Skip to main content

Table 3 Estimates of intervention effect on potential mediators among women, comparing outcome in intervention versus control communities

From: Ecological pathways to prevention: How does the SASA! community mobilisation model work to prevent physical intimate partner violence against women?

  Baselinea Follow-up  
  Intervention Control Intervention Control aRRb (95 % CI)
COMMUNITY LEVELc EA-level mean risk % (sd) n = 16 EA-level mean risk % (sd) n = 16 EA-level mean risk % (sd) n = 23 EA-level mean risk % (sd) n = 24 Mean difference (95 % CI)
Community responses to prevent violence:      
  Okay for others in community to intervene if they know IPV is occurring - - 79.2 (7.1) 58.7 (6.5) 20.3 (10.2–30.4)
  People who have witnessed/heard violence who have responded appropriately - - 47.2 (16.3) 29.8 (13.6) 13.0 (−14.6–40.6)
Norms around violence:      
  Acceptable for a man to use violence against his partner 57.0 (15.3) 59.1 (15.5) 28.1 (6.7) 51.1 (12.6) −26.7 (−49.6– −3.7)
Norms around women’s control over sex:      
  Acceptable for a woman to refuse sex with her partner 40.4 (14.7) 35.3 (15.1) 91.3 (3.2) 74.7 (10.3) 18.4 (6.0–30.9)
  Okay for a woman to ask her husband to use a condom - - 78.5 (3.2) 59.2 (7.3) 20.4 (13.5–27.4)
Broader gender norms:      
  Others in community would respect a man who made decisions jointly with his wife - - 75.2 (0.5) 57.2 (16.6) 22.8 (−2.7–48.3)
  Man’s role to decide if his wife can work - - 39.3 (2.3) 58.6 (9.9) −21.9 (−36.9– −7.0)
RELATIONSHIP LEVEL (PARTNERED IN PAST YEAR) n/N (%) n/N (%) n/N (%) n/N (%) aRR (95 % CI)
Communication:      
  Discuss things that happen in day 243/605 (80 %) 232/274 (85 %) 402/482 (83 %) 269/398 (68 %) 1.23 (1.01–1.48)
  Discuss worries 255/305 (84 %) 231/274 (84 %) 433/482 (90 %) 295/398 (74 %) 1.21 (1.02–1.44)
  Discuss what you both like during sex - - 321/481 (67 %) 183/398 (46 %) 1.49 (0.91–2.43)
  Appreciate work partner does around house - - 269 /397 (68 %) 155/303 (51 %) 1.27 (1.08–1.50)
  Appreciate work partner does outside house - - 346/410 (86 %) 244/308 (79 %) 1.08 (0.97–1.19)
Power dynamics:      
  Joint decision making 219/266 (82 %) 205/246 (83 %) 279/421 (66 %) 154/332 (46 %) 1.42 (1.14–1.76)
  Man helps around house 156/292 (53 %) 159/272 (58 %) 285/392 (73 %) 164/299 (55 %) 1.33 (0.94–1.88)
  Woman refused a job because husband doesn’t want her to work 76/300 (25 %) 52/273 (19 %) 54/454 (12 %) 65/376 (17 %) 0.78 (0.15–4.10)
  Woman participated in deciding how household finances spent 216/297 (73 %)* 169/272 (62 %)* 321/406 (79 %) 217/313 (69 %) 1.12 (1.01–1.24)
Additional sex partners:      
  Concurrent partners 18/247 (7 %) 8/215 (4 %) 25/429 (6 %) 20/341 (6 %) 1.25 (0.37–4.22)
  Male partner often suspicious that female partner is unfaithful - - 68/504 (13 %) 98/425 (23 %) 0.65 (0.24–1.73)
Relationship dissolution:      
  Separated/divorced in past year 8/299 (3 %) 3/264 (1 %) 9/486 (2 %) 17/401 (4 %) 0.44 (0.08–2.52)
INDIVIDUALS (PARTNERED IN PAST YEAR)      
Attitudes around violence:      
  Acceptable for a man to use violence against his partner 181/304 (60 %) 166/274 (61 %) 168/504 (33 %) 260/426 (61 %) 0.56 (0.38–0.82)
  Okay for a woman to tell others if she is experiencing violence - - 409/504 (81 %) 241/427 (56 %) 1.45 (1.22–1.72)
Attitudes towards women’s control over sex:      
  Acceptable for a woman to refuse sex with her partner 124/605 (41 %) 95/274 (35 %) 465/504 (92 %) 305/427 (71 %) 1.30 (1.03–1.65)
  Okay for a woman to ask her husband to use a condom - - 401/504 (80 %) 242/427 (57 %) 1.41 (1.18–1.69)
Broader gender attitudes:      
  Others in community would respect a man who made decisions jointly with his wife - - 385/504 (76 %) 227/427 (53 %) 1.49 (0.91–2.44)
  Man’s role to decide if his wife can work - - 226/504 (45 %) 288/427 (67 %) 0.67 (0.54–0.81)
Behaviours:      
  Drunk at least once a month - - 67/502 (13 %) 56/422 (13 %) 0.98 (0.56–1.70)
  Woman experiencing (man perpetrating) violence who has told someone 78/132 (59 %) 54/112 (48 %) 184/271 (68 %) 170/301 (56 %) 1.22 (0.81–1.85)
  1. aQuestion wording/item construction changed between baseline and follow-up to improve face validity - those baseline measures closest to the follow-up outcomes are presented here to assess underlying intervention/control community comparability, but baseline/follow-up comparisons are not possible
  2. bRisk ratios calculated at the cluster-level, adjusted for community-pair, and weighted according to the number of observations per village. 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 and marital status
  3. c Mean number of respondents per EA = 28.0 (range 18–35)
  4. *χ2 p-value <0.005