Abstract
BACKGROUND: The overwhelming uptake of contraception in Kenya at 58% suggests huge potential for a continued increase, but discontinuation threatens efforts to achieve new targets. Further increases in contraceptive prevalence will depend more on continuation and re-adoption amongst past users because unintended pregnancies would increasingly result from discontinuation. Eliminating discontinuations from side effects and method failure could increase continuation rates by 10%. AIM: To establish the prevalence and factors associated with contraceptive discontinuation. SETTING: Kenya, with a successful family planning programme, but also the challenge of discontinuation rates of 31%. METHODS: Contraceptive calendar data from the 2014 Kenya Demographic and Health Survey were used in the survival analysis approach. RESULTS: Overall discontinuation rates were 37% (24 months) and 74% at (36 months), whilst discontinuation in need was 36%. Side effects accounted for 40% of discontinuations, whilst injection and pill recorded the highest rates. Current method emerged as a predictor of discontinuation at 24 months with the following hazard ratio (HR) at 95% confidence interval [CI]; intrauterine device (IUD) (HR = 0.466, CI = 0.254-0.857), injection (HR = 0.801, 95% CI = 0.690-0.930), implants (HR = 0.580, 95% CI = 0.429-0.784) and at 36 months, injection (HR = 0.808, 95% CI = 0.722-0.904) and implants (HR = 0.585, 95% CI = 0.468-0.730). Age (15-24 years) displayed influence only at 36 months (HR = 1.219, 95% CI = 1.044-1.424). CONCLUSION: The study showed a close link between contraceptive method used and discontinuation and thus the need to address method-related issues in an attempt to minimise discontinuation in Kenya. Expanding contraceptive options and improving the quality of service can scale up switching and thus help reduce discontinuation and unintended births.