Rates of contraceptive discontinuation are high in many low and middle countries contributing to unmet need for contraception and other adverse reproductive health outcomes. Few studies have investigated how women's beliefs about methods and strength of fertility preferences affect discontinuation rates. This study examines this question using primary data collected in Nairobi and Homa Bay counties in Kenya.
We used data from two rounds of a longitudinal study of married women ages 15–39 years (2,812 and 2,424 women from Nairobi and Homa Bay respectively at round 1). Information on fertility preferences, past and current contraceptive behavior, and method-related beliefs about six modern contraceptive methods were collected, along with a monthly calendar of contraceptive use between the two interviews. The analysis focused on discontinuation of the two most commonly used methods in both sites, injectables and implants. We carry out competing risk survival analysis to identify which method related beliefs predict discontinuation among women using at the first round.
The percentages of episodes discontinued in the 12 months between the two rounds was 36%, with a higher rate of discontinuation in Homa Bay (43%) than in the Nairobi slums (32%) and higher for injectables than implants. Method related concerns and side effects were the major self-reported reasons for discontinuation in both sites. The competing risk survival analysis showed that the probability of method related discontinuation of implants and injectables was significantly lower among respondents who believed that the methods do not cause serious health problems (SHR = 0.78, 95% CI: 0.62–0.98), do not interfere with regular menses (SHR = 0.76, 95% CI: 0.61–0.95) and do not cause unpleasant side effects (SHR = 0.72, 95% CI 0.56–0.89). By contrast, there were no net effects of three method related beliefs that are commonly cited as obstacles to contraceptive use in African societies: safety for long-term use, ability to have children after stopping the method, and the approval of the husband.
This study is unique in its examination of the effect of method-specific beliefs on subsequent discontinuation for a method-related reason, using a longitudinal design. The single most important result is that concerns about serious health problems, which are largely unjustified and only moderately associated with beliefs about side effects, are a significant influence on discontinuation. The negative results for other beliefs show that the determinants of discontinuation differ from the determinants of method adoption and method choice.