Effects of a Psychosocial Intervention on Gender-Based Violence, Antepartum Depression, General Health and Adoption of Safety Behaviors in Pregnant Women in Kisumu County, Kenya

Abstract

Gender-based violence is a prevalent problem affecting pregnant women globally. Many adverse and often fatal maternal and fetal outcomes have been linked to violence during pregnancy. The study objectives were; to determine the prevalence of GBV during pregnancy and the factors associated with it, and to establish the effect of a psychosocial intervention on GBV, antepartum depression, perceived general health and adoption of safety behaviors in pregnant women in Kisumu County. This was a longitudinal study, in which 12 public health centers were randomized to provide a psychosocial intervention together with the usual antenatal care or the usual antenatal care alone. Antenatal care attendees in the selected facilities were first surveyed and screened for GBV, and eligible GBV positive women enrolled for a follow-up. The facility-based psychosocial consisted of three 30 minutes sessions of active non-judgemental interviewing, sharing of unbiased information on the cycle, magnitude and adverse effects of GBV, empathetic listening, encouragement and validation of participants’ feelings, safety planning and provision of a resource card. Data was collected using a semi-structured questionnaire and validated standardized tools and analyzed in SPSS. Multivariable logistic regression was used to determine the risk factors and analysis of covariance used to estimate the intervention effect on the quantitative outcomes, with the baseline scores as covariates. Chi-square was used to test for equality of proportions at 5% level of significance for categorical outcomes. Six hundred and ninety-one (691) pregnant women were surveyed and screened for GBV. Two hundred and eighty-eight (288) were enrolled for follow-up, 144 in each arm. The mean age of the survey participants was 24.5 years, standard deviation (SD) = 4.3 years and the age at sexual debut was 16.7, SD=2.2 years. Almost half of the pregnant women (48.6%) had experienced some form of GBV (physical, emotional, sexual or a combination of these). Forty-two percent and 23.4% reported experiencing physical and sexual violence respectively, in the past one year, and 39.2% had experienced physical violence during the current pregnancy. The individual and intimate partner characteristics associated with an increased risk of violence during pregnancy were; having a post-primary level of education in the women [OR=2.088], occasional alcohol consumption by an intimate partner, [OR=2.843], witnessing violence as a child [OR=3.380] and prior experience of physical [OR=13.116] and sexual violence, [OR=4.208]. Relationship factors associated with violence were; male partner dominance in decision making, [OR=5.930] and infidelity by the woman [OR=3.442] or her intimate partner, [OR=9.906]. The belief in the social superiority of a man [OR=3.949], man’s right to assert over a woman [OR=3.163] and the belief that women should tolerate violence to save a xxii relationship/marriage [OR=9.493] were associated with increased risk of violence during pregnancy. The difference between the psychosocial intervention and the usual care groups in the total intimate partner violence and physical violence scores was significant post-intervention, with small effect sizes of partial eta = 0.20 and 0.31 respectively, but the groups did not differ significantly in the proportion reporting other acts GBV by an intimate and non-intimate partners (p 0.05). The intervention arm had a significantly lower mean depression score of 5.34, SD=4.23 versus 12.46, SD=4.22 in the usual care group post-intervention, (effect size=0.50), higher mean general quality of life score, 40.03, SD=8.3 versus 27.36, SD=16.7(effect size=0.25) and higher mean of adjusted safety behaviors, 8.82, SD=2.3 versus 5.56, SD=2.0 (effect size =0.61). GBV during pregnancy, particularly violence perpetrated by intimate partners is high in Kisumu County. Individual characteristics of women and their partners, relationship factors (decision making and infidelity) and beliefs that influence a woman’s attitude and response to GBV were risk factors for violence during pregnancy. The ANC based psychosocial intervention resulted in meaningful and significantly lower total IPV, physical violence by an intimate partner and antepartum depression scores, and improved the women’s perception of their general health as well as the adoption of safety behaviors. The intervention however did not have a meaningful effect in the reduction of severe combined violence, emotional violence, harassment, acts of GBV by intimate partners (refusal to use a condom and economic violence) and non-partner physical and sexual violence. The County Government of Kisumu in collaboration with other stakeholders needs to engage in advocacy against GBV and come up with systematic community-led initiatives to promote changes in social norms, beliefs and behavior that entrench gender inequality hence GBV against women in general and pregnant women. There is an urgent need to routinely screen pregnant women attending ANC for GBV in order to identify those experiencing violence for follow up support and care. In the context of human resource constraints, trained GBV advocates (community workers and volunteers) can be used to engage with abused women to offer psychosocial support, empower them and link them with community services including available psychological interventions.