Skip to main content

Table 3 Multivariate secondary analysis of intervention effect on trial primary and secondary outcomes at follow-up

From: Working with men to prevent intimate partner violence in a conflict-affected setting: a pilot cluster randomized controlled trial in rural Côte d’Ivoire

Trial outcomes (secondary outcomes in italics)

Baseline

Follow-up

Unadjusted RR¥(95% CI)

Adjusted RR¥(95% CI)

 

Intervention

Control

Intervention

Control

  

Intimate Partner Violence (IPV) (Women’s reports)

    

Experience of physical and/or sexual IPV, last 12 months

26/106 (25%)

35/149 (23%)

13/113 (12%)

22/126 (17%)

0.57 (0.21 – 1.53)

0.52* (0.18 – 1.51)

Experience of physical IPV, last 12 months

21/106 (20%)

22/149 (15%)

9/113 (8%)

9/126 (7%)

1.06 (0.39 – 2.86)

0.64* (0.24 – 1.73)

Experience of sexual IPV, last 12 months

12/106 (11%)

20/149 (13%)

7/112 (6%)

18/126 (14%)

0.51 (0.18 – 1.45)

0.50* (0.14 – 1.80)

Intention and Attitudes towards IPV (Men’s reports)

    

Intention to use physical violence against an intimate partner (in at least one circumstance)

n/a

n/a

64/159 (40%)

75/155 (48%)

0.81 (0.66 – 0.99)

0.83*± (0.66 – 1.06)

Believes a woman can refuse sex in all circumstances

44/166 (27%)

43/180 (24%)

63/159 (40%)

56/155 (36%)

1.26 (0.83 – 1.91)

1.21* (0.77 – 1.91)

Hostility & Conflict Management Skills (Men’s report of skill, women’s reports of no threats)

  

Man uses at least one hostility/conflict management technique and none of his female partners report him threatening her during arguments

n/a

n/a

130/159 (82%)

100/156 (64%)

1.26 (1.08 – 1.48)

1.30*± (1.06 – 1.58)

Male Involvement in Household (Men’s reports)

    

Man involved in at least two household tasks, last 12 months

69/149 (46%)

40/151 (26%)

76/142 (54%)

25/144 (17%)

4.04 (1.53 – 10.65)

2.47* (1.24 – 4.90)

  1. ¥Risk ratios calculated at the cluster-level using data from follow-up. Both crude and adjusted ratios were adjusted for village-pair and weighted according to the number of observations per village.
  2. *Adjusted risk ratios generated on the basis of expected number of events from a logistic regression model on individual data with independent variables including man’s age, cohabitation status, self-reported ability to read, above median level of exposure to war-related traumatic events, and baseline measure of outcome indicator (or most similar baseline indicator available).
  3. ±Attitudes towards a man’s use of physical IPV against his wife, used as most similar baseline measure in calculation of adjusted risk ratio.
Back to article page

BMC Public Health

ISSN: 1471-2458

Contact us