Objective To determine the contribution of specific contraceptive side effects to method switch and modern-method discontinuation among Kenyan women.Design A prospective cohort study.Setting Five counties in Western Kenya.Participants Women aged ≥18 years old and emancipated female minors ≥14 years old using modern, reversible contraception were recruited while attending 10 public health facilities.Methods Patient-reported adverse effect symptoms, method switch and discontinuation were reported through weekly text message-based surveys for 24 weeks.
Main outcome measurements Prevalence, hazards ratio (HR).Results Among 825 women, 44% were using implants, 43% injectables, 7% an intrauterine device and 6% oral contraceptive pills at enrolment. Most (61%) women were continuing a method used in the previous month. During the 24-week follow up, incidence of contraceptive switch was 61.3 per 100 person-years (95% confidence interval [CI] 52.4-71.8) and incidence of discontinuation was 38.5 per 100 person-years (95% CI 31.6-47.0). On average, one-quarter (prevalence [Pr] 0.24, 95% CI 0.22-0.26) of participants reported side effects or method problems weekly, with sexual side effects the most prevalent symptom (Pr 0.15, 95% CI 0.13-0.16). Lack of expected bleeding was associated with higher risk of method switch (adjusted hazard ratio [aHR] 2.36, 95% CI 1.22-4.57). Risk of all-modern method discontinuation was higher among women experiencing irregular bleeding (aHR 2.39, 95% CI 1.20-4.77), weight changes (aHR 2.72, 95% CI 1.47-4.68) and sexual side effects (aHR 2.42, 95% CI 1.40-4.20).Conclusions Addressing irregular bleeding, weight changes and sexual side effects through development of new products that minimise these specific side effects and anticipatory counseling may reduce method-related discontinuation.