Violence, mental health, and physical symptoms in an academic internal medicine practice

Abstract

Objective: To assess how physical and/or sexual intimate partner violence (IPV), child abuse, and community violence relate to long-term mental and physical problems; to examine the overlap between different forms of violence and the impact of experiencing multiple forms of violence.

Design: Cross-sectional survey.

Setting: Three general internal medicine practices affiliated with an academic medical center.

Participants: English-speaking women aged 25 to 60.

Measurements: Telephone or in-person interview and chart review.

Results: One hundred seventy-four women completed interviews. A majority of participants experienced more than one form of violence. In separate multivariate analyses, each form of violence was associated with depressive symptoms or with at least 6 chronic physical symptoms, after adjustment for demographic factors and substance abuse. The degree of association with health outcomes was similar for each form of violence (odds ratio [OR], 2.4 to 3.9; P < .003). The association with chronic physical symptoms remained significant for IPV (OR, 3.3; P < .002) and community violence (OR, 3.4; P < .003), even after adjustment for depression and posttraumatic stress disorder. There were dose-response relationships between the number of forms of violence experienced and the odds of depressive symptoms and the odds of multiple chronic physical symptoms.

Conclusions: Multiple types of victimizations may contribute to patients' current mental health and physical problems. Research or clinical protocols that only focus on one form of violence may underestimate the complexity of women's experiences and needs.

FIGURE 1

Lifetime prevalence of physical and sexual assaults. IPV, intimate partner violence; CA, child abuse; CV, community violence; any violence, any of the above.

FIGURE 2

Overlap of different forms of violence. Numbers represent number of participants reporting each possible combination of violence, followed by the prevalence rate in parentheses. IPV, intimate partner violence; CA, child abuse; CV, community violence.

FIGURE 3

Adjusted odds ratios of 6 or more physical symptoms by overlapping forms of violence. Number in each area represents odds ratio of having 6 or more physical symptoms after adjustment for age, race, income, tobacco use, and alcohol use. Further adjustment for household income, educational attainment, or drug use did not alter results. Numbers in parentheses represent P value. The group of women with no history of violence is used as the reference category. IPV, intimate partner violence; CA, child abuse; CV, community violence.