Defense Date

2009

Document Type

Thesis

Degree Name

Master of Public Health

Department

Epidemiology & Community Health

First Advisor

Derek Chapman

Abstract

Background: Intimate partner violence (IPV) during pregnancy increases the risks of adverse outcomes for both mothers and their unborn children, including maternal and fetal death. However, more research is needed to determine if IPV increases in frequency or severity during pregnancy and to determine what the risk factors are for IPV during pregnancy. Objectives: To use data from the Chicago Women’s Health Risk Study to determine (1) if abuse is more prevalent during the pregnancy period, 2) if abuse during the pregnancy period increases in frequency or severity, 3) if pregnant women who are abused are at increased risk for intimate partner homicide, and 4) what the risk factors are for intimate partner violence during pregnancy. Methods: A chi square test of independence was performed on the crosstabulation of the pregnancy and the abuse variables. The means of the scores on three validated abuse measures for women recently pregnant and not recently pregnant at the time of interview were compared using an independent samples t-test. Chi square tests of independence were performed on crosstabulations of abuse frequency and severity variables and the pregnancy variable. Logistic regressions were performed to generate crude and adjusted odds ratios for IPV for the sample characteristics, first for the complete sample and then for the recently pregnant subsample. Results: The prevalence of IPV was about the same in the recently pregnant (68.2%) and recently not pregnant samples (71.1%). The chi square value for the crosstabulation of the pregnancy and the abuse variable were not significant (X2 = 0.606, df = 1, p = 0.436). HARASS scores were not significantly different for recently pregnant and recently not pregnant women. Power and Control scores were significantly lower for recently pregnant women (t = -2.081, df = 483, p = 0.038), however this difference was very small (mean difference = -0.317, SE = 0.152). Danger Assessment scores were not significantly different for recently pregnant and recently not pregnant women. The chi square value on the crosstabulation of the abuse frequency variable and the pregnancy variable was not significant (X2 = 0.344, df = 1, p = 0.557). The chi square value on the crosstabulation of the abuse severity variable and the pregnancy variable was not significant as well (X2= 0.412, df = 1, p-value = 0.521). Adjusted odds ratios for IPV for the pregnant subsample indicated that the only factor that increased risk was having between 0 and 6 social supports (aOR = 12.39, 95% CI = 3.27 to 46.88). Conclusions: In this high-risk sample, abuse was not more prevalent during the pregnancy period. Abuse during the pregnancy period did not increase in severity or intensity. Furthermore, pregnant women were not at greater risk for intimate partner homicide. Having fewer social supports put recently pregnant women at greatest risk for abuse. This may be because abusers frequently employ tactics to isolate victims from social supports in order to better maintain control of their victims. Having fewer social supports is particularly risky for this group, as pregnant women need more outside support to negotiate the demands of childbearing. More research is needed to determine the unique risk factors for domestic violence during pregnancy.

Rights

© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission

December 2009

Included in

Epidemiology Commons

Share

COinS