Madden, Tessa MD, MPH; Eisenberg, David L. MD, MPH; Zhao, Qiuhong MS; Buckel, Christina MSW; Secura, Gina M. PhD, MPH; Peipert, Jeffrey F. MD, PhD
Division of Clinical Research, Department of Obstetrics and Gynecology, Washington University in St. Louis School of Medicine, St. Louis, Missouri.
Corresponding author: Tessa Madden, MD, MPH, Department of Obstetrics and Gynecology, Washington University in St. Louis School of Medicine, 4533 Clayton Avenue, Box 8219, St. Louis, MO 63110; e-mail: firstname.lastname@example.org.
Supported in part by the Susan Thompson Buffett Foundation and award number K23HD070979 from the Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD). Its contents are solely the responsibility of the authors and do not necessarily represent the official view of NICHD or the National Institutes of Health.
Presented at the North American Forum on Family Planning, October 27-29, 2012, Denver, Colorado.
Financial Disclosure Dr. Madden is a former speaker for Bayer HealthCare Pharmaceuticals. The other authors did not report any potential conflicts of interest.
OBJECTIVE: To estimate whether women who receive an immediate postabortion etonogestrel subdermal implant have higher discontinuation rates than women who undergo an interval insertion (insertion not associated with a recent pregnancy).
METHODS: This was a secondary analysis of the Contraceptive CHOICE Project. We compared 12-month continuation rates for the subdermal implant between women undergoing immediate postabortion placement and women who underwent an interval placement. Continuation rates were compared using Kaplan-Meier estimator of survival function and Cox proportional hazard models.
RESULTS: Over the study period, 141 women received the implant immediately postabortion and 935 women underwent an interval placement. The 12-month continuation rates were similar: 81.5% among women undergoing immediate postabortion placement and 82.8% among women undergoing interval placement (P=.54). In the survival analysis, women who received an immediate postabortion implant were no more likely to discontinue their method than women who had an interval placement (adjusted hazard ratio 1.04, 95% confidence interval 0.68–1.60). The most common reason given for discontinuation was frequent or irregular bleeding, which was reported by 65.4% of women discontinuing in the immediate postabortion group and 56.7% in the interval group.
CONCLUSION: Women who undergo immediate postabortion placement of the subdermal implant have high rates of continuation that are equivalent to women undergoing interval insertion. Immediate postabortion long-acting reversible contraceptives should be available to women undergoing induced abortion to decrease the risk of repeat unplanned pregnancy.
LEVEL OF EVIDENCE: II
The rates of unintended pregnancy and abortion have not changed significantly over the past few decades and remain a public health challenge.1,2 The reasons for the high rate of unintended pregnancy and abortion are multifactorial, but it is clear that contraceptive failure, including incorrect or inconsistent use, is an important factor.3 Although nearly all women have used a contraceptive method, not all methods are continued at the same rate or have the same inherent effectiveness.4–6 Long-acting reversible contraceptive (LARC) methods, including the most effective reversible methods such as the intrauterine device (IUD) and subdermal implant, have the potential to affect the rate of both unintended pregnancy and abortion significantly.7
Repeat abortion accounts for 47% of all abortions in the United States8; therefore, it is important to provide access to LARC methods at the time of abortion to reduce this risk. In fact, one study demonstrated that women who received immediate postabortion IUDs were 63% less likely to experience a repeat abortion.9 A recently published prospective cohort study from New Zealand showed that women who initiated depo-medroxyprogesterone acetate or received an IUD immediately postabortion had significantly lower rates of repeat abortion up to 2 years later.10 We have previously shown that, compared with women without a recent abortion, women offered immediate postabortion LARC were three times more likely to choose IUDs and 50% more likely to choose implants.11
Choosing a contraceptive method at the time of abortion may influence continuation in a positive or negative direction. There is some concern among clinicians that women who receive an immediate postabortion etonogestrel implant may have higher discontinuation rates than women who have an interval insertion. We conducted this study to estimate whether women who received an immediate postabortion implant would have equivalent 12-month continuation rates as women who received the implant not in association with a recent pregnancy.
MATERIALS AND METHODS
This study is a secondary analysis of the Contraceptive CHOICE Project. CHOICE is a prospective cohort study designed to promote the use of LARC methods, to remove financial barriers, to contraception, and evaluate continuation and satisfaction for reversible methods.12 The objective of this analysis was to compare the 12-month continuation rates of the subdermal etonogestrel implant between women who had the implant placed immediately after a surgical abortion and women who underwent an interval placement. We defined interval placement as 6 weeks or greater from the time of the last pregnancy. We also evaluated continuation rates in women who had “delayed” insertion of an implant after an abortion, which was performed within 6 weeks but not on the same day as the abortion. This group included women undergoing both medical and surgical abortion.
Between August 2007 and September 2011, CHOICE enrolled 9,256 women. Women were eligible to participate if they were 14–45 years of age, resided in St. Louis City or County, had been sexually active with a male partner in the past 6 months or anticipated sexual activity in the next 6 months, had not had a tubal sterilization or hysterectomy, did not desire pregnancy in the next year, and were interested in starting a new reversible contraceptive method. Potential participants were read a short script about long-acting reversible contraception, which stated that IUDs and implants are the most effective methods. The script was read at the time of screening for study eligibility. Participants also underwent comprehensive contraceptive counseling, which emphasized the effectiveness of all reversible contraceptive methods. A participant could choose any U.S. Food and Drug Administration–approved reversible method of contraception as long as she was not currently using the method. A health care provider at our university research site or one of the community partner clinics would determine if the desired contraceptive method was appropriate for the participant according to evidence-based criteria. At baseline, participants provided written informed consent, completed a comprehensive survey, and received their desired reversible contraceptive method at no cost. Participants completed follow-up surveys by telephone at 3 and 6 months and every 6 months for the study duration (3 years for the first 5,090 participants and 2 years for the remainder of the cohort). We obtained approval from the Washington University in St. Louis School of Medicine Human Research Protection Office before recruitment of participants.
We compared demographic, socioeconomic, and behavioral characteristics of women who had been enrolled in the study for at least 12 months after implant placement. Comparisons were made using χ2 for categorical variables and analysis of variance test for continuous variables between women undergoing immediate postabortion, delayed, and interval placement. Twelve-month continuation rates were estimated using the Kaplan-Meier estimator of survival function. Participants who were lost to follow-up were censored at the time of the last completed survey. Hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) were calculated using Cox proportional hazard models. Satisfaction was measured on a 3-point scale (“very satisfied,” “somewhat satisfied,” and “not satisfied”). For this analysis, women who discontinued the implant were considered not satisfied.
We performed univariable and multivariable Cox proportional hazards regression to calculate the HR for factors associated with discontinuation. We planned to include immediate postabortion insertion in the multivariable model regardless of whether it was significant in the univariable model because this was our a priori hypothesis. All covariates that were associated with implant discontinuation in the univariable model (at the .05 level) or altered the HR by 10% or greater were included in the multivariable model. We performed all statistical analyses using STATA 11.
Between August 2007 and November 2010, there were 7,472 women enrolled into the CHOICE Project who had initiated their baseline method by 3 months and been enrolled in the study for at least 12 months. Of these, 1,178 women received the subdermal implant; 141 received the implant immediately postabortion, 102 women had a “delayed” placement, and 935 women had an interval placement. There were an additional 37 women in this cohort who underwent immediate postpartum implant placement and were excluded from this analysis because continuation of immediate postpartum implants may differ from immediate postabortion or interval placement.
Table 1 shows the demographic and reproductive characteristics for women who received an immediate postabortion implant, “delayed” implant, and interval placement. The three groups were similar across demographic and reproductive characteristics, although women who received the implant immediately postabortion were less likely to be Hispanic, have a college education, and report having health insurance. They were more likely to be single, to have higher parity, and report a history of abortion (before the index abortion). Overall, the 12-month continuation rate for the implant was high at 83.0%.
There were a total of 189 women who discontinued the implant during the first 12 months of use. The rates of continuation at 12 months are shown in Table 2 and were similar for the three groups: 81.5% of women who received the implant immediately postabortion were still using the implant at 12 months compared with 82.8% of women who underwent interval placement and 86.3% of women who underwent “delayed” placement. There was no statistically significant difference between these rates (P=.54). Satisfaction with the implant was also similar across the three groups with 73.2% of immediate postabortion implant users, 79.1% of women undergoing “delayed” placement, and 78.4% of women undergoing interval insertion reporting being “somewhat” or “very satisfied” with the implant at 12 months (P=.70). There were 87 women (7.4%) who had been lost to follow-up by 12 months. There was no further information about implant use available for these women. Loss to follow-up did not differ by the timing of implant placement (P=.46)
Table 3 shows the univariable and multivariable Cox proportional hazards model for factors associated with discontinuation of the implant. In the univariable analysis, women who received the implant immediately postabortion did not have a higher rate of discontinuation compared with women who underwent an interval placement (HR 1.15, 95% CI 0.76–1.74). Women who had a “delayed” placement were also no more likely to discontinue the implant compared with women who had an interval placement (HR 0.79, 95% CI 0.45–1.40). Of the demographic and reproductive characteristics described in Table 1, only separated, divorced, and widowed marital status; some college education; nulliparity; history of a prior abortion; and history of a sexually transmitted infection were significantly associated with discontinuation of the implant in the univariable model. Women who reported being separated, divorced, or widowed were twice as likely to discontinue their implant (HR 2.15, 95% CI 1.17–2.94). Nulliparous women and women with a history of a prior abortion or sexually transmitted infection were slightly more likely to discontinue the implant (HR 1.60, 95% CI 1.19–2.14; HR 1.39, 95% CI 1.01–1.91; and HR 1.39, 95% CI 1.04–1.86, respectively).
After adjusting for the factors significant in the univariable model, we still did not observe any association between immediate postabortion timing of implant placement and discontinuation (adjusted HR 1.04, 95% CI 0.68–1.60) compared with interval placement. In the adjusted model, separated, divorced, and widowed marital status; nulliparity; and history of a prior sexually transmitted infection remained significantly associated with discontinuation.
Table 4 shows the reported reasons for discontinuation of the implant. Women could provide up to three reasons. Most women provided a single reason for discontinuation (74.6%). There were 48 (19.6%) women who provided two reasons and 11 (5.8%) who provided three reasons for discontinuation. The most common reason given for discontinuation was frequent or irregular bleeding. Among discontinuers, 65.4% (17 of 26) of the immediate postabortion group, 76.7% (10 of 13) of the “delayed” group, and 56.7% (85 of 150) of the interval group reported frequent or irregular bleeding as the reason for discontinuation (P=.29). The second most common reason given for discontinuation was weight gain with 19.2% of discontinuers in the immediate postabortion group and 15.3% in the interval group reporting weight gain. The percentage of women discontinuing the implant because they were seeking pregnancy was similar between the immediate postabortion group and the interval group (3.8% compared with 4.7%).
We found that the continuation of the etonogestrel subdermal implant at 12 months does not differ among women undergoing immediate postabortion placement, delayed postabortion placement, and interval placement. In addition, the reasons reported for discontinuation were similar among the three groups. Our observed continuation rate of 83.0% overall is similar to the other published 12-month continuation rates of the etonogestrel implant.13,14 Few studies have specifically evaluated continuation of the implant after immediate postabortion insertion. Women undergoing immediate postabortion placement may have had less time to consider their contraceptive decision and may be more likely to regret their choice, leading to higher discontinuation rates. Recently, Mark et al15 presented findings from a study of 12-month continuation rates of the implant in women who underwent immediate postabortion insertion compared with women undergoing an interval insertion. The rate of discontinuation was higher among women undergoing immediate postabortion insertion compared with the interval group (31.4% compared with 19.2%), although this was not a statistically significant difference as a result of a small sample size. This discontinuation rate among the immediate postabortion placement group is quite a bit higher than our observed rate of 18.5%. We observed higher continuation rates in all three groups of implant users than have previously been reported in the same cohort among women using shorter-acting reversible contraceptive methods such as oral contraceptive pills, contraceptive vaginal ring, transdermal contraceptive patch, and depot medroxyprogesterone acetate with estimated 12-month continuation rates of 55%, 54%, 49%, and 57%, respectively.16
Strengths of our study include a relatively large sample size with a high rate of follow-up, racial and socioeconomic diversity, and systematic assessment of continuation and satisfaction. However, we recognize the limitations of our study. Our cohort was comprised of women willing to enroll in a contraceptive study from a single geographical area, which could limit generalizability. In addition, we did not randomize women to immediate postabortion compared with interval insertion. At baseline, these groups did differ in demographic and reproductive characteristics. We performed a multivariable Cox proportional hazard regression to adjust for these baseline differences. There were 141 women who received their implant immediately postabortion and a smaller number who discontinued the method. A post hoc sample size and power calculation indicates that we have greater than 90% power to detect a twofold increased risk of discontinuation with 141 immediate postabortion implant insertions and a 5:1 ratio of exposed to comparison participants. With relatively few discontinuers, we had limited ability to find differences in the reasons for discontinuation based on timing of the implant placement. However, the discontinuation rates of 18.5% in the immediate postabortion group and 17.2% in the interval insertion group appear quite similar and this difference was not statistically significant.
Availability of immediate postabortion contraception, especially LARC methods, varies depending on the abortion provider, upfront costs, and other institutional barriers.17 Rates of repeat abortion have previously been shown to be lower in women undergoing immediate postabortion IUD insertion.9 A recent study by Tocce et al18 found that postpartum adolescents who had an immediate postpartum implant placement had a 12-month continuation rate of 86%. The same study found that postpartum implant users were significantly less likely to have a repeat pregnancy within 12 months compared with women who chose another postpartum contraceptive method (2.9% compared with 21.3%). It is reasonable to extrapolate that a reduction in repeat unintended pregnancy would also be observed for women undergoing immediate postabortion implant placement if continuation rates are high. Our results show that these women have equivalent rates of continuation at 12 months compared with interval insertion and that continuation is significantly higher than that seen with the shorter-acting reversible methods of contraception. Immediate postabortion LARC should be offered to all women undergoing abortion as a strategy to decrease the high rate of unintended pregnancy and repeat abortions observed in the United States.
1. Finer LB, Zolna MR. Unintended pregnancy in the United States: incidence and disparities, 2006. Contraception 2011;84:478–85.
2. Jones RK, Zolna MR, Henshaw SK, Finer LB. Abortion in the United States: incidence and access to services, 2005. Perspect Sex Reprod Health 2008;40:6–16.
3. Frost JJ, Singh S, Finer LB. Factors associated with contraceptive use and nonuse, United States, 2004. Perspect Sex Reprod Health 2007;39:90–9.
4. Mosher WD, Jones J. Use of Contraception in the United States: 1982–2008. Vital Health Stat 23 2010:1–44.
5. Trussell J. Contraceptive failure in the United States. Contraception 2011;83:397–404.
6. Trussell J, Vaughan B. Contraceptive failure, method-related discontinuation and resumption of use: results from the 1995 National Survey of Family Growth. Fam Plann Perspect 1999;31:64–72, 93.
7. Speidel JJ, Harper CC, Shields WC. The potential of long-acting reversible contraception to decrease unintended pregnancy. Contraception 2008;78:197–200.
8. Jones RK, Kost K, Singh S, Henshaw SK, Finer LB. Trends in abortion in the United States. Clin Obstet Gynecol 2009;52:119–29.
9. 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.
10. Rose SB, Lawton BA. Impact of long-acting reversible contraception on return for repeat abortion. Am J Obstet Gynecol 2012;206:37.e1–6.
11. 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.
12. 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.
13. Short M, Dallay D, Omokanye S, Hanisch JU, Inki P. Acceptability of the levonorgestrel releasing-intrauterine system and etonogestrel implant: one-year results of an observational study. Eur J Contracept Reprod Health Care 2012;17:79–88.
14. Croxatto HB, Urbancsek J, Massai R, Coelingh Bennink H, van Beek A. A multicentre efficacy and safety study of the single contraceptive implant Implanon. Implanon Study Group. Hum Reprod 1999;14:976–81.
15. Mark A, Sonalkar S, Borgatta L. One-year continuation of etonorgestrel contraceptive implant in women with postabortal versus interval placement. North American Forum on Family Planning. Washington, DC: Contraception; 2011. p. 306–7.
16. 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.
17. Kavanaugh ML, Jones RK, Finer LB. Perceived and insurance-related barriers to the provision of contraceptive services in US abortion care settings. Womens Health Issues 2011;21:S26–31.
18. Tocce K, Jheeder J, McHugh C, Teal S. Offering adolescents immediate postpartum etonorgestrel implant:continuation and repeat pregnancy rates. North American Forum on Family Planning. Washington, DC: Contraception; 2011. p. 303.
Figure. No available...Image Tools