Domestic Violence and Associated Factors Among Women of Reproductive Age in Liberia: A Cross-Sectional Study Using a Recent Liberian Demographic and Health Survey | BMC Women’s Health

Design and implementation of the study

This study was based on a large community cross-sectional survey, Liberia Demographic Health Survey (LDHS), conducted from October 16, 2019 to February 12, 2020 in Liberia. The 2019-2020 EDD used a two-stage stratified cluster design, the first stage involved clusters and the second stage involved systematic household sampling. Liberia is a West African country bounded by Sierra Leone to the northwest, Guinea to the north, Côte d’Ivoire to the east, and the Atlantic Ocean to the south and west. The total population of Liberia was 5.18 million in 2021. The Liberia DHS 2019-2020 used a standardized module of questions designed to elicit information on the extent to which women in Liberia experience violence domestic. These questions asked women about their experience of intimate partner violence by perpetrators other than husbands and other intimate partners. The questionnaire was administered to a subsample of 50% of the households selected for the men’s survey. Only one eligible woman aged 15-49 per household was randomly selected for the survey. A total of 3,120 women aged 15 to 49 completed the module with a response rate of 98.5%. Out of a total of 3120 women of childbearing age 15-49 invited for the domestic violence module, we included 2331 married women of childbearing age and after weighting the sample the final sample size becomes 2100. All previously married women of childbearing age who reported their experience of IPV were included in the study. However, women of childbearing age who were never married/single and who did not report their experience of IPV were excluded from the study.

Data analysis

Stata version 14.0 was used for statistical analysis. Descriptive studies such as number of frequencies and proportion of categorical data were used to summarize the descriptive data. Bivariate logistic regression was used to select candidate variables for multivariate logistic regression. In bivariate logistic regression, variables with a p value less than 0.2 were considered candidate variables for multivariate logistic regression analysis. Multivariate logistic regression was used to identify independent predictors of intimate partner violence among women of reproductive age in Liberia and to control for confounding factors. We performed a multivariate logistic regression by entering all the variables at once and taking the variables with p value less than 0.05 as statistically significant. Adjusted odds ratio and confidence interval (CI) were used to report statistical significance in the final model. Multicollinearity was assessed and there was no potential multicollinearity in which the VIF was 1.1.

Study variables

Dependent variable

Domestic violence previously married women’s experience of one or more emotional, physical, or sexual domestic violence in the past 12 months [11, 16, 18].. In the 2019-2029 LDHS survey, women were asked questions beginning with “Has your husband/partner ever…” Women who answered yes to a specific question were asked about the frequency of stock in the past 12 months (often/sometimes/not at this time). The three types of spousal violence were combined into a single spousal violence variable with binary results indicating whether a woman had ever experienced at least one type of spousal violence (“yes/no”) in the past 12 months. A woman was considered to have experienced IPV in the past 12 months if she had said yes to at least one form of violence 12 months prior to the survey. Series of questions were asked to each woman about physical, sexual and emotional abuse as follows;

Physical domestic violence to identify physical spousal abuse, women were asked to confirm that if their husband pushed, shook or threw something, slapped the face, twisted his arm or pulled his hair, punched or with anything that could hurt, gave kicked, dragged, beat, tried to choke, burn, threatened to attack with a knife, gun or other weapon in the past 12 months [11, 16, 18].

Spousal sexual violence to identify spousal sexual violence, women were asked to confirm that if their husband physically forced them to have sex with him even when they did not want to, physically forced them to perform any other sexual act they didn’t want, or forced them with threats or in any other way to perform sexual acts they didn’t want [11, 16, 18].

Emotional domestic violence to identify emotional domestic violence, women were asked to confirm that if their husband said or did something to humiliate them in front of others, threatened to hurt or hurt them or someone close, or insulted or hurt themselves [11, 16, 18].

Independent variables

Independent study variables were extracted from 2019/2020 LDHS data. Socio-demographic characteristics of the mother (Age of women, religion, region, level of education, profession, residence, wealth index), sex of head of household (headed by a woman, headed by a man), education of husbands , professional status of husbands, attitude towards wife beating (positive, negative), smoking (yes, no), alcohol consumption (yes, no), current pregnancy, women’s decision-making autonomy (yes, no) and media exposure was calculated as a composite variable that includes frequency of reading newspapers/magazines, frequency of listening to radio, frequency of viewing television, whether a woman was exposed to at least one form of media in the last week she will say “yes”, otherwise “no” were included in the study. The wealth index was categorized into poorest, poorest, middle, richest, and richest wealth quintiles according to the Liberia DHS standard. Religion was initially categorized as Christian, Muslim, traditional religion no traditional religion (0.47), and no religion (0.83) was less than 5% we recode religion into Christian and Muslim by grouping traditional religion and no religion under the Muslim religion giving it the name as Muslim/other. These variables were identified as factors associated with IPV after reviewing different publications [13, 15, 16, 23,24,25,26].

Comments are closed.