Adolescent Medicine Fellow The Children's Hospital at Montefiore Bronx, New York, United States
Background: Eating disorders have become increasingly prevalent in adolescents. Like their peers without eating disorders, many of these adolescents are sexually active and require contraception. There is a paucity of data on long-acting reversible contraception (LARC) use by those with an eating disorder diagnosis (EDO). Objective: We hypothesized that weight concerns may lead to lower acceptance and higher device removal rates in adolescents with an EDO compared to peers. Design/Methods: We conducted a secondary analysis of data from a prospective cohort study from 2017-2021 of LARC insertions from 4 academic U.S. Adolescent Medicine practices. Each of the 4 practices obtained IRB approval at their institutions. We compared groups with and without an EDO, by participant characteristics, selected device type, one-year continuation rate, and reasons for insertion & removal. Results: Overall, 2361 LARCs were inserted; 80 (3.4%) in adolescents with a pre-existing EDO (mean time from diagnosis 1.8 ± 1.8 years). Of these, 27 (34%) had anorexia nervosa, 17 (21%) unspecified feeding or eating disorder, 13 (16%) binge eating disorder, 12 (15%) other specified feeding or eating disorder, 6 (8%) bulimia nervosa, & 5 (6%) avoidant restrictive food intake disorder. Mean age at insertion was 18.6 ± 2.2 years for those with an EDO & 18.8 ± 2.5 for those without. Of those with an EDO, 36 (45%) were white, 12 (15%) Hispanic, 12 (15%) Black vs 711 (31.2%), 645 (28.3%), 352 (15.4%), respectively for those without an EDO (p < 0.01). Of those with an EDO, 85% had LARC inserted for contraception as did 84% of those without. Of those with an EDO, 56 (70%) had an IUD inserted & 24 (30%) had an implant vs 1144 (50%) & 1137 (50%), respectively of those without an EDO, (p < 0.001). Most IUDs inserted in both groups were levonorgestrel releasing, 52 (93%) in those with an EDO and 1087 (95%) in those without an EDO. One-year continuation rate for all LARCs was 77% for both groups. No devices in the EDO group were removed due to concern for weight changes.
Conclusion(s): Overall, the IUD was preferred over implants by adolescents with an EDO when compared to those without an EDO. This may be an indicator of their interest in a more localized hormonal delivery method, IUDs are often cited by clinicians as associated with less weight gain than implants. One-year continuation of LARC methods was similar for adolescents with and without an EDO suggesting these are satisfactory methods of contraception for adolescents with an EDO.
