Six binary outcome variables were measured: (a) receipt of FP counseling from a community-based health worker who visited the household in the past 12 months; (b) obtaining the current contraceptive method from a community-based health worker (defined to include Momentum nursing students who were community-based distributors); (c) informed choice (i.e., whether the provider informed the FTM about all of the following: other FP methods that she could use, possible side effects or problems that she might have with the method, and what to do if she experienced any side effects or problems, i.e., the MII); (d) current use of implants; (e) current use of injectables; and (f) current use of LARCs vs. short-term modern contraceptive methods (female and male condoms, injectables, pills, Cycle Beads, and emergency contraception). As the analysis was based on users of modern contraceptive methods, women who reported not doing anything to prevent pregnancy and those using traditional methods (withdrawal and rhythm) were not included in the analysis. Secondary outcomes included the FTM's participation in decision making about her current contraceptive method and method satisfaction.
All treatment effect models controlled for baseline measures of age, marital status, years of schooling, ethnicity, parents' education, and weekly television viewing. For dose response, we used multiple measures of intervention exposure: level of exposure categorized as full (participation in both home visits and group education), partial (one of the two) and no exposure (neither); the number of home visits (none, 1–3, 4–6, 7+), the number of group education sessions (none, 1–2, 3–4, 5+), and the total number of exposures to Momentum, defined as the number of home visits plus the number of group education sessions (none, 1–3, 4–6, 7–9, and 10+).
Like the treatment effect models, the multivariable regression model of the choice of LARCs over short-term methods was restricted to women who were currently using a modern contraceptive method at the time of the endline survey. The regression controlled for level of exposure to Momentum interventions (none (comprising users in the comparison health zones as well as 45 users in the intervention health zones who were not exposed to any Momentum interventions), partial (either home visits or group education sessions), and full (both home visits and group education sessions)); receipt of counseling on FP and/or birth spacing during the prenatal period, which was measured at baseline and consisted of the following categories: none, FP or birth spacing, and both FP and birth spacing); being never married at baseline (yes vs. no); Bakongo ethnicity (yes vs. no); worked in the past 12 months at baseline (yes vs. no); awareness of LARCs (a binary variable indicating that the respondent had ever heard of IUDs and implants); and household wealth at baseline [low (reference group), medium, and high]. We also controlled for the FTM's perceived ability to say “no” to unwanted sex (yes vs. no) and to ask her husband/partner to use a condom if she wanted him to (yes vs. no); whether the pregnancy was unintended at baseline (yes vs. no); and age group (15–19 vs. 20–24).
All treatment effect models controlled for baseline measures of age, marital status, years of schooling, ethnicity, parents' education, and weekly television viewing. For dose response, we used multiple measures of intervention exposure: level of exposure categorized as full (participation in both home visits and group education), partial (one of the two) and no exposure (neither); the number of home visits (none, 1–3, 4–6, 7+), the number of group education sessions (none, 1–2, 3–4, 5+), and the total number of exposures to Momentum, defined as the number of home visits plus the number of group education sessions (none, 1–3, 4–6, 7–9, and 10+).
Like the treatment effect models, the multivariable regression model of the choice of LARCs over short-term methods was restricted to women who were currently using a modern contraceptive method at the time of the endline survey. The regression controlled for level of exposure to Momentum interventions (none (comprising users in the comparison health zones as well as 45 users in the intervention health zones who were not exposed to any Momentum interventions), partial (either home visits or group education sessions), and full (both home visits and group education sessions)); receipt of counseling on FP and/or birth spacing during the prenatal period, which was measured at baseline and consisted of the following categories: none, FP or birth spacing, and both FP and birth spacing); being never married at baseline (yes vs. no); Bakongo ethnicity (yes vs. no); worked in the past 12 months at baseline (yes vs. no); awareness of LARCs (a binary variable indicating that the respondent had ever heard of IUDs and implants); and household wealth at baseline [low (reference group), medium, and high]. We also controlled for the FTM's perceived ability to say “no” to unwanted sex (yes vs. no) and to ask her husband/partner to use a condom if she wanted him to (yes vs. no); whether the pregnancy was unintended at baseline (yes vs. no); and age group (15–19 vs. 20–24).
Free full text: Click here
