|
|
| |||||||||||||||||||||||||||||||||||||||
Background Although levonorgestrel implants (Norplant) would appear to be a good contraceptive option for adolescent mothers, there is little information about the use of Norplant in this population.
Methods We studied 100 postpartum adolescents who chose a contraceptive method at an urban teaching hospital between September 1991 and July 1992. Structured interviews were conducted and medical records were reviewed soon after delivery and at a mean (±SD) of 15.5 ±2.9 months post partum.
Results Forty-eight of the adolescent mothers chose Norplant, 50 chose oral contraceptives, and 2 (not further studied) chose barrier methods of contraception. The factors significantly associated with the choice of Norplant were older age (16.7 years among those who chose Norplant, vs. 16.2 years among those who chose oral contraceptives), multiparity (24 subjects vs. 6 subjects), and previous use of oral contraceptives (34 subjects vs. 21 subjects). During follow-up, there were no differences between the Norplant group and the oral-contraceptive group in the frequency of clinic visits (an average of 2.3 visits per subject in each group), failure to return after the postpartum visit (9 subjects vs. 11 subjects), or the incidence of sexually transmitted diseases (42 percent vs. 36 percent). At follow-up, 95 percent of the subjects in the Norplant group and 33 percent of those in the oral-contraceptive group were still using the method they had chosen (P<0.001). During the first postpartum year 1 subject in the Norplant group and 19 in the oral-contraceptive group became pregnant (P<0.001). Norplant users did not differ from all other adolescents studied with regard to sexual activity or condom use.
Conclusions The selection of Norplant by adolescent mothers as a method of contraception is associated with higher rates of continued use and lower rates of new pregnancy than the selection of oral contraceptives and does not affect the use of health care services, sexual activity, condom use, or the rate of sexually transmitted diseases.
Although to date oral contraceptives have been the most popular and effective contraceptive method used by adolescents,7 pregnancy rates among adolescents who use oral contraceptives range from 8 to 18 percent, as compared with 2 percent in adults8. From 45 to 66 percent of adolescents discontinue the use of oral contraceptives within 12 months9,10,11. Compounding the risk of pregnancy among adolescent oral-contraceptive users is the tendency to stop and start use, depending on sexual activity12.
Levonorgestrel implants (Norplant, Wyeth-Ayerst Laboratories, Philadelphia) would seem to provide a good contraceptive option for adolescent mothers. Norplant is safe, easy to use, effective, and does not require the cooperation of a partner or planning before intercourse13. However, the rate of acceptance of Norplant by adolescents is unknown. Irregular bleeding, fear of insertion, and visibility of the implant may deter them from using it.
It is also possible that some of the appealing characteristics of Norplant may lead to adverse health consequences for adolescents. Unlike oral-contraceptive users, Norplant users need not return for prescription refills and may miss opportunities to be screened for sexually transmitted diseases and cervical dysplasia. Furthermore, concern that the effectiveness of Norplant in preventing pregnancy may lead to decreased condom use among its users has led to considerable controversy over the suitability of the method for adolescents14,15. However, there are no data on actual condom use by Norplant users.
This study is an examination of the choice of Norplant and its use by inner-city adolescent mothers. Participants were followed prospectively from the time they chose a contraceptive post partum. The study objectives were to determine the factors associated with the choice of Norplant as compared with other contraceptive methods and to assess the effect of Norplant use on the use of health care services, sexual activity, additional pregnancy, and condom use.
Methods
Study Subjects
The study protocol was approved by the institutional review boards of the Hospital of the University of Pennsylvania, the Family Planning Council of Southeastern Pennsylvania, and New York Hospital.
The study population consisted of all adolescents 17 years of age or younger who gave birth at the Hospital of the University of Pennsylvania from September 1991 through July 1992 and who were scheduled to return post partum to its family-planning clinic. Adolescents receiving postpartum care at outside sites were not included in the study. All the participants gave written informed consent.
Study Design
A family-planning counselor visited eligible adolescents before discharge from the hospital and presented Norplant as a contraceptive option during an individualized, nondirective counseling session. A structured interview and medical-record review were performed to determine sociodemographic variables; to obtain a history of contraception, sexually transmitted diseases, and pregnancy; to assess the degree of support available from the family; and to evaluate prenatal care. The interview included measures of beliefs about the acquired immunodeficiency syndrome (AIDS) (20 true-false items derived from the AIDS Beliefs Survey of the Children's Hospital of Philadelphia16); knowledge about contraception (5 true-false and 5 two-item choice questions); attitude toward contraception (8 agree-disagree statements derived from the Perinatal HIV Reduction and Education Demonstration Project Questionnaire of the Family Planning Council of Southeastern Pennsylvania); and knowledge about sexually transmitted diseases (10 true-false-unsure items derived from the Initial Client Survey of the Family Planning Council).
The study participants were scheduled for a visit to the family-planning clinic two weeks post partum. At this visit, the chosen type of contraception was dispensed free of charge, and the participants were interviewed about sexual activity in the interval, concern about future pregnancy and contraception, and expectations about sexual activity and condom use. The participants completed the Rosenberg Self-Acceptance subscale17 and the Coopersmith Family and Self-Esteem subscales,18 which have been previously validated in studies of inner-city adolescents19.
Follow-up visits were scheduled for six weeks post partum and every six months thereafter. Participants using oral contraceptives for the first time were scheduled for an additional visit at three months. Follow-up gynecologic care was provided routinely. Screening for Chlamydia trachomatis (with Microtrak [Syva, Palo Alto, Calif.]), Neisseria gonorrhea (by culture), trichomonas (by wet smear), and syphilis (by the rapid-plasma-reagin test) was performed at the discretion of the providers at the family-planning clinic. Free condoms were dispensed, and condom use was encouraged regardless of the contraceptive method chosen.
In April 1993, the study participants were telephoned and interviewed about the continuation of contraceptive use, satisfaction with the method, any additional pregnancy, sexual activity, and condom use. The records of the family-planning clinic and the hospital were reviewed for visits during the interval, pregnancies, sexually transmitted diseases, and pelvic inflammatory disease, as diagnosed by hospital physicians using standard criteria.
Statistical Analysis
Data entry and analysis were performed with the Statistical Package for the Social Services on an IBM-compatible 486 computer. Statistical analyses included Student's t-test20 for continuous variables and the chi-square test21 for categorical variables. Ordinal values were ranked and compared with use of the Kruskal-Wallis test22; their means are reported for descriptive purposes only. All P values reported are two-tailed, and a P value below 0.05 was considered to indicate statistical significance. Logistic regression23 was used to identify the variables most predictive of a specific contraceptive choice. Variables significantly associated with the choice of contraceptive method on bivariate analysis were made eligible for logistic regression with the use of both forward and backward stepwise selection. Adjusted odds ratios and 95 percent confidence intervals were calculated from the coefficients of the variables in the model. Life-table analysis24 was used to compare the outcomes of continued use of contraception and additional pregnancy. Data were aggregated monthly from the date of delivery, and outcomes were compared by the log-rank method for censored data according to the algorithm of Lee and Desu25.
Outcomes related to follow-up visits, additional pregnancies, and sexually transmitted diseases were determined for all participants. Outcomes related to continuation of and satisfaction with contraception, menstrual patterns, sexual activity, and condom use were determined only for participants who completed the telephone interviews.
Results
During the 10-month enrollment period, 166 adolescent girls gave birth; 130 (78 percent) received in-hospital contraceptive counseling and scheduled visits to the family-planning clinic post partum. Twelve of the 130 (9 percent) declined to participate in the study. Of the 118 subjects in the initial study sample, 100 returned for contraception -- 74 (63 percent) as scheduled, and 26 (22 percent) after telephone contact.
Forty-eight subjects chose Norplant, and 50 chose oral contraceptives. The two who chose foam and condoms were not studied further. Of the 98 subjects in the final study sample, 42 in the Norplant group and 42 in the oral-contraceptive group (86 percent of the sample) completed the follow-up telephone interview.
Base-Line Findings
The study subjects were predominantly single, black, and poor and lived in single-parent households (Table 1). The Norplant group was significantly older than the oral-contraceptive group. The two groups did not differ significantly with respect to grade in school, performance in school as they described it, rate of dropping out of school before 28 weeks' gestation, or plans to attend college.
|
The subjects in the Norplant group and those in the oral-contraceptive group did not differ significantly with respect to their own or their partners' desire for another pregnancy within the year; 37 and 38 percent, respectively, planned abortions if they conceived. There were no significant differences between the groups in the degree of perceived family support during pregnancy. The mean scores related to family (Norplant group, 4.6; oral-contraceptive group, 4.7), assertiveness (3.8 vs. 3.9), and self-acceptance (8.4 vs. 8.6) did not differ significantly between the groups and were similar to the scores reported for inner-city adolescents19.
The Norplant and oral-contraceptive groups did not differ with respect to rates of participation in the comprehensive prenatal program for adolescents or in the proportion who received additional antepartum contraceptive counseling in that program. Although the groups did not differ in regard to gestational age at their first prenatal visit, the subjects in the Norplant group had poorer compliance with prenatal visits and prenatal vitamin therapy. These associations between choice of method and prenatal compliance were not significant when the analyses were controlled for parity. Despite their higher parity and presumably greater need for child care, the subjects in the Norplant group were more likely than those in the oral-contraceptive group to return post partum for contraception as scheduled.
There were no significant differences between the Norplant group and the oral-contraceptive group on any measure of sexual activity. There were also no differences in AIDS-belief scores, knowledge of sexually transmitted diseases, or rates of such disease during the index pregnancy (Table 2). The diagnoses made in the two groups included infections with chlamydia (20 subjects) and trichomonas (15), gonorrhea (5), syphilis (3), condyloma infection (5), and herpes infection (3).
|
Five variables significantly associated with the contraceptive method chosen (age, gravidity, parity, previous oral-contraceptive use, and compliance with postpartum visits) were made eligible for a logistic-regression model predicting the choice of Norplant. Of these variables, parity and previous oral-contraceptive use were entered into the model with adjusted odds ratios of 5.8 percent (95 percent confidence interval, 3.4 to 10.0) and 2.2 (95 percent confidence interval, 1.4 to 3.6), respectively.
When subjects in the Norplant group were asked why they chose Norplant, they reported difficulty remembering to take pills (71 percent), side effects of oral contraceptives (38 percent), fear of pregnancy (57 percent), ease of use of Norplant (48 percent), and encouragement from others (34 percent). When subjects in the oral-contraceptive group were asked why they did not choose Norplant, they reported a fear of insertion, needles, or both (55 percent); concern about irregular bleeding (24 percent); concern about other side effects (15 percent); and fear that the implant would be visible (9 percent).
Follow-up
Subjects who chose Norplant did not differ significantly from those who chose oral contraceptives with respect to the duration of follow-up, the number of follow-up clinic visits, or the rate of return after the postpartum visit (Table 3).
|
|
Nineteen subjects in the oral-contraceptive group (38 percent) and one subject in the Norplant group (2 percent) became pregnant during follow-up. The latter pregnancy was conceived shortly after the removal of the Norplant implant. These pregnancies resulted in six miscarriages, four elective abortions, four births, and six pregnancies ongoing at the end of the study. Five subjects in the oral-contraceptive group who became pregnant subsequently chose Norplant. In the life-table analysis (Figure 2), the risk of pregnancy within the first postpartum year was 2 percent in the Norplant group and 38 percent in the oral-contraceptive group (P<0.001).
|
Sexually transmitted diseases were documented during follow-up in 42 percent of the Norplant group and 36 percent of the oral-contraceptive group (P not significant). The diagnoses included gonorrhea (15 subjects), chlamydia infection (18), herpes infection (3), trichomonas infection (10), syphilis (3), and hepatitis B (1). Two cases each of gonorrhea and chlamydia infection were second infections. Four subjects in the oral-contraceptive group and two in the Norplant group were treated for pelvic inflammatory disease.
Because the contraceptive method chosen at base line was not necessarily the method used at follow-up, the follow-up (currently used) method was used in the analysis of recent sexual activity (Table 4) and condom use (Table 5). There was no difference in recent sexual activity between current Norplant users and users of all other methods at follow-up. However, only 36 percent of the subjects using no method of contraception reported having intercourse in the past month, as compared with 80 percent of the Norplant users and 88 percent of those using other methods (P = 0.003). Condom use reported at follow-up did not differ significantly between subjects using Norplant and all other subjects. Of the nine subjects using condoms for contraception at follow-up, only two reported condom use "all the time."
|
|
Our study explores the factors and outcomes associated with postpartum use of Norplant by inner-city adolescent mothers, a group of particular importance because of their high rates of both pregnancy and sexually transmitted disease. The demographic homogeneity of our sample facilitated control for many factors that have confounded previous research on contraceptive use by adolescents.
Our findings suggest that Norplant will be a popular contraceptive option among adolescents who have had contraceptive failure and pregnancy. However, it appears that many adolescents will not consider Norplant until oral contraceptives have been tried and have failed. Until that point, fears about insertion and possible side effects are deterrents, suggesting that counseling about contraceptives for adolescents should include a discussion of the procedure for inserting Norplant.
The rate of continued use of Norplant in our population was similar to previously reported rates in adults26. However, when the removal of a contraceptive implant requires a provider's visit, rates of continued use may not necessarily reflect the user's satisfaction. Therefore, it was important that adolescent users of Norplant also reported high rates of satisfaction with the method and willingness to recommend it to others. The patterns of bleeding reported by users in our study were similar to those reported by adult users,27 suggesting that adolescents tolerate the abnormalities of bleeding that are associated with Norplant use as well as adults.
In contrast, the adolescents who chose oral contraceptives had low rates of continued use and satisfaction with that method. An important consequence was a 38 percent pregnancy rate in the first postpartum year. The need for daily compliance with oral contraceptives, cited by 70 percent of Norplant users and one quarter of those who discontinued oral contraceptives as a reason for choosing Norplant, continues to be problematic for adolescents. Furthermore, oral-contraceptive users, who can easily discontinue their method, may be less likely to wait out the minor side effects of hormonal contraception than Norplant users, for whom discontinuation requires surgical removal of the implant.
We found that users of Norplant and oral contraceptives have similar rates of gynecologic follow-up. This suggests that interventions to improve adolescents' compliance with health care need not focus specifically on Norplant users. Moreover, the need to return for prescription refills does not appear to impart any particular advantage to oral-contraceptive use.
In our study, sexual activity did not differ between adolescents using Norplant and those using other methods. This finding is consistent with previous reports on the sexual behavior of teenagers28 and provides important and reassuring information. As would be expected, teenagers reporting no use of contraception at follow-up tended to be those who reported no recent intercourse. Research has shown that adolescents who report regular intercourse use contraception more consistently than those who report sporadic activity29 -- a finding that is supported in this study by a 30 percent pregnancy rate among the adolescents with less frequent sexual activity.
The good news is that Norplant use does not appear to decrease condom use by adolescent mothers. The bad news is that this population in general uses condoms poorly and has a high rate of sexually transmitted disease, regardless of the contraceptive method. Although Norplant cannot be singled out as a risk factor for poor condom use or for sexually transmitted diseases in adolescents, merely recommending condoms for contraception cannot be heralded as the solution to the problem, since even adolescents who reported using condoms for contraception used them inconsistently. Attention needs to be shifted from contraceptive methods as antecedents of high-risk sexual behavior and to be focused instead on interventions that will decrease such behavior and increase condom use among all teenagers, regardless of the contraceptive method they choose.
Although outcomes associated with pregnancy and continued use of the contraceptive method clearly differed between adolescent mothers who chose Norplant and those who chose oral contraceptives, the size of our sample may have limited our ability to detect smaller differences in sexual activity and condom use. In addition, attitudes specific to the program or the provider may have influenced the choice of contraceptive. Our study patients all had access to Norplant free of charge, but the effect of health insurance and prescription-drug coverage on the choice of this method is likely to be important in more diverse socioeconomic groups.
Our results provide information on the factors that influence the choice of Norplant and on the ramifications of its use by inner-city adolescent mothers. Although our findings may not be generalizable to adolescents who have never been pregnant or to adolescents from other social groups, they should allow more objective discourse by family-planning providers, community leaders, policy makers, and the public about contraceptive use by sexually active adolescents.
Supported by grants from the Research Foundation of the University of Pennsylvania, the John Merck Fund, the Craig-Dalsimer Fund, the President's Council of Cornell Women, and the J.D. Hatcher Award Committee.
We are indebted to Michael T. Mennuti, M.D., and Luigi Mastroianni, Jr., M.D., for their support; to Ellen W. Freeman, Ph.D., for her advice and suggestions; to Najma Khalid for assistance with the statistical analysis; and to Gina Fikes and Michele Cousins for their assistance with data collection.
Source Information
From the Division of Women's Health, Department of Obstetrics and Gynecology, New York Hospital-Cornell Medical Center, New York (M.P.); the Division of General Internal Medicine, Department of Medicine (G.S., C.F.), and the Department of Obstetrics and Gynecology (S.S.), University of Pennsylvania School of Medicine, Philadelphia; and Queens University, Kingston, Ont., Canada (A.R.). Presented in part at meetings of the Association for Reproductive Health Professionals, Los Angeles, November 5-6, 1993, and the Society for Adolescent Medicine, Chicago, March 18-21, 1993.
Address reprint requests to Dr. Polaneczky at Box 392, Department of Obstetrics and Gynecology, New York Hospital-Cornell Medical Center, 525 E. 68th St., New York, NY 10021.
References
| |||||||||||||||||||||||||||||||||||||||
Related Letters:
Norplant
Shihata A. A., Salzetti R. G., Lewy R. I., Polaneczky M., Sondheimer S., Slap G. B.
Extract |
Full Text
N Engl J Med 1995;
332:821-822, Mar 23, 1995.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
