Association between Intimate Partner Violence and Posttraumatic Stress Disorder: A Case-Control Study

Background: Intimate partner violence (IPV) against women is a common form of interpersonal violence in both developed and developing countries, and represents a forensic and public health problem. IPV is related to Post-traumatic Stress Disorder (PTSD). This relationship however, has not been investigated in Colombian population. Objective: To determine the strength of the association between IPV and PTSD in women referred for forensic psychiatric evaluation in Bucaramanga, Colombia. Method: A case-control study was designed. A total of 132 cases involving women referred for forensic psychiatric evaluation met criteria for PSTD according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV). A group of 262 women without PTSD were taken as controls. First, odds ratio (OR) was computed. Logistic regression was used to control confounding variables. Results: A total of 76 (56.6%) in the case group reported IPV during the past year compared to 85 (32.6%) in the control group. IPV and PTSD were associated (OR=3.09, 95%CI: 1.58-6.03) after controlling for age, employment, medico-legal loss or injury, and current aggressor (partner). Conclusions: IPV increased the risk for PTSD three-fold among women attending forensic assessment in Bucaramanga, Colombia.


Introduction
Traumatic experiences are relatively frequent. The intimate partner violence (IPV) against women is a common exposure to interpersonal violence in developed and developing countries and represents a medico-legal and public health pro-blem (1). A comprehensive deÀ nition of IPV includes physical, sexual, emotional and À nancial abuse by current or former partner (2). Last year's prevalence of IPV was around 15% among women attending American health care services (3,4), and up to 50% in several settings of other countries and cultures (5-7). We should bear in mind that the prevalence varies according to the deÀ nition of IPV used (8).
The IPV results in signiÀ cant negative physiological and psychological consequences, with longterm and immediate negative health effects (9)(10)(11). IPV is associated with chronic physical symptoms and poor physical health (12-14), and signiÀ cantly increases the risk of emotional distress and formal mental disorders (15-21).
The posttraumatic stress disorder (PTSD) is the most diagnosed and deteriorating mental disorder that appears after traumatic events among women (22). Physical violence is associated with a higher risk of PTSD (23). In addition, females are at a higher risk of PTSD than men (23)(24)(25)(26). Several studies report a signiÀ cant relationship between IVP and PTSD among women. They have found Odds Ratios (OR) between 2.0 and 3.0, after controlling other variables (27,28). However, the strength of this link has not been explored within any Colombian population using a case-control design study. In forensic settings, there are special considerations; people may distort information and symptoms for many reasons, for instance, self-justiÀ cation or À nancial gains (29). Women who report IPV are more likely to access mental health and emergency services (20,30,31) and need forensic evaluation (32). The forensic evaluation must consider physical and mental aspects of the IPV of the woman's health in order to reduce its negative impact, suggesting integral intervention (33). Moreover, Colombian and other countries' laws increase the sentence if PTSD is diagnosed as a consequence of an assault (34).
The objective of this research was to À nd out the strength of association between IPV and PTDS, after controlling some confounding variables, among women attending a medico-legal setting in Bucaramanga, Colombia.

Method
An un-match case-control study was carried out. The research protocol was approved by the ScientiÀ c Research Division of the Instituto Nacional de Medicina Legal y Ciencias Forenses of Colombia (INML-CF). Participants signed an informed consent form according to the Helsinki Declaration and Colombian research laws.
Participants included women older than 14 years residing in Bucaramanga, Colombia. They were referred for medico-legal clinical assessment due to personal injuries, determining age, level of alcohol consumption, and pregnancy between April and June in 2004. A total of 599 women referred to INML-CF by judicial authorities were consecutively assessed. The Structured Interview for Axis I Diagnosis was administered to diagnose PTSD (35), according to Diagnostic and Statistical Manual for Mental Disorders DSM-IV (36). A total of 132 women met criteria for PTSD, only or with comorbidity, and were deÀ ned as cases. Then, two controls were selected by each case (n=262). Controls were participants without PTSD. Cases and controls with a same-sex partner or those who refused to participate were excluded from the study. Cases and controls were asked about any kind of violence experienced (past and current), including IPV by former or current partners. The mean age of cases was 30.6 years (SD=12.2) compared to 30.4 years (SD=12.7) in controls (t=0.181, df=394, p=0.857), and mean years of formal schooling was 8.0 years (SD=3.8) in cases versus 8.3 years (SD=4.0) in the control group (t=0.715, df=394, p=0.475). Both variables were dichotomized for the À nal analysis. See these categories and other characteristics of cases and control in Table 1.
For statistical analysis IPV was taken as the independent variable; PTSD as the dependent variable; and others as covariables (age, formal education, employment, last stable partner within the past year, socio-economic status, religion, current intimate partner aggressor, and medico-legal incapacity greater than 8 days as an index of the severity of the assault). Unconditional logistic regression was computed to control some of the confounding covariables (variables with values of probability lower than 0.20). Odds Ratios were calculated with 95% conÀ dence interval (95%CI). All analyses were carried out with STATA 9.0 software (37).

Results
A total of 76 (56.6%) in the case group reported IPV during the past year compared to 85 (32.6%) in the control group. The difference was statistically signiÀ cant (OR=2.86, CI95% 1.86-4.40, p=0.001). All confounding variables (older than 18 years, unemployed, medico-legal incapacity greater than 8 days, and current intimate partner aggressor) were included in an unconditional regression logistic model in order to control confusion. This model is presented in Table 2. Hosmer-Lemeshow's goodness-of-fit was adequate.

Discussion
This un-match case-control study corroborates the important association between IPV and PSTD among women referred for medicolegal assessment in a developing country. In the present research, it was found that IVP increases the risk of PTSD almost three-fold. Similar À ndings have been reported by other investigations. O'Campo et al. observed OR of 2.3 in a sample of 89 relatively highly educated, middleclass working women, between 21 and 55 years old, drawn from a metropolitan area in the United States (27). And, Fedovsky et al. reported OR of 3.0 in Spanish-speaking women, aged between 18 and 64 years, attending a primary care clinic in a large, urban, public hospital in the US (28). However, Yoshihama & Horrocks did not À nd a statistically signiÀ cant association between IPV and PSTD, OR of 2.1 and 95%CI 0.87-5.5, in a community-based random sample of 211 women, between 18 and 49 years of age, of Japanese descent born in the United States or Japan (7). The lack of association was probably due to the relatively small sample size used in the study.
The PSTD is a complex mental disorder associated with many important negative experiences, both past and present. Clinical presentation of PTSD needs interaction between constitutional and environmental factors, not well understood yet (38). The IPV is highly prevalent in the community of developed and developing countries (3)(4)(5)(6)(7)39,40); and it is a common factor for PTDS in women around the world (22,23). Often, female victims of IPV report the assaults to legal authorities and are referred for medico-legal, physical and psychiatric assessment. Forensic psychiatrists must evaluate these women without any bias (29). As it was expected, an important number of these women met full criteria for PTSD (22). In consequence, IPV represents an epidemic public health concern that predominantly affects women and contributes to the burden of PTSD in females (38). Forensic psychiatrists have to refer any women victim of IPV to mental health services. Lipsky & Caetano reported that women who experienced IPV were at increased risk of not receiving mental health care (41). This is the À rst case-control study demonstrating the relationship between IPV and PTSD in a forensic setting. However, it has some limitations with regards to its design and population. PTSD is a mental disorder associated with multiple factors and direction of causality cannot be determined. All participants were women, and it is possible that these correlations do not exist among men. In addition, medico-legal assessment could be affected by other issues, such as simulations by women who making false claims for compensation from male aggressors. It concludes that IPV increases the risk for PTSD three-fold among women attending medico-legal assessment in Bucaramanga, Colombia. More complex design researches need to be developed in order to establish the exact strength of the relation between IPV and PTSD.