Long-acting reversible contraceptive (LARC) methods, including the subdermal implant and intrauterine device (IUD), may be highly desired by women who have recently had an induced abortion1 and are associated with high continuation and satisfaction.2–4 Moreover, compared with women using other postabortion contraception, those choosing LARC have lower repeat abortion rates.5–7
However, the high upfront costs of LARC often prevent clinics from providing them immediately postabortion and serve as a barrier to women desiring these methods.8–13 When the cost of LARC is waived or subsidized, both women who have14 and have not15–17 recently had an abortion often choose these methods, yet no studies have isolated the effect of cost by directly comparing postabortion LARC use among women who can and cannot obtain these methods for free.
Our study takes advantage of a specialized county-based funding program offering no-cost LARC to uninsured, low-income women. Recent Texas legislation has resulted in significant reductions of state family planning funds18,19 and abortion facilities.20 Thus, many women have limited access to contraception and must travel cross-county to obtain abortion care. In this unusual setting, we were able to prospectively evaluate a natural experiment comparing women who had access to immediate postabortion no-cost LARC (Travis County residents) with women who were ineligible for the program (residents of other counties). Our objectives were to evaluate: 1) preabortion contraceptive preference for LARC, 2) immediate postabortion LARC use and 1-year continuation, 3) contraceptive use among non-LARC users in the year after abortion, and 4) pregnancy risk in this period according to eligibility for the funding program.
MATERIALS AND METHODS
Medicaid 1115 waiver programs use a combination of local and federal funds to pay for innovative and cost-effective health care programs for low-income, uninsured populations. In 2014, a multi-institution health delivery system in Travis County (located in south–central Texas and encompassing Austin) began using 1115 waiver funding to completely subsidize LARC for low-income women in a program we refer to as the LARC Access Program. Eligibility for the LARC Access Program was restricted to uninsured Travis County residents with household incomes less than 200% of the federal poverty level (from here on referred to as low-income eligible). Ineligible participants were divided into two categories: 1) low-income, uninsured, women who did not live in Travis County (referred to as low-income ineligible); and 2) higher income or insured women (referred to as higher income).
Between October 2014 and March 2015, we recruited women seeking abortion care from Planned Parenthood of Greater Texas in Austin to participate in this prospective study. The study site was the only facility in Travis County with the ability to provide immediate postabortion no-cost LARC under the LARC Access Program. Study participants were 18–44 years old, stated a desire to use contraception after their abortion, did not intend to become pregnant for at least the next year, and spoke English or Spanish.
Each participant received standardized, comprehensive contraceptive counseling before her abortion as described in the literature.21 Using a tiered-effectiveness model,22 unbiased information about the effectiveness, advantages, and disadvantages of each reversible method was provided. Cost was discussed after a method was chosen or sooner if brought up by the patient. Low-income eligible participants were told about the availability of immediate postabortion LARC free of charge. Program-ineligible participants were also counseled about the immediate availability of these methods at a cost of $952 for the levonorgestrel IUD (Mirena), $636 for the copper IUD (Paragard), and $985 for the etonogestrel implant (Nexplanon). When requested, LARC methods were placed immediately postabortion for those who were eligible for the LARC Access Program, had insurance coverage, or had the ability to cover the unsubsidized cost. For those requesting other methods, the contraceptive injection was administered on the day of the abortion procedure at a cost of $87 or a 1-month supply and prescription for oral contraceptive pills, the contraceptive patch, or the vaginal ring were provided. Otherwise, patients were offered male condoms and emergency contraception if desired.
After providing informed consent, study participants completed a baseline structured survey before their abortion. Despite attempts to recruit participants before contraceptive counseling, logistic constraints often precluded this. The baseline survey included questions about demographic characteristics and obstetric history. Preabortion contraceptive preference was assessed with the survey question; “If you could use any birth control method you wanted, which method would you be most interested in using after your abortion?” Response options included all hormonal and nonhormonal methods and an option of “no method.” Participants were then asked about the main reason for their interest in that method. Participants who would not consider using LARC were asked about the primary reason for that response.
Data were electronically entered into REDCap, an online data collection tool, using an iPad by the study participant or by a trained bilingual interviewer depending on participant preference. Data on the contraceptive method provided to each study participant either immediately postabortion or at the 2-week follow-up visit were provided by the clinic's sole provider (A.D.) and also entered into REDCap.
For the follow-up telephone surveys, a trained interviewer read the structured survey questions and response options to each participant and then recorded the results in REDCap. Follow-up surveys were scheduled to occur between February 2015 and March 2016 at 4, 8, and 12 months after the baseline interview to assess continuation of the contraceptive method provided immediately postabortion, current contraceptive use, and repeat pregnancy. For those who could not be reached by telephone, text message, or using alternate contacts after several attempts, or for those who preferred to complete follow-up interviews by email, an identical structured follow-up survey was sent to the email address provided by the participant through a secure, password-protected online link. Some participants were contacted multiple times before actually completing the survey. The distribution of follow-up surveys according to the time elapsed since the baseline interview ranged between 3 and 17 months.
Participants received $10 for completing the baseline survey and $15–20 for completing each of the three follow-up surveys. Institutional review board approval for this study was obtained from the University of Texas at Austin.
Power calculations indicated that a total sample size of 256 participants would be sufficient to detect a 20% difference in postabortion LARC use (based on an estimate of 45% postabortion LARC among low-income eligible participants and 25% for low-income ineligible participants) at an α level of 0.05 and power of 0.9 for a two-tailed test. However, because we expected to enroll nearly as many high-income as low-income participants, we set our total target sample size at 500.
We assessed differences in baseline characteristics between low-income eligible and ineligible participants and among all three eligibility groups using t tests and one-way analysis of variance for age and χ2 tests for race–ethnicity, parity, relationship status, education, previous abortion, contraceptive preferences, and postabortion contraceptive use. We used Poisson regression models with robust standard errors to assess whether any difference in LARC use between eligibility groups might have been the result of potential confounding variables (age, race–ethnicity, education, and preabortion contraceptive preferences).
The interval between follow-up and baseline surveys was calculated and then grouped as follows: 3–6, 7–10, and 11–14 months after baseline. Surveys completed 15 months or greater postabortion were excluded from the following analyses. For each follow-up interval, we used current status information from the follow-up surveys to calculate the proportion of participants who had immediate postabortion LARC placement and continued to use LARC. We also calculated the proportion of participants using a given contraceptive method to determine contraceptive method mix by eligibility group at successive points.
Finally, we used Cox proportional hazard models to estimate hazard ratios for risk of repeat pregnancy over the follow-up period among the three eligibility groups. All statistical analyses were performed using Stata 14.
Of those approached to participate in the study, 50 declined, 101 had incomplete baseline interviews primarily because they were called to their appointment before survey completion, and 121 did not meet the eligibility criteria. In total, 518 women completed the baseline survey among whom 315 (61%) completed an interview at 3–6 months, 357 (69%) completed an interview at 7–10 months, and 375 (72%) completed an interview at 11–14 months.
Among the baseline study participants, 143 (28%) were low-income eligible, 112 (22%) were low-income ineligible, and 263 (51%) were higher income. We found no significant differences in age, relationship status, or previous abortion among the three LARC Access Program eligibility groups. However, low-income eligible participants were more likely to be Hispanic compared with the two ineligible groups. Higher income participants were less likely to have any living children compared with the two other groups. Also, low-income ineligible participants were less likely to have completed college or attended postgraduate education compared with the other two groups (Table 1).
The majority of participants reported a preabortion contraceptive preference for LARC use (Table 2). The most important reasons why participants stated a preference for LARC included not having to remember to use these methods regularly (25%), that they last several years (25%), and that they were very effective at preventing pregnancy (14%). Among those who would not consider using LARC, the most common reasons included discomfort about having a foreign body in place (34%), concerns about side effects (19%), and concerns about safety (8%).
Low-income eligible participants with a preference for LARC were far more likely to receive postabortion LARC compared with low-income ineligible and higher income participants (Table 2). After adjusting for age, race–ethnicity, and education, low-income eligible participants had a 10-fold greater incidence of receiving postabortion LARC compared with low-income ineligible participants (incidence rate ratio 10.13, 95% confidence interval [CI] 4.68–21.91).
Continuation of LARC placed postabortion was high in the both low-income eligible group and high-income ineligible group. At 3–6 months, 94% (95% CI 88–101%) of low-income eligible and 91% (95% CI 82–100%) of higher income participants were still using the LARC method they received immediately postabortion. At 7–10 months postabortion, 93% (95% CI 87–99%) of low-income eligible and 87% (95% CI 77–96%) of higher income participants were still using LARC. By 11–14 months postabortion, continuation remained high and comparable between these two groups with 90% (95% CI 82–97%) and 86% (95% CI 76–97%) of low-income eligible and higher income participants, respectively, continuing to use LARC.
These high LARC continuation rates as well as the substantial variation in postabortion LARC use across eligibility groups are reflected in the mix of contraceptive methods used at the three durations (Fig. 1). Use of LARC among low-income ineligible participants remained steadily low at each follow-up interval, whereas use of less effective methods increased at each interval, mostly at the expense of short-term hormonal methods. Additionally, a higher proportion of low-income ineligible participants reported using no method of contraception at each follow-up interval compared with participants in the other two groups.
The risk of pregnancy over the follow-up period was significantly higher for low-income ineligible compared with eligible participants. The risk of pregnancy was also higher for the higher income group compared with the low-income eligible group, but not significantly so (Table 3).
We observed far higher uptake of LARC immediately after abortion among low-income women eligible for the LARC Access Program than among low-income women ineligible for the program. Long-acting reversible contraception uptake was also greater among low-income eligible women than among high-income ineligible women who either had insurance or whose incomes exceeded 200% of the federal poverty level. These differences in uptake greatly exceeded the differences in stated preference for LARC, which was high across all three eligibility groups.
The differences in postabortion LARC uptake were reflected in LARC use throughout the year after induced abortion through high continuation rates. Overall, low-income eligible women used a more effective mix of contraceptive methods, especially in comparison with low-income ineligible women. The observed differences in method mix by eligibility group were, in turn, reflected in pregnancy rates during the year after abortion with low-income eligible women having significantly lower pregnancy rates than low-income ineligible women.
Our study was a natural rather than a randomized experiment. Although we did observe differences between the low-income eligible participants who resided in Travis County and those who were ineligible on account of living in another county, our multivariable analysis indicated that those differences did not have a substantial influence on LARC uptake. The difference between the two groups resulted mainly from the likelihood of using LARC given a preference rather than from a difference in preference between the groups. Moreover, the somewhat smaller stated preference for LARC among low-income ineligible women might have resulted from a difference in counseling or perceived accessibility rather than a difference in underlying preference.
Our findings of increased LARC use among those who can obtain these methods without cost-sharing are consistent with prior work, although few data come from the abortion setting. A New Zealand study comparing 500 women at two time points, before and during an intervention when LARC fees were waived, demonstrated that postabortion IUD use increased from 26% (129/500) to 49% (243/500), respectively.14 Similar findings have been shown in nonabortion settings. In a retrospective study evaluating contraceptive use 2 years before and after a Kaiser Foundation Health Plan in California policy change to include 100% universal coverage for the most effective methods, provision of cost-free LARC increased from 4,884 per 2,319,493 (0.2%) to 9,771 per 2,654,707 (0.4%) contraceptives procured.15 In the Contraceptive CHOICE Project, where all contraceptive methods were offered free of charge to adolescents and women at risk for unintended pregnancy living in the St Louis, Missouri, region, 75% (6,928/9,256) of participants requested LARC.16 Among privately insured women requesting IUD placement from a university hospital in Philadelphia, 54% (19/35) of those with out-of-pocket expenses less than $50 had the device placed compared with 8% (5/60) who had out-of-pocket expenses of $50 or more.23
Our study has several strengths. Because all women in this study were primarily presenting for abortion care and not solely seeking contraception, we were able to isolate the effect of cost on LARC use without potential selection bias resulting from a predetermined desire for LARC use. Another strength of our study is the high retention rate of participants, which is in line with the highest follow-up rates obtained in prospective studies of postabortion patients.9,24
Despite these strengths, our study also has limitations. It is based on the experience of patients undergoing abortion at a single clinic, and our results cannot be generalized to all abortion facilities. Although cost is a significant barrier to provision of immediate postabortion LARC, other barriers include lack of sufficiently trained providers, limited array of available contraceptive methods, and inadequate contraceptive counseling.9,25 All of these barriers were obviated in our study, but they may still prevent provision of desired postabortion LARC at other clinics, even if cost barriers are removed. Indeed, such barriers likely explain the lower use of LARC among high-income ineligible women, although our study did not assess barriers encountered in that group.
A larger question concerns the overall nature of the LARC Access Program. Providing a large subsidy for LARC methods may have created an incentive for LARC use greater than that which would arise from removing the cost barrier for all methods of contraception.
Our results do, however, indicate high demand for postabortion LARC among low-income, uninsured women. In states like Texas where abortion providers are prohibited from many funding programs,18,19 uninsured low-income women who desire LARC postabortion are effectively excluded from accessing this method, leaving them at high risk for repeat unintended pregnancy.26 Specialized funding programs such as the one evaluated in this study may play an important role in bridging the gap for these women.
1. Madden T, Secura GM, Allsworth JE, Peipert JF. Comparison of contraceptive method chosen by women with and without a recent history of induced abortion. Contraception 2011;84:571–7.
2. McNicholas C, Hotchkiss T, Madden T, Zhao Q, Allsworth J, Peipert JF. Immediate postabortion intrauterine device insertion: continuation and satisfaction. Womens Health Issues 2012;22:e365–9.
3. Madden T, Eisenberg DL, Zhao Q, Buckel C, Secura GM, Peipert JF. Continuation of the etonogestrel implant in women undergoing immediate postabortion placement. Obstet Gynecol 2012;120:1053–9.
4. Mark A, Sonalkar S, Borgatta L. One-year continuation of the etonogestrel contraceptive implant in women with postabortion or interval placement. Contraception 2013;88:619–23.
5. Rose SB, Lawton BA. Impact of long-acting reversible contraception on return for repeat abortion. Am J Obstet Gynecol 2012;206:37. e1–6.
6. Goodman S, Hendlish SK, Reeves MF, Foster-Rosales A. Impact of immediate postabortal insertion of intrauterine contraception on repeat abortion. Contraception 2008;78:143–8.
7. Ames CM, Norman WV. Preventing repeat abortion in Canada: is the immediate insertion of intrauterine devices postabortion a cost-effective option associated with fewer repeat abortions? Contraception 2012;85:51–5.
8. Kavanaugh ML, Jones RK, Finer LB. Perceived and insurance-related barriers to the provision of contraceptive services in U.S. abortion care settings. Womens Health Issues 2011;21:S26–31.
9. Rocca CH, Thompson KM, Goodman S, Westhoff CL, Harper CC. Funding policies and postabortion long-acting reversible contraception: results from a cluster randomized trial. Am J Obstet Gynecol 2016;214:716.e1–8.
10. Thompson KM, Speidel JJ, Saporta V, Waxman NJ, Harper CC. Contraceptive policies affect post-abortion provision of long-acting reversible contraception. Contraception 2011;83:41–7.
11. Stanek AM, Bednarek PH, Nichols MD, Jensen JT, Edelman AB. Barriers associated with the failure to return for intrauterine device insertion following first-trimester abortion. Contraception 2009;79:216–20.
12. Morse J, Freedman L, Speidel JJ, Thompson KM, Stratton L, Harper CC. Postabortion contraception: qualitative interviews on counseling and provision of long-acting reversible contraceptive methods. Perspect Sex Reprod Health 2012;44:100–6.
13. 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.
14. Rose SB, Lawton BA, Brown SA. Uptake and adherence to long-acting reversible contraception post-abortion. Contraception 2010;82:345–53.
15. Postlethwaite D, Trussell J, Zoolakis A, Shabear R, Petitti D. A comparison of contraceptive procurement pre- and post-benefit change. Contraception 2007;76:360–5.
16. Peipert JF, Madden T, Allsworth JE, Secura GM. Preventing unintended pregnancies by providing no-cost contraception. Obstet Gynecol 2012;120:1291–7.
17. Fanarjian N, Drostin C, Garrett J, Montalvo A. Does the provision of free intrauterine contraception reduce pregnancy rates among uninsured low-income women? A cohort study: a two North Carolina clinics. Contraception 2012;85:160–5.
18. White K, Grossman D, Hopkins K, Potter JE. Cutting family planning in Texas. N Engl J Med 2012;367:1179–81.
19. White K, Hopkins K, Aiken AR, Stevenson A, Hubert C, Grossman D, et al. The impact of reproductive health legislation on family planning clinic services in Texas. Am J Public Health 2015;105:851–8.
20. Grossman D, Baum S, Fuentes L, White K, Hopkins K, Stevenson A, et al. Change in abortion services after implementation of a restrictive law in Texas. Contraception 2014;90:496–501.
21. Madden T, Mullersman JL, Omvig KJ, Secura GM, Peipert JF. Structured contraceptive counseling provided by the Contraceptive CHOICE Project. Contraception 2013;88:243–9.
23. Gariepy AM, Simon EJ, Patel DA, Creinin MD, Schwarz EB. The impact of out-of-pocket expense on IUD utilization among women with private insurance. Contraception 2011;84:e39–42.
24. Foster DG, Steinberg JR, Roberts SC, Neuhaus J, Biggs MA. A comparison of depression and anxiety symptom trajectories between women who had an abortion and women denied one. Psychol Med 2015;45:2073–82.
25. Sonfield A, Gold RB. Beyond consent to sterilization: facing up to the full range of barriers to post-abortion contraception. Am J Obstet Gynecol 2014;211:3–4.
26. Jones RK, Singh S, Finer L, Frohwirth L. Repeat abortion in the United States. New York (NY): Guttmacher Institure; 2006.