Table 3.

Weighted logistic regression models using constructs derived from the Theory of Gender and Power to explain healthcare utilisation in the past 12 months among non-pregnant women in Liberia who are currently married or living with a partner

	OR (95% CI) N=3894
Sexual division of labour
Age	0.99 (0.98 to 1.00)
Education
No education	0.76 (0.62 to 0.93)**
Primary or higher	1.00
Currently working	1.18 (0.97 to 1.43)
Urban area	1.27 (0.90 to 1.79)
Region
Monrovia	1.00
North Western	0.69 (0.39 to 1.22)
South Central	1.02 (0.66 to 1.58)
South Eastern A	0.90 (0.57 to 1.43)
South Eastern B	0.81 (0.52 to 1.27)
North Central	1.00 (0.66 to 1.51)
Wealth index	1.22 (1.10 to 1.36)**
Sexual division of power
Percentage of household decisions in which respondent participates†	1.10 (1.01 to 1.20)*
Percentage of scenarios in which IPV is justifiable†	0.97 (0.90 to 1.05)
History of abuse
Physical abuse	1.25 (0.93 to 1.69)
Sexual abuse	0.65 (0.45 to 0.95)*
Structure of Social exposure and affective attachment
Older partner	0.93 (0.67 to 1.29)
Married	0.69 (0.54 to 0.88)**
Partner lives in the house	1.07 (0.83 to 1.37)
Partner’s education compared to respondent
Less education	0.93 (0.61 to 1.42)
Equal education	1.05 (0.87 to 1.27)
More education	1.00
Additional controls
Number of household members	0.95 (0.91 to 0.99)*
Number of children 5 years old or younger in household	1.23 (1.11 to 1.37)**

^*p<0.05.

^**p<0.01.

^†Estimate is per 25% increase.

IPV, intimate partner violence.