Women’s Health
Patterns of depot medroxyprogesterone acetate (DMPA) injectable contraceptive use and predictors of discontinuation in young Ugandan women ages 16-25 years Kristen Upson* Kristen Upson Upson Upson Upson Upson Upson Upson Upson Upson Upson Upson Department of Epidemiology and Biostatistics, College of Human Medicine, Michigan State University
The injectable contraceptive depot medroxyprogesterone acetate (DMPA) is the leading hormonal contraceptive in Uganda. Yet, data on DMPA use patterns and predictors of discontinuation in Ugandan adolescent and young women are sparse. Thus, we used data from the Kampala Women’s Bone Study, a cohort of 499 young Ugandans ages 16-25 initiating DMPA or condom use. Participants were enrolled in years 2018-2020 and followed for 2 years with quarterly study visits. Among 268 DMPA initiators, we estimated the probability of DMPA continuation using Kaplan-Meier survival curves. We conducted Cox proportional hazard regression to estimate the adjusted hazard ratios (HR) and 95% CIs for associations between predictors and time to DMPA discontinuation. Predictors included baseline age, education, and travel time to study clinic and time varying factors of COVID-19 lockdown, vaginal sex frequency, menstrual bleeding, weight change (from baseline), and DMPA type (IM or SQ). Probability of DMPA continuation at 3-, 6-, 12-, and 24-months were 63%, 39%, 15%, and 2%, respectively. However, among those who discontinued, 32% restarted DMPA use. Factors associated with increased DMPA discontinuation risk included being age 20 (vs. age 19; HR 1.5, 95%CI: 1.0-2.3), COVID-19 lockdown (HR 2.9 95%CI: 1.9-4.2) and no vaginal sex in past 3 months (vs. 13-35 times; HR 1.6, 95%CI: 0.9-3.0). In contrast, vaginal sex >45 times in past 3 months was associated with decreased DMPA discontinuation risk (HR 0.7, 95%CI: 0.5-1.1) as well as having ≤7 years of education (vs. 12-16 years; HR 0.7, 95%CI:0.5-1.1), no menstrual bleeding (vs. regular bleeding; HR 0.6, 95%CI: 0.4-0.9) and only spotting (HR 0.4, 95%CI: 0.2-0.9). No associations were observed with travel time, weight change, or DMPA type. To our knowledge, this is the first study of DMPA use patterns and predictors of discontinuation in adolescent and young Ugandan women. Our results can inform efforts to reduce unintended pregnancy.