas City, Missouri, and nonmetropolitan Missouri. Kansas City is of similar size and demographic profile to St. Louis (Table 1).
Although not an a priori primary outcome of interest, we also estimated abortion rates because the majority of abortions result from unintended pregnancies. We obtained abortion data from two sources. Reproductive Health Services of Planned Parenthood is the major abortion provider in the St. Louis area and represents 90% of the abortions among St. Louis-area females reported to the Missouri Department of Health and Senior Services. We obtained the total number of abortions per year among Missouri residents served at Reproductive Health Services grouped by county-level residence. We also obtained the total number of abortions per year and the number of abortions among females who reported a prior abortion at the time of the current abortion at the state-level from the Missouri Department of Health and Senior Services. The Missouri Department of Health and Senior Services also maintains reciprocal vital statistics reporting with states to obtain information on abortions that occur in Missouri residents at facilities located outside of Missouri. The Missouri Department of Health and Senior Services reported the total and repeat abortion numbers by four geographic locations based on patient ZIP code and grouped by St. Louis City or County (our study catchment area), Kansas City, and all remaining ZIP codes grouped as nonmetropolitan Missouri.
Abortion rates among participants aged 15–44 years and births among participants aged 15–19 years within CHOICE were compared with regional and national rates. Because the CHOICE cohort represents a higher-risk population (median age of 25 years and 50% black) than the general population, we standardized the CHOICE abortion rate to the age and racial (black and white) distribution of females who reside in the St. Louis region using data from the 2010 U.S. Census (direct standardization). We compared the CHOICE standardized rate with the St. Louis regional rate using data from the Missouri Department of Health and Senior Services and with the national rate using the most recent published data from 2008.11
Statistical analyses were performed using STATA software. The significance level was set at 0.05. Means, standard deviations, frequencies, and percentages were used to describe the demographic characteristics of the study participants. We calculated 95% confidence intervals around the percentage of abortions that were repeat for each year from 2006 through 2010. We used a Pearson χ2 to determine a significant difference in the proportion of repeat abortions each year between the St. Louis region and Kansas City and a Mantel-Haenszel score test for trend of odds of repeat abortions for 2006–2010 for the St. Louis , Kansas City, and nonmetropolitan Missouri regions, individually. We used negative binomial regression models to estimate the time trend for total abortions in the St. Louis region and in non-St. Louis Missouri. We estimated the population-attributable risk by calculating the risk difference between CHOICE and the St. Louis region rates and multiplying by the St. Louis region population divided by 1,000, and we calculated the number needed to treat by taking the inverse of the absolute risk reduction. Confidence intervals for the number needed to treat were estimated using the Bender method.12
For sample size calculations, we considered a 50–60% reduction within the study population or a 6–7% St. Louis region reduction in teen births (baseline rate 39 per 1,000 15–19-year-old females) and repeat abortion (baseline rate 430 per 1,000 abortions) to be of clinical and public health significance. Further, we assumed no change in the remainder of the St. Louis population not participating in the study, an alpha of 0.05, and power of 80%. Our intent was to interact with 11% of the population in the region at greatest risk for unintended pregnancy. Further, we estimated that 10% of the cohort would choose an IUD and 3% of the cohort would choose an implant, a rate almost double that of IUD and implant use in the United States in 2008.7 Based on these assumptions, an enrollment of 2,000 participants aged 15–19 years would represent 11% of the population at risk for teenage pregnancy and an enrollment of 5,000 participants with a history of abortion would reach 11% of the population at risk of repeat abortion. The uptake of IUD and implant was much greater than originally predicted,9 leading to greater observed reductions in pregnancy risk among the accrued study population than anticipated, allowing us to exceed our initial targets for population effect.
Between August 2007 and September 2011, 9,256 adolescents and women enrolled in CHOICE; of these participants, 16% were recruited at the abortion facilities. The baseline demographic and reproductive characteristics of the cohort are provided in Table 2. The mean age of the total population was 25 years; 51% were black, 35% had a high school education or less, 37% received public assistance, and 39% had trouble paying for basic expenses. Forty-seven percent were nulliparous, 63% had a prior unintended pregnancy. Participants chose the following contraceptive methods at baseline: 46% levonorgestrel IUD, 12% copper IUD, 17% subdermal implant, 9% OCPs, 7% contraceptive vaginal ring, 7% depot medroxyprogesterone acetate, and 2% contraceptive patch. Thus, 75% of our study population chose a LARC method.
Participants recruited at the abortion clinics were more likely to be black and to report a high school education or less, trouble paying for basic necessities, receipt of public assistance or no insurance, and greater parity and a history of three or more unintended pregnancies. We also found that these participants were more likely to choose a LARC method at enrollment compared with adolescents and women enrolled at the other recruitment clinics (84.5% compared with 72.9%, P<.001).
We evaluated teenage birth (births per 1,000 females aged 15–19) as a proxy for unintended pregnancy because up to 80% of these births are unintended.3,13 The number of teenage births in the United States has decreased markedly in the past several years, with a 44% drop between 1991 and 2010 to a level of 34.3 per 1,000.14 The birth rate among participants aged 15–19 years within the CHOICE cohort was 6.3 per 1,000, a rate far below the national level.
Figure 1 shows the number of abortions among Missouri residents reported by Reproductive Health Services stratified by region (St. Louis City and County compared with the rest of Missouri). Between 2008 and 2010, the number of abortions performed at Reproductive Health Services among women and teenagers who resided in St. Louis City and County declined by 20.6% (P<.001), compared with no appreciable change (0%, P=.39) in the number of abortions among women and teenagers who resided in the rest of Missouri.
Our primary outcome of interest was the percentage of abortions that were repeat abortions for the following reasons: 1) this information is tracked by providers and in government statistics and 2) our objective was to have the maximum population effect through the provision of our intervention to participants at highest risk for unintended pregnancy. Women and teenagers seeking pregnancy terminations are at risk for future unintended pregnancy and repeat abortion as well as potentially motivated to seek contraceptive services. Using vital statistics data from the state health department, Figure 2 shows a significant difference in the proportion of repeat abortions between the St. Louis region and Kansas City in 2009 (P=.02) and 2010 (P<.01). We also detected a significant decline in the proportion of repeat abortions over time in the St. Louis region (P=.002).
Between 2008 and 2010, abortion rates in CHOICE participants ranged from 4.4 to 7.5 per 1,000 after adjusting for age and race (Table 3). These rates are considerably less than the rates in St. Louis City and County for the same years (P<.001) and far below the national rate of 19.6 per 1,000. Using these data, we then estimated the difference in abortion rates and number of abortions prevented each year if CHOICE were available to the entire population of the region. Based on the number needed to treat, one abortion could be prevented for every 79–137 women and teenagers provided the CHOICE intervention.
The Institute of Medicine recently recommended that eight primary preventive health services for women be covered without cost to patients under the Patient Protection and Affordable Care Act of 2010.15 Among these eight services, the Institute of Medicine recommended “a fuller range of contraceptive education, counseling, methods, and services so that women can better avoid unwanted pregnancies and space their pregnancies to promote optimal birth outcomes.” As a result, all FDA-approved contraceptive methods would be covered without cost. The Contraceptive CHOICE Project essentially simulated this recommendation in our region for reversible contraceptive methods. All women received any FDA-approved contraceptive method of their choice as well as the ability to change methods at no cost. In addition, the project provided education to promote the use of the most effective contraceptive methods, IUDs, and implants in an effort to alter population outcomes.
There are few studies that have investigated whether increasing the uptake of LARC methods decreases unintended pregnancy. A recent analysis of the family planning expansion program in California known as the Family Planning, Access, Care and Treatment Program examined the association between increased contraceptive access and unintended pregnancy.16 Using a Markov model, the authors estimated that increasing contraceptive access to low-income women averted more than 286,000 unintended pregnancies. Interestingly, uptake of LARC was substantially lower among Family Planning, Access, Care and Treatment Program participants than among CHOICE participants; however, the increased access to no-cost contraception still resulted in a significant reduction in the number of unintended pregnancies. A qualitative study of women undergoing abortion found that there was a high level of interest in LARC methods for postabortion contraception, but the majority of women also identified cost as a barrier to obtaining a LARC method.17 Participation in CHOICE provided participants with access to IUDs and implants that they otherwise might not have had, including teenagers and women recently postabortion who are at high risk for repeat unintended pregnancy.
The birth rate to females 15–19 years of age in our cohort is markedly less than the U.S. rate despite a remarkable decline in teenage births nationally.14 We also noted a substantially lower abortion rate in our cohort compared with regional and national statistics and a significant decline in repeat abortions in the St. Louis region. Increasing access to the most effective contraceptive methods by removing cost and access as barriers has greatly increased the number of adolescents and women in the St. Louis region using the most effective methods of birth control. Providing no-cost contraception and promoting the use of highly effective contraceptive methods has the potential to reduce unintended pregnancies in the United States. In fact, based on our calculations in Table 3, changes in contraceptive policy simulating the Contraceptive CHOICE Project would prevent as many as 62–78% of abortions performed annually in the United States.
The strengths of our study include the prospective design of the Contraceptive CHOICE Project—a large sample with a high uptake of the most effective contraceptive methods—the use of systematically collected state-mandated data, and partnership with community clinics serving adolescents and women at highest risk for unintended pregnancy. Our study also has several limitations. Intendedness of pregnancy is not captured in the state vital statistics; therefore, we used proxy markers of unintended pregnancy. However, teenage births, abortions, and repeat abortions are indicators of unintended pregnancy.18 Also, we previously compared CHOICE participants with National Survey of Family Growth participants and found that they were similar except that CHOICE participants were more likely to be black, single, and have a lower income,19 all characteristics associated with an increased risk for unintended pregnancy.3 Thus, although our CHOICE cohort may not be similar demographically to many other geographic areas, limiting the study’s generalizability, CHOICE participants are at high risk for unintended pregnancy. Another limitation is that the analysis comparing repeat abortion in the St. Louis region with that in Kansas City and nonmetropolitan Missouri is essentially an ecological study. There may be several factors that affect the rates of repeat abortion, such as the economic recession, federal changes in Title X funding for family planning, and Missouri state laws that limit access to abortion. It is not possible to conclude that the changes observed in repeat abortion were due solely to the Contraceptive CHOICE Project. However, the weight of the evidence, including a marked reduction in teenage births and abortion rates compared with regional and national statistics, provides evidence of a population effect of the CHOICE intervention.
Unintended pregnancy remains a stubborn problem in the United States, with higher proportions among adolescents and young women, racial and ethnic minorities, and women with less education and lower socioeconomic status.3 Approximately half of unintended pregnancies are the result of contraceptive failure,20 with the majority of women using reversible contraception using OCPs or condoms.7 The American College of Obstetricians and Gynecologists and many family planning experts believe that LARC methods should be first-line contraceptive options and that increased uptake of LARC methods is essential to decreasing the rate of unintended pregnancy.21–23 In addition, because LARC methods have been shown to have higher continuation rates than other reversible methods, the number of adolescents and women using no contraception would decline, further decreasing the unintended pregnancy rate.5 Increased access to contraception, particularly highly effective LARC methods, and providing contraception at no cost may result in a significant decrease in the rate of unintended pregnancy in the United States.
1. Gold RB, Sonfield A. Publicly funded contraceptive care: a proven investment. Contraception 2011;84:437–9.
2. Singh S, Sedgh G, Hussain R. Unintended pregnancy: worldwide levels, trends, and outcomes. Stud Fam Plann 2010;41:241–50.
3. Finer LB, Zolna MR. Unintended pregnancy in the United States: incidence and disparities, 2006. Contraception 2011;84:478–85.
4. Frost JJ, Darroch JE, Remez L. Improving contraceptive use in the United States. Issues Brief (Alan Guttmacher Inst) 2008;1:1–8.
5. Peipert JF, Zhao Q, Allsworth JE, et al.. Continuation and satisfaction of reversible contraception. Obstet Gynecol 2011;117:1105–13.
6. Bajos N, Leridon H, Goulard H, Oustry P, Job-Spira N. Contraception: from accessibility to efficiency. Hum Reprod 2003;18:994–9.
7. Mosher WD, Jones J. Use of contraception in the United States: 1982-2008. Vital Health Stat 2010;23:1–54.
8. 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.
9. 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.
10. Madden T, Mullersman JL, Omvig KJ, Secura GM, Peipert JF. Structured contraceptive counseling provided by the Contraceptive CHOICE Project. Contraception 2012; S0010–7824(12)00722–6 [Epub ahead of print].
11. Jones RK, Kavanaugh ML. Changes in abortion rates between 2000 and 2008 and lifetime incidence of abortion. Obstet Gynecol 2011;117:1358–66.
12. Bender R. Calculating confidence intervals for the number needed to treat. Control Clin Trials 2001;22:102–10.
13. Kissin DM, Anderson JE, Kraft JM, Warner L, Jamieson DJ. Is there a trend of increased unwanted childbearing among young women in the United States? J Adolesc Health 2008;43:364–71.
14. Hamilton BE, Ventura SJ. Birth rates for U.S. teenagers reach historic lows for all age and ethnic groups. NCHS data brief, no. 89. Hyattsville (MD): National Center for Health Statistics; 2012.
15. Committee on Preventive Services for Women, Institute of Medicine. Clinical preventive services for women: closing the gaps. Washington, DC: The National Academies Press; 2011.
16. Foster DG, Biggs MA, Rostovtseva D, de Bocanegra HT, Darney PD, Brindis CD. Estimating the fertility effect of expansions of publicly funded family planning services in California. Women Health Issue 2011;21:418–24.
17. Rose SB, Cooper AJ, Baker NK, Lawton B. Attitudes toward long-acting reversible contraception among young women seeking abortion. J Womens Health (Larchmt) 2011;20:1729–35.
18. Jones R, Singh S, Finer L, Frohwirth L. Repeat abortion in the United States: occasional report. New York (NY): Guttmacher Institute; 2006.
19. Kittur ND, Secura GM, Peipert JF, Madden T, Finer LB, Allsworth JE. Comparison of contraceptive use between the Contraceptive CHOICE Project and state and national data. Contraception 2011;83:479–85.
20. 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.
21. Blumenthal PD, Voedisch A, Gemzell-Danielsson K. Strategies to prevent unintended pregnancy: increasing use of long-acting reversible contraception. Hum Reprod Update 2011;17:121–37.
22. Cleland K, Peipert JF, Westhoff C, Spear S, Trussell J. Family planning as a cost-saving preventive health service. N Engl J Med 2011;364:e37.
23. Adolescents and long-acting reversible contraception: implants and intrauterine devices. Committee Opinion No. 539. American College of Obstetricians and Gynecologists. Obstet Gynecol 2012;120:983–8.