Associations between self-reported mood-related symptoms and early contraceptive method discontinuation.

Abstract

There have been mixed findings regarding the psychological impacts of hormonal contraceptive methods. Some studies suggest negative mood symptoms with hormonal methods, while others report improved mood or no effect. To better inform clinical counseling, this study examines associations between self-reported mood-related symptoms and contraception discontinuation within the first 6 months of use, accounting for self-reported baseline mood symptoms and method type. This secondary analysis uses data from the HER Salt Lake initiative, a prospective cohort study of 4425 individuals initiating a new contraceptive method between September 2015 and March 2017. Participants were followed for 3 years, with method switching and discontinuation offered at no cost. Surveys were administered at baseline and at months 1, 3, and 6, capturing self-reported mood-related symptoms in the prior 4 weeks, contraceptive use status, and reasons for discontinuation. Chi-square tests examined the self-reported reasons for discontinuation across method types. Mixed effects logistic regression models evaluated the effect of self-reported baseline mood-related symptom burden on discontinuation, stratified by hormone category (ie, progestin-only, combined hormone, and nonhormonal) and controlling for baseline symptom frequency. Given the small sample size of those who discontinued their methods in the first 6 months, no additional covariates were included to avoid overfitting. Among participants who discontinued their method within 6 months (n=880), 28% of progestin-only users (110/389), 21% of combined hormonal users (80/386), and 10% of nonhormonal users (11/105) self-reported mood-related symptoms (ie, moodiness and/or depressive symptoms) as a contributing reason for discontinuation (P=<.001). Within progestin-only users, self-reported mood-related discontinuation was highest for the implant (40%; 34/86), followed by the injection (33%; 49/147), hormonal intrauterine device (25%; 22/88), and progestin-only pill (7.4%; 5/68). Individuals with self-reported frequent (experienced moodiness and/or depression once a week to a couple of days a week) to daily (experienced moodiness and/or depression every day) baseline symptoms were less likely to discontinue a progestin-only method (odds ratio, 0.63; 95% confidence interval, 0.44-0.92), whereas no significant association was found among users of nonhormonal (odds ratio, 1.56; 95% confidence interval, 0.44-5.52) or combined hormonal methods (odds ratio, 2.62; 95% confidence interval, 0.56-12.2). This suggests a possible moderating effect between preexisting symptom burden, method type, and (dis)continuation behaviors. A higher proportion of individuals using progestin-only contraception reported mood-related symptoms as a reason for discontinuation within the first 6 months compared to those using combined or nonhormonal methods. However, self-reported baseline mood-related symptoms were associated with lower discontinuation rates among progestin-only users, indicating a complex interaction between mood symptom burden, method choice, and (dis)continuation behaviors. These findings underscore the need for nuanced patient counseling, acknowledging mood-related concerns as legitimate and potentially method-specific, and for future research to explore these relationships using validated mood assessment tools and longer-term outcomes.