 |
|
Correlates and Consequences of Early Removal of Levonorgestrel Implants Among Teenaged Mothers
Catherine Stevens-Simon, MD;
Lisa Kelly, PA
Arch Pediatr Adolesc Med. 1998;152:893-898.
ABSTRACT
| |
Objective To determine if adolescent mothers who request early removal of levonorgestrel implants differ from those who do not in ways that might predispose them to repeated conceptions and in their concerns about adverse effects. We hypothesized that adolescent mothers who request removal of levonorgestrel implants within 2 years of insertion have more risk factors for repeated pregnancy than those who do not.
Methods We studied the prevalence of 21 characteristics associated with repeated adolescent pregnancy and 16 adverse effects of levonorgestrel implants in 181 postpartum, adolescent levonorgestrel implant recipients, 66 (36%) of whom had the levonorgestrel implants removed within 20 months of insertion (hereafter, removers).
Results Removers (n = 66) had significantly more risk factors for repeated pregnancy and reported significantly more adverse effects than did those who continued to use levonorgestrel implants (hereafter, users) (n = 115). Concerns about adverse effects rose in tandem with risk factors for repeated pregnancy (r= 0.26; P = .001) and were the most important determinant of levonorgestrel implant removal (relative risk, 9.72; 95% confidence interval, 4.62-19.49). However, the number of risk factors for repeated pregnancy was also a significant, independent predictor of levonorgestrel implant removal (relative risk, 2.34; 95% confidence interval, 1.10-4.66). Following removal, contraceptive use was poor and conception occurred rapidly; 24 (37%) of the removers conceived again within 2 years of the index delivery.
Conclusions The study hypothesis was supported. Our findings that concerns about the adverse effects of levonorgestrel implants rise in tandem with risk factors for repeated pregnancy suggest that the efficacy of counseling before and after levonorgestrel implant insertion could be improved by addressing those aspects of the user's life that undermine the motivation to use contraception.
REPEATED pregnancies occur commonly even among teenagers in comprehensive, adolescent-oriented, maternity programs.1 The frequency and rapidity with which these conceptions occur indicate that unprotected sexual practices persist even in these intensive reproductive health care settings. The adverse consequences of repeated adolescent pregnancies emphasize the need for new interventions.1-4
The use of levonorgestrel implants (Norplant, Wyeth-Ayerst, Philadelphia, Pa) could meet this need by decreasing the risk of accidental and unintended conceptions and helping teenagers bridge the gaps in contraceptive vigilance that arise when they begin to feel that becoming pregnant would help them cope with other aspects of their lives.5-12 Reported differences in the rate of repeated conception among adolescent mothers who use levonorgestrel implants (hereafter, users) following delivery and those who use other contraceptive methods are particularly impressive because they do not appear to reflect antecedent differences between teenaged users and nonusers.5-7,11 However, caution is critical since the efficacy of levonorgestrel implants for preventing repeated adolescent pregnancies at the population level depends on which young mothers request early removal.
The motivation to use any form of contraception reflects the balance individuals strike between the costs of use and nonuse.12-15 Since the use of levonorgestrel implants is almost always associated with some adverse effects, teenaged mothers who live in environments that fail to foster negative (and/or promote positive or ambivalent) feelings about adolescent childbearing may quickly lose the motivation to use levonorgestrel implants.12-15 In such settings, adolescents who use levonorgestrel implants may be more likely to report adverse effects and request early removal.
The objective of this longitudinal extension of our previously published, cross-sectional study11 of postpartum adolescent users was to determine whether adolescent mothers who request early removal of levonorgestrel implants (hereafter, removers) differ from those who do not in ways that might predispose them to repeated conceptions and in their concerns about adverse effects. We hypothesized that adolescent mothers who request removal of levonorgestrel implants within 2 years of insertion have more risk factors for repeated pregnancy than those who do not.
SUBJECTS AND METHODS
SUBJECTS
Members of the original study population were enrolled consecutively at the time of levonorgestrel implant insertion. They comprised a multiracial (114 white subjects [54%], 44 black subjects [21%], 51 Hispanic subjects [24%], and 2 other [1%]) group of 211 poor (198 subjects [94%] were Medicaid dependent), adolescent mothers who had levonorgestrel implants inserted during the first postpartum year. At delivery, their ages ranged from 13 to 18 years, with a mean ± SD age of 16.9 ± 1.2 years, approximately half lived with at least 1 biological parent, two thirds were enrolled in school, and all obtained prenatal and/or postnatal care in the Colorado Adolescent Maternity Program, Denver. As described in prior publications, the program is a comprehensive, multidisciplinary, adolescent-oriented prenatal, delivery, postpartum, and infant care program.11, 16-19 Details of our educational program for prospective levonorgestrel implant users have also been published previously.11
Analyses for this article are based on the 181 young mothers whose contraceptive and reproductive behavior was tracked for at least 20 months following levonorgestrel implant insertion. Characteristics of this portion of the study population are presented below.
The remaining 30 (14%) of the original 211 levonorgestrel implant users were unavailable for follow-up (most because they moved out of the region leaving no forwarding address or contact person). Attrition analysis revealed that the demographic features of the study sample (n = 181) were representative of the larger population (N = 211) from which it was drawn. However, the mean ± SD age of those who were unavailable for follow-up (n = 30) was significantly older (17.3 ± 1.0 compared with 16.9 ± 1.3 years; P = .05), and they were less likely to be black (6.7% compared with 23.8%; P = .003). Since young, black teenagers are at highest risk for rapid repeated conception, our study could overstate the repeated pregnancy rate.1-2
DATA COLLECTION
The data used to quantify the risk of repeated adolescent pregnancy were obtained prospectively from patient interviews. Following delivery, a female research assistant who was not involved in the prenatal contraceptive counseling and who was unaware of the study hypothesis conducted a structured interview with all the study subjects on the postpartum ward of the university hospital in Denver.
A precoded, multiple-choice questionnaire was used to collect information about the participants' sexual and reproductive histories, the social context of their pregnancies, and the 21 widely accepted demographic and psychosocial risk factors for repeated adolescent pregnancy, which are listed below.
*Asterisk indicates that these were individual risk factors that differed significantly between removers and users.
The individual items were identified a priori from a review of the literature.1-2,4, 11-15,17-18 They are defined in prior publications in which we note that none of these characteristics of adolescent mothers and the environments they live in includes an inherent mechanism that necessarily leads to repeated pregnancy.11, 18 Indeed, we have found that the presence of 5 or more of the risk factors is a better predictor of rapid repeated conception during adolescence than the presence of any single high-risk characteristic.18 Thus, we surmise that these characteristics are associated with an increased risk of repeated conception because they create an atmosphere that favors adolescent childbearing.18 This hypothesis is consistent with other data concerning the antecedents of risk behavior during adolescence.15, 20-21
To gain insight into the factors motivating early removal of levonorgestrel implants, the research assistant used the same 21-item questionnaire to reinterview the study participants at 4- to 6-month intervals for 2 years following levonorgestrel implant insertion. During the postinsertion interviews, information was also obtained about the adverse effects of levonorgestrel implants and the desirability of having another child. Participants rated the distress caused by 16 common adverse effects of levonorgestrel implants on a scale of 0 to 3 to obtain a distress score. The teenagers were also asked when they wanted to have another child, when they thought their boyfriend wanted to have another child, how they would feel, and how they thought their boyfriends would feel if they were to become pregnant again immediately. Responses to the last 2 questions were quantified on a 4-point pictorial Likert scale, with choices ranging from 0 (a face showing a sad or mad expression) to 3 (a face showing a happy expression). Both scales are available from us on request.
Study participation ceased with the diagnosis of pregnancy.
The study was approved by the Institutional Review Board at the University of Colorado Health Sciences Center, Denver.
DATA ANALYSIS
We used univariate analyses to describe the study population and to assess the prevalence of risk factors for repeated pregnancy and adverse effects related to levonorgestrel implant use (Table 1) in the study population. Relationships between continuous variables were examined with Pearson correlations. Initial comparisons between adolescent mothers who requested removal of levonorgestrel implants and those who did not were carried out with bivariate analyses (Student t and  2 tests).
|
|
|
|
Table 1. Reasons for Removal of Levonorgestrel Implants
|
|
|
Multivariate analyses using logistic regression were conducted to determine whether findings at the bivariate level would be supported after adjusting for other factors that might affect their association with early removal of levonorgestrel implants. Adjusted odds ratios and their 95% confidence intervals were calculated from the logistic coefficients and SEs for each variable in the models. To simplify the model and its application in clinical practice, the independent predictor variables were dichotomized (present or absent). For example, based on the results of our prior studies,18 individuals who had more than 5 repeated pregnancy risk factors were coded as 1 for repeated pregnancy risk. The statistical test for the logistic regression model was the 2 likelihood ratio. All analyses were performed with Statistical Package for the Social Sciences software.22
RESULTS
INSERTION AND REMOVAL OF LEVONORGESTREL IMPLANTS
Study participants had the levonorgestrel implants inserted a mean ± SD of 11.1 ± 9.9 weeks after delivery. The incidence of levonorgestrel implant removal increased from 7 subjects (4%) at 6 months following insertion to 20 subjects (11%) at 6 to 12 months following insertion to 39 subjects (22%) at 12 to 20 months following insertion. Overall, 66 (36%) of 181 study participants were removers who had their levonorgestrel implants removed within 20 months of insertion (range, 2-20 months; mean ± SD, 13.2 ± 4.8 months). The remaining 115 young mothers remained users. Neither the duration of levonorgestrel implant use nor the prevalence of levonorgestrel implant removal was related to the site of prenatal or postpartum care or to any of the demographic or behavioral characteristics.
RISK OF REPEATED CONCEPTION
At the time of levonorgestrel implant insertion, the 66 removers had significantly more risk factors for repeated pregnancy than the 115 users (mean ± SD, 6 ± 2 compared with 5 ± 2; P = .007). They were also significantly more likely to have more than 5 risk factors for repeated pregnancy (40 [61%] compared with 40 [35%]; P = .001). Regarding specific risk factors for repeated adolescent pregnancy, the removers were more likely than users to be black (25 [38%] compared with 18 [16%]; P = .002); depressed (14 [21%] compared with 12 [10%]; P = .05); and think that their boyfriends wanted another child within the next 2 years (6 [9%] compared with 1 [1%]; P = .03).
Removers were no more enthusiastic about becoming pregnant again immediately than users at delivery or at any other time during the study period. Nevertheless, at removal, the 39 young mothers who had their levonorgestrel implants removed during the second year of use were significantly more likely to admit that they wanted to have another child within the next 2 years than their 115 user peers (50% compared with 29%; P = .05). At the time of levonorgestrel implant removal, these 39 young mothers were also significantly more likely to feel that their boyfriends would be pleased by a pregnancy than were the 115 remaining users (55% compared with 27%; P = .01). By contrast the 27 young mothers who had their levonorgestrel implants removed during the first year of use were significantly less likely to ascribe such sentiments to their boyfriends than were the 154 users (10% compared with 33%; P = .009). Users and removers did not differ significantly with regard to school enrollment, marital status, or living arrangements during the follow-up period.
ADVERSE EFFECTS
Most (165 [91%] of 181) of the teenagers reported some adverse effects related to levonorgestrel implant use.The most common adverse effect was moodiness (70%) followed by irregular vaginal bleeding (65%), headaches (61%), depression (46%), and weight gain (44%). The study participants reported a mean ± SD of 5 ± 4 adverse effects (range, 0-13), with a mean ± SD score of 9.5 ± 8.5 on the adverse effects distress scale (range, 0-33). There were no significant group differences in the prevalence of these symptoms at the time of levonorgestrel implant insertion. However, the 66 removers reported significantly more adverse effects than the 115 remaining users (mean ± SD, 7 ± 3 compared with 4 ± 3; P<.001). They were also significantly more distressed by the adverse effects they experienced (mean ± SD score, 15.6 ± 8.4 compared with 5.7 ± 6.0; P<.001).
Table 1 shows the primary and contributory reasons for levonorgestrel implant removal in relation to the duration of levonorgestrel implant use. Irregular vaginal bleeding was the most commonly cited reason for removal, and the desire for repeated conception was the second most commonly cited reason for removal, except among those who used the levonorgestrel implant for less than 6 months.
RELATIONSHIP BETWEEN RISK FACTORS, ADVERSE EFFECTS CONCERNS, AND REMOVAL
Both the frequency and the severity of the distress about the adverse effects of levonorgestrel implant use and the removal rate rose in tandem with the number of risk factors for repeated pregnancy (r = 0.2, P = .007 and r = 0.26, P = .001, respectively). The 80 adolescents who had more than 5 risk factors for repeated pregnancy were significantly more distressed by the adverse effects of levonorgestrel implants than the other 101 adolescents (mean ± SD, 12.0 ± 8.6 compared with 7.6 ± 7.9; P = .001). They were also significantly more likely to request levonorgestrel implant removal (40 [50%] compared with 26 [26%]; P = .001).
The results of the multivariate analyses presented below support the study hypothesis. None of the individual variables with a hypothesized relationship to repeated adolescent pregnancy entered the logistic regression model. Although distress about the adverse effects related to levonorgestrel implants was the most important determinant of removal, the number of risk factors for repeated pregnancy remained a significant, independent predictor of levonorgestrel implant removal (especially during the second year of levonorgestrel implant use).
CONSEQUENCES OF LEVONORGESTREL IMPLANT REMOVAL
The data presented in Table 2 show that contraceptive use was poor and conceptions occurred quickly and commonly among teenagers who had the levonorgestrel implants removed after the first 6 months of use. This was true although on average these young mothers reported that they and their boyfriends did not want another child for 3 to 4 years. Neither the prevalence of postremoval conception nor the duration of the removal-to-conception interval was related to postremoval contraceptive plans. Overall, 22 (58%) of the 38 teenagers who planned to use a specific contraceptive agent following levonorgestrel implant removal conceived again (on average within 10 months of levonorgestrel implant removal), whereas 18 (64%) of those who did not have specific postremoval contraceptive plans conceived again (on average within 6 months of removal); however, the differences are not statistically significant. Ultimately, 24 (37%) of the 66 teenagers who had their levonorgestrel implants removed conceived during the first 2 postpartum years. There were no pregnancies among users during the study period.
|
|
|
|
Table 2. Contraception and Conception After Removal of Levonorgestrel Implants*
|
|
|
COMMENT
Repeated adolescent pregnancy remains a significant public health problem in the United States.1-4 In the absence of intensive postpartum intervention, nearly half of first-time adolescent mothers become pregnant again within 2 years.1 There have been numerous attempts to prevent (or at least delay) these medically and socially risky conceptions,1-2,17-18,23 but postpartum levonorgestrel implant insertion is the only intervention that has been consistently associated with a significant reduction in the repeated adolescent pregnancy rate in this country.5-7,18, 23-24 While the importance of this finding should not be diminished, in the absence of prospective randomized controlled trials, enthusiasm for this intervention strategy must be tempered by concerns about the possibility of selection bias.11 Although sociodemographic homogeneity probably minimizes this threat to validity, differences in the long-term consequences of childbearing within seemingly homogeneous populations of adolescent mothers raise the concern that inherent, antecedent differences between users, nonusers, and removers account for the dramatic group differences in the repeated pregnancy rates that have been reported.11 If levonorgestrel implant use is not randomly distributed in the population of adolescent mothers and those who are at highest risk for repeated conception are also least likely to use levonorgestrel implants and/or most likely to request early levonorgestrel implant removal, the efficacy of this approach to the prevention of repeated adolescent pregnancies could be compromised at the population level.
Within the context of this concern, our previous finding that adolescent mothers who had a levonorgestrel implant inserted following delivery of their child (users) were not at lower demographic and psychosocial risk for repeated pregnancy compared with adolescent mothers who chose to use other forms of contraception (nonusers) was reassuring.11 However, the results of this longitudinal extension of that study renew the concern about selection bias; the removers we studied had significantly more risk factors for repeated adolescent pregnancy than the users did. Although concerns about adverse effects are the most important determinant of levonorgestrel implant removal in this and other populations,5-9 multivariate analyses indicated that the number of risk factors for repeated adolescent pregnancy is also a significant, independent predictor of levonorgestrel implant removal.
These findings support the study hypothesis; adolescent mothers who request early levonorgestrel implant removal differ from those who do not in ways that could predispose them to repeated conception. Taken together, the results of this study suggest that the demographic and behavioral risk factors for repeated adolescent pregnancy are associated with an increased risk of early levonorgestrel implant removal because collectively the risk factors foster positive and/or ambivalent feelings about the desirability of childbearing during adolescence, which in turn make users more vulnerable to concerns about adverse effects that prompt removal. Further study is needed to determine how these relationships affect the efficacy of levonorgestrel implant use for preventing repeated adolescent pregnancies at the population level.
The frequency and rapidity with which we and others6-7,25 have found that conceptions occur following levonorgestrel implant removal emphasize the need for more effective counseling, particularly when concerns about adverse effects prompt removal after the first 6 months of use. Within the context of this need, our finding that the repeated pregnancy rate was lowest when the levonorgestrel implant was removed during the first 6 months of use is important. (It was the only time during the study period when the removers neither wanted to become pregnant again nor thought that their boyfriends wanted them to become pregnant again.) It suggests that the efficacy of counseling before and after levonorgestrel implant insertion may be improved by addressing those aspects of the user's life that undermine the motivation to use contraception, rather than focusing primarily on specific adverse effects of levonorgestrel implant use. The results of our study suggest that more attention needs to be given to the motivational component of the adolescent pregnancy problem. While most of us who provide contraceptive counseling for teenagers are quick to point out the costs associated with early childbearing, we rarely take the time to promote alternative behaviors that can meet their needs.14 As a result of our preoccupation with the societal costs of adolescent pregnancy, we tend to disregard the potential benefits of childbearing for individual adolescents. Our failure to acknowledge that pregnancy can serve important social and personal functions for adolescents may be undermining our ability to combat the feelings that make risky sexual behavior (eg, unprotected sexual activity after levonorgestrel implant removal) so difficult to change.
AUTHOR INFORMATION
| Editor's Note: Unlike so many behaviors of adolescents, the one hypothesized in this study makes sense—and it proved to be true.—Catherine D. DeAngelis, MD
|
|
Accepted for publication May 4, 1998.
Supported in part by grant APH000166-5, Office of Adolescent Pregnancy, and grant 5 MO1 RR00069, General Clinical Research Centers Program, National Center for Research Resources, National Institutes of Health, Bethesda, Md, and a grant from the Colorado Trust, Denver.
Presented in part at The Society for Pediatric Research Annual Meeting, Washington, DC, May 2, 1997.
We thank the staff and patients of the Colorado Adolescent Maternity Program for their help with data collection.
Reprints: Catherine Stevens-Simon, MD, Department of Pediatrics, Division of Adolescent Medicine, University of Colorado Health Sciences Center, The Children's Hospital, 1056 E 19th St, Denver, CO 80218.
From the Department of Pediatrics, University of Colorado Health Sciences Center, Denver.
REFERENCES
|
1. Stevens-Simon C, White M. Adolescent pregnancy. Pediatr Ann. 1991;20:322-331.
ISI
| PUBMED
2. Brown SS, Eisenberg L. The Best Intentions. Washington, DC: National Academy Press; 1995.
3. Stevens-Simon C, Roghmann KJ, McAnarney ER. Repeat adolescent pregnancy and low birth weight: methods issues. J Adolesc Health Care. 1990;11:114-118.
FULL TEXT
| PUBMED
4. Furstenberg FF, Brooks-Gunn J, Morgan SP. Adolescent mothers and their children in later life. Fam Plann Perspect. 1987;19:142-151.
FULL TEXT
|
ISI
| PUBMED
5. Polaneczky M, Slap G, Forke C, Rappaport A, Sondheimer S. The use of levonorgestrel implants (Norplant) for contraception in adolescent mothers. N Engl J Med. 1994;331:1201-1230.
FREE FULL TEXT
6. Polaneczky M, Forke C, Armstrong J. Norplant vs oral contraceptive use by adolescent mothers: long term outcome [abstract]. J Adolesc Health. 1996;18:120.
7. Blumenthal PD, Wilson LE, Remsburg RE, Cullins VE, Huggins GR. Contraceptive outcomes among post-partum and post-abortal adolescents. Contraception. 1994;50:451-460.
FULL TEXT
|
ISI
| PUBMED
8. Dinerman L, Wilson M, Duggan A, Joffe A. Outcomes of adolescents using levonorgestrel implants vs oral contraceptives or other contraceptive methods. Arch Pediatr Adolesc Med. 1995;149:967-972.
FREE FULL TEXT
9. Cromer B, Smith D, Blair J, Dwyer J, Brown RT. A prospective study of adolescents who choose among levonorgestrel implant (Norplant), medroxyprogesterone acetate (Depo-Provera), or the combined oral contraceptive pill as contraception. Pediatrics. 1994;94:687-694.
FREE FULL TEXT
10. Ricketts SA. Repeat fertility and contraceptive implant use among Medicaid recipients in Colorado. Fam Plann Perspect. 1996;28:278-280.
FULL TEXT
|
ISI
| PUBMED
11. Stevens-Simon C, Wallis J, Allan-Davis J. Which teen mothers choose Norplant? J Adolesc Health. 1995;16:350-353.
FULL TEXT
|
ISI
| PUBMED
12. Miller WB. Why some women fail to use their contraceptive method: a psychological investigation. Fam Plann Perspect. 1986;18:27-31.
FULL TEXT
|
ISI
| PUBMED
13. Zabin LS, Astone NM, Emerson MR. Do adolescents want babies? the relationship between attitudes and behavior. J Res Adolesc. 1993;3:67-86.
FULL TEXT
| PUBMED
14. Stevens-Simon C. Contraceptive care of adolescents. Contemp Pediatr. 1997;14:35-58.
15. Stevens-Simon C, Lowy R. Teenage childbearing: an adaptive strategy for the socioeconomically disadvantaged or a strategy for adapting to socioeconomic disadvantage? Arch Pediatr Adolesc Med. 1995;149:912-915.
FREE FULL TEXT
16. Stevens-Simon C, Wallis J, Allan-Davis J. Antecedents of preterm delivery among adolescents: relationship to type of prenatal care. J Matern Fetal Med. 1995;4:186-193.
FULL TEXT
17. Stevens-Simon C, Kelly LS, Singer D. Reasons pregnant adolescents give for not using contraceptives prior to conception: implications for recidivism. Arch Pediatr Adolesc Med. 1996;150:1037-1043.
FREE FULL TEXT
18. Stevens-Simon C, Dolgan JI, Kelly LS, Singer D. The effect of monetary incentives and peer support groups on repeat adolescent pregnancies: a randomized trial of the Dollar-A-Day Program. JAMA. 1997;277:977-982.
FREE FULL TEXT
19. Stevens-Simon C, Kelly LS, Singer D. Late prenatal care: a risk factor for incomplete immunization among infants of teen-mothers? Arch Pediatr Adolesc Med. 1996;150:829-833.
FREE FULL TEXT
20. Jessor R. Risk behavior in adolescence: a psychosocial framework for understanding and action. J Adolesc Health. 1991;12:597-605.
FULL TEXT
|
ISI
| PUBMED
21. Resnick MD, Bearman PS, Blum RW, et al. Protecting adolescents from harm. JAMA. 1997;278:823-832.
FREE FULL TEXT
22. Norusis MJ. Statistical Package for the Social Sciences. Chicago, Ill: SPSS/PC+;1990.
23. Maynard R, Rangarajan A. Contraceptive use and repeat pregnancies among welfare-dependent teenage mothers. Fam Plann Perspect. 1994;26:198-205.
FULL TEXT
|
ISI
24. O'Dell CM, Forke CM, Polaneczky MM, Sondheimer SJ, Slap GB. Depot medroxyprogesterone acetate or oral contraception in postpartum adolescents. Obstet Gynecol. 1998;91:609-614.
FULL TEXT
|
ISI
| PUBMED
25. Harel Z, Biro FM, Kollar LM, Rauh JL. Adolescents' reasons for and experience after discontinuation of long-acting contraceptives Depo-Provera and Norplant. J Adolesc Health. 1996;19:118-123.
FULL TEXT
|
ISI
| PUBMED
 CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati  Twitter
What's this?
|
