Skip to main content


Table 5 Estimates of effect on primary outcome indicators a – comparison of results from primary and secondary analyses

From: Findings from the SASA! Study: a cluster randomized controlled trial to assess the impact of a community mobilization intervention to prevent violence against women and reduce HIV risk in Kampala, Uganda

Primary outcome indicators Primary analysis – including all individuals in communities Secondary analysis – including individuals exposed to SASA!, and matched controlsb
  Adjusted RRa(95% CI) Adjusted RRa(95% CI)
Reduced social acceptance of gender inequality and IPV   
Acceptability of physical violence by a man against his partner   
• Male attitudes 0.13 (0.01 to 1.15) 0.09 (0.01 to 1.24)
• Female attitudes 0.54 (0.38 to 0.79) 0.44 (0.30 to 0.63)
Acceptability that a woman can refuse sex   
• Male attitudes 1.31 (1.00 to 1.70) 1.32 (1.02 to 1.72)
• Female attitudes 1.28 (1.07 to 1.52) 1.37 (1.14 to 1.65)
Decrease in women’s experience of IPV   
Past year physical IPV 0.48 (0.16 to 1.39) 0.57 (0.32 to 1.03)
Past year sexual IPV 0.76 (0.33 to 1.72) 0.78 (0.41 to 1..49)
Improved response to women experiencing IPV   
Appropriate community response to women experiencing IPV in past year 2.11 (0.52 to 8.59)c 3.53 (0.91 to13.62)c
Decrease in sexual risk behaviors   
Past year concurrent sexual partners among non-polygamous men partnered in past year 0.57 (0.36 to 0.91) 0.53 (0.32 to 0.87)
  1. aRisk ratios calculated at the cluster-level, both crude and adjusted ratios adjusting for community-pair, and weighted according to the number of observations per cluster. Adjusted risk ratios generated on the basis of expected number of events from a logistic regression model on individual data with independent variables including age, marital status and EA-level summary baseline measure of outcome indicator. bMatched on propensity for exposure, with propensity for exposure predicted using a logistic regression model including age, marital status, duration of relationship, duration living in community, whether stayed elsewhere in past year, work and its location, time spent out in community, whether live in gated compound, community pair, % of EA households in gated compounds, number of households in EA. cBaseline measure controlled for: disclosed past year IPV and got helpful response. CI, confidence interval; EA, Enumerated Areas; IPV, intimate partner violence; RR, risk ratio.