Continuation of injectable contraception when self-injected vs. administered by a facility-based health worker: a nonrandomized, prospective cohort study in Uganda

Abstract

Objective: The purpose of this study was to compare 12-month continuation rates for subcutaneous depot medroxyprogesterone acetate (DMPA-SC) administered via self-injection and DMPA-IM administered by a health worker in Uganda.

Study design: Women seeking injectable contraception at participating health facilities were offered the choice of self-injecting DMPA-SC or receiving an injection of DMPA-IM from a health worker. Those opting for self-injection were trained one-on-one. They self-injected under supervision and took home three units, a client instruction guide and a reinjection calendar. Those opting for DMPA-IM received an injection and an appointment card for the next facility visit in 3 months. We interviewed participants at baseline (first injection) and after 3 (second injection), 6 (third injection) and 9 (fourth injection) months, or upon discontinuation. We used Kaplan-Meier methods to estimate continuation probabilities, with a log-rank test to compare differences between groups. A multivariate Cox regression identified factors correlated with discontinuation.

Results: The 12-month continuation rate for the 561 women self-injecting DMPA-SC was .81 [95% confidence interval (CI) .78-.84], and for 600 women receiving DMPA-IM from a health worker, it was .65 (95% CI .61-.69), a significant difference at the .05 level. There were no differences in pregnancy rates or side effects. The multivariate analysis revealed that, controlling for covariates, self-injecting reduced the hazard for discontinuing by 46%. A significant interaction between injection group and age suggests that self-injection may help younger women continue injectable use.

Conclusions: The significant difference in 12-month continuation between women self-injecting DMPA-SC and women receiving DMPA-IM from a health worker - which remains significant in a multivariate analysis - suggests that self-injection may improve injectable contraceptive continuation.

Implications: While injectable contraceptives are popular throughout much of sub-Saharan Africa, they have high rates of discontinuation. This study is the second from an African country to demonstrate that self-injection may improve injectable continuation rates and may do so without increasing the risk of pregnancy or adverse events.

Keywords: DMPA-SC; Discontinuation; Sayana® Press; Self-administration; Self-injection.

Fig. 1

Participant flow diagram.

Fig. 2

Kaplan–Meier cumulative probability of continuation. Log-rank test for equality of survivor function, p value=.0000.