Physical violence against U.S. women around the time of pregnancy, 2004-2007.

BACKGROUND: Previous research shows that the prevalence of intimate partner violence (IPV) around the time of pregnancy varies from 4% to 9%, but no studies have distinguished between abuse rates by former versus current partners. PURPOSE: This study aims to estimate the prevalence of IPV among U.S. women shortly before and during pregnancy and to compare the rates and predictors of abuse perpetrated by current partners with the rates and predictors of abuse perpetrated by former partners. METHODS: Using data from 27 states and New York City, the prevalence of physical abuse by current and former intimate male partners was estimated among 134,955 women who delivered a singleton, full-term infant in 2004-2007. Multivariable logistic regression was used to determine the demographic, pregnancy-related, and stress factors that predicted the risk of IPV. RESULTS: Prevalence of IPV from either a former or current partner was 5.3% before and 3.6% during pregnancy. Prevalence of abuse by a former partner was consistently higher than the prevalence of abuse by a current partner. The three strongest predictors of IPV during pregnancy were the woman's partner not wanting the pregnancy (current: AOR=3.47, 95% CI=3.13, 3.85; former: AOR=3.22, 95% CI=2.90, 3.76); having had a recent divorce or separation (current: AOR=3.23, 95% CI=2.92, 3.58; former: AOR=3.54, 95% CI=3.20, 3.91); and being close to someone having a drug or alcohol problem (current: AOR=3.05, 95% CI=2.78, 3.36; former: AOR=2.97, 95% CI=2.70, 3.27). Maternal characteristics (age, education, race, marital status, woman did not want the pregnancy) were less important predictors. CONCLUSIONS: Assessments of abuse should ask specifically about actions by both current and ex-partners.