Study area
The study was conducted in St. Paul millennium hospital medical college and affiliate hospitals in Addis Ababa, Ethiopia. The hospital has one of the largest maternity center in the country. It provides ANC service with an average of 1500 mothers per month. SPHMMC has affiliates with three hospitals in Addis Ababa for obstetrics and gynecology services. These hospitals are RDDMH (Ras Desta Damtew Memorial Hospital), Abebech Gobena mothers and children Hospital (AGMCH) and Minilik II referral hospital. Their monthly client flow at ANC averages to 500,750,550 clients respectively.
Study design
Multicenter prospective analytic Cross sectional study was conducted. All pregnant women above 28 weeks of gestation (Estimated fetal weight (EFW) above 1KG if unknown date) attending ANC in the study hospitals, included in the study after fulfilling the inclusion criteria, and randomly selected.
Inclusion and Exclusion criteria
Inclusion criteria
- All pregnant women above 28 weeks of gestation (EFW > 1kg if unknown date) attending ANC in the selected hospitals and live with their partner in the past 06 months.
Exclusion criteria
- Pregnant women who have not lived with their partner in the past 06 months.
- Pregnant women whose gestational age is below 28 weeks/ estimated fetal weight < 1kg for unknown date.
- Acute onset psychiatric conditions and critically ill patients who are unable to communicate.
Sample size determination and sampling procedure
Sample size was calculated using single population formula, where z is the normal standard deviation set at 1.96, with a confidence level specified at 95% and a tolerable margin of error (d) at 5%, considering 10% nonresponse rate and prevalence of violence (p) 46.1% from a previous similar study in Debre Markos, Ethiopia. The total sample size after considering 10% non-response rate is 420. Based on pre-determined sample size, probability proportional to size (PPS) sampling employed to allocate number of participants in each of the respective hospitals.
Data collection tools and procedures
The standard WHO multicountry study questionnaire for assessing women’s health and violence was used for data collection as modified to fit local setting in previous studies. The questionnaire had four items for psychological violence, five items for physical violence, and three items for sexual violence, and the response to each item was either “Yes” or “No.” Participants who respond “Yes” to one or more items of violence during recent pregnancy were incident cases of intimate partner violence victimization [2]. The adopted questionnaire was assessed among 5% of the total sample size before the actual data collection. Data collected by ten midwives (all females, which enable better rapport with better disclosure from study participants). The principal investigator supervised the data collection procedures. Training given for two days regarding interviewing techniques, the purpose of the study, the importance of privacy, discipline and approach to the interviewees, and confidentiality of the respondents. The investigator assessed collected data weekly for consistency, completeness and double entry of data.
Data processing and analysis
Data entry achieved using Epi-info version 7.1. The raw data then exported into SPSS version 25 for analysis. Descriptive statistical analyses computed to describe the characteristics of participants/partners. Then, information presented using tables and figures. Bivariate analysis carried out to see the association of each independent variable with the outcome variable (IPV). Variables with p value <0.2 in the bivariate analysis transferred in to multivariate analysis model. Odds ratios with 95% CI reported to show the relation pattern among variables. Variables with p values less than 0.05 were taken to be significantly associated.
Study variables
Dependent variable:
- Intimate partner violence (IPV)
Independent variables
Socio-demographic and socio-economic characteristics, intimate partner related factors, family related factors, pregnancy and reproductive history of the participants are predictors included for assessment of any association.
Data quality control
Training was given to data collectors on the purpose of the study and data collection process. Revision of the collected data and checking for the completeness before data entry was done. Pretest among (5%) of the sample prior to the actual data collection carried out to test the accuracy of the questionnaire to collect the intended data.
Operational definition
Intimate partner violence (IPV): A study participants who responds at least one YES response for the Likert Scale (sometimes, usually, and always) among thirteen items (five item for Psychological/emotional, five for physical, and three items for sexual) qualifies the respondent as victim for IPV[2].
Psychological or emotional violence: at least one YES response for the Likert Scale (sometimes, usually, and always) among five items for psychological violence qualifies the respondent for being faced with psychological/ emotional violence[2].
Physical violence (beating): at least one YES response for the Likert Scale (sometimes, usually, and always) among five items for physical violence qualifies the respondent for being faced with physical violence[2].
Sexual Violence: at least one YES response for the Likert Scale (sometimes, usually and always) among three items for sexual violence qualifies the respondent for being faced with sexual violence[2].