Contraceptive discontinuation is high in Kenya--a quarter of family planning (FP) users discontinue within 12 months, despite having FP needs. The purpose of this dissertation was to provide the foundation for interpretation of the findings of a four-year, quasi-experimental intervention study being conducted to encourage the continued use of contraceptives by developing, implementing and evaluating a communication campaign intended to increase continuation rates among contraceptive injectable users in Nyando District, Kenya. The pre-intervention research in this dissertation is presented in two papers. For the first paper, fourteen focus group discussion (FGDs) were conducted in Nyando District among current contraceptive injectable users and their salient references, people who influence women's use and discontinuation of contraception. The purpose of the FGDs was to understand why women discontinue using contraceptives. For the second paper, continuation rates were measured in two districts in western Kenya before implementing a campaign to increase contraceptive continuation. Prospective data were collected over 9-months to: (a) describe factors which predict occurrence and timing of contraceptive discontinuation among first-time contraceptive injectable users, and (b) determine if rates and predictors of discontinuation differ between districts. The second study incorporated novel time-dependent attitudinal and motivational factors. In the first paper, discontinuation occurred for logistical, social and medical reasons. Common reasons for discontinuation included side effects, husbands' opposition, provider and/or clinic restrictions, misconceptions about injectables, stock outs, and lack of cash to purchase contraceptives or pay for FP services. The following factors predicted discontinuation in the second paper: side effects or health concerns, nervousness about using contraception, no previous use of modern FP, unmarried at study enrollment, preferring more privacy during FP appointment, and paying more for FP services. Associations between predictors and discontinuation differed between the districts, as did rates of discontinuation. The findings from both papers highlight the importance of side effects and health concerns on premature discontinuation of contraception suggesting that FP services more strongly address these topics. Both studies also indicate that interventions aimed at increasing continuation be tailored to address community-specific concerns. Results are used to suggest intervention points for increasing continuation of FP in western Kenya.