Avaliação da segurança, eficácia e custo-efetividade dos métodos contraceptivos de longa duração em comparação aos métodos convencionais em adolescentes

• Ano de defesa: 2020
• Autor(a) principal: Farah, Daniela De Moura Azevedo Tuma [UNIFESP]
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Paulo (UNIFESP)
• Palavras-chave em Português: Adolescent; Adolescence Pregnancy; Contraception; Meta-Analysis; Health Economics; Adolescente; Gravidez Na Adolescência; Anticoncepção; Metanálise; Economia Da Saúde
• Link de acesso: https://sucupira.capes.gov.br/sucupira/public/consultas/coleta/trabalhoConclusao/viewTrabalhoConclusao.jsf?popup=true&id_trabalho=10005551; https://hdl.handle.net/11600/64811

Adolescent pregnancy is a worldwide concern, affecting not only the teenager's life but also the community. The unmet need for acceptable and effective contraceptive methods in this population contributes to high adolescent pregnancy rates. Our objective was to assess the safety and efficacy of long-term reversible contraceptives (LARC) for adolescent girls and estimate these methods' cost-effectiveness compared to non-LARC methods. Within this scope, we consider the willingness to pay for contraceptive methods of adolescent girls and their parents. Thus, we conducted a meta-analysis to assess the efficacy, safety, adherence, and preference of choice of contraception LARC and SARC methods in young women. We performed an incidence meta-analysis to estimate the average rates of continuation and pregnancy of four contraceptive methods. We carry out systematic searches in the PUBMED, EMBASE, LILACS, and Cochrane databases. There was no restriction on the language. We included observational studies or RCTs. Two authors independently selected abstracts and reviewed full-text articles, and extracted data. We used RevMan 5.3 software in the first meta- analysis to combine the results of the studies. The data were derived in risk ratios (RR) and mean differences with 95% CI using a meta-analytic model of random effects. For the meta- analysis of incidence, we used the Metafor and Meta packages in RStudio software (version 1.1.4). We collected the pregnancy and continuation rates for each contraceptive method, used the inverse of variance in all calculations, and applied the LOGIT transformation. We used the random-effects meta-analytical model. The Cochrane collaboration tool and New Castle- Ottawa were used to assess all included ECR and observational studies' quality and bias, respectively. We assessed the quality of the evidence using the GRADE criteria. Concerning cost-effectiveness, we developed a Markov model to mirror the clinical and economic impact of adopting LARC methods compared to combined oral contraceptives in sexually active Brazilian adolescent girls from the perspective of the public health system and with a time horizon of five years. Costs are expressed in 2018 US dollars (US $), and we applied a 5% discount rate to clinical outcomes and costs. The model's inputs included the costs of contraceptive methods, the rate of discontinuation and compliance of the contraceptives, abortion rates, and the costs of childbirth and neonatal care associated with pregnancies. The model's outcomes were the number of pregnancies, abortions, and the number of preterm and term births avoided. We conducted univariate and probabilistic sensitivity analyzes. Finally, we conducted a study of willingness to pay for contraceptive methods. The willingness to pay was assessed by a cross-sectional study using a questionnaire applied to Brazilian adolescents aged 13 to 19 and their parents. Values are expressed as mean ± SD in reais (R $). Spearman's correlation was performed for the socioeconomic status of the parents and the age of adolescents. Chi-square analyzes were carried out between the types of methods and the adolescents 'and parents' perspectives. We performed the Bland-Altman chart to determine an agreement between the pairs and their willingness to pay. The first meta-analysis included 25 studies, of which eight compared the non-LARC and LARC methods, and 17 studies performed comparisons between the LARC methods. There was greater adherence favoring LARC methods in 12 months than non-LARC methods [RR 1.60 (95% CI 1.21–2.12) I2 = 88%; 1,606 young women]. However, young women still opt for non-LARC methods [RR 0.37 (95% CI 0.17–0.80) I2 = 99%; 2,835 young women]. Pregnancies in LARC methods were rare, with xii very few events. In the incidence meta-analysis, we showed that the oral contraceptive pill's continuation rate was estimated at 51%. At the same time, for the copper IUD, LNG-IUS, and implant, it was 77%, 84%, and 86%, respectively. The oral contraceptive pill's combined pregnancy rate was 11%, while for the copper IUD, LNG-IUS, and implant, it was 5%, 1.7%, and 1.8%, respectively. Our cost-effectiveness model showed that the copper IUD was dominant 100% of the time for all outcomes and a five-year budgetary impact of - $ 422,431,269.10. The LNG-IUS and the subdermal implant were cost-effective for all outcomes. However, in the budgetary impact analysis of the subdermal implant, it represented a savings of - US $ 851,686,904.96 in five years. A total of 165 pairs of questionnaires were answered. The non-LARC method was significantly more acceptable to pay out of pocket than the LARC method for adolescents and parents. Parents and their daughters are willing to pay R $ 52.25 ± 22.48 and R $ 51.63 ± 21.24 out of their pocket by the non-LARC method, while by the LARC method, when they are subsidized, they are willing to pay R $ 176.83 ± 130.34 and R $ 174.83 ± 143.64, respectively. The Bland-Altman analysis indicated agreement between peers and the price they are willing to pay for each contraceptive method. The LARC methods are more effective in preventing pregnancy in adolescent girls, being more cost-effective than the birth control pill. However, non-LARC methods are still more willing to pay for parents and their daughters.