When studying or treating violence against women, patients are often considered survivors of a particular form of trauma. Our study, however, illustrates the importance of looking at the full picture of violence. The majority of participants (53%) reported experiencing multiple forms of violence. While the magnitude of association between violence and health outcomes was similar regardless of the type of violence, the strong dose-response relationship between depressive or physical symptoms and the number of forms of violence suggests the need to examine the cumulative impact of violence on health.
This study has several limitations including modest response rate. Women who recognized an association between their traumatic experiences and health may have been more likely to participate, biasing findings toward greater association. Moreover, we had few nonwhite participants. Thus, it may not be possible to generalize our results to women of color. As the sample size is rather small, no conclusions can be drawn about undetected associations. We did not have adequate power to assess for associations between different forms of violence and individual physical symptoms. Larger studies have found associations between IPV and a host of physical symptoms or illnesses.8,9
Given the great breadth of these symptoms and illnesses, we feel that the presence of multiple physical symptoms may be a better clue to victimizations than any particular clinical symptom. Similarly, our study was not powered to make distinctions based on severity and frequency of violence or time since abuse. Less than 5% of participants disclosed that they were experiencing current violence (i.e., within the past year). Limiting the analyses to women with past violence alone did not change the results. It is possible that some of the events described did not represent the full picture of what occurred, since we only used self-report data. We used very strict criteria in our definition of violence, excluding events where there was only emotional abuse or low-level assaults, to control for differences in what women may consider abusive. As a cross-sectional survey, this study can only examine associations, not causations. We cannot determine whether the violence described by participants led to their health problems, or whether it is simply a marker for other unmeasured circumstances that cause depression or chronic physical concerns.
Despite these limitations, there are important findings and implications. We found a higher than expected lifetime prevalence of violence. It is possible, given our modest response rate, that our sample was biased toward women with a history of violence. However, even if we assumed that all nonresponders had never experienced violence, the prevalence in our population would remain high. The rates of violence in our study are higher than those found in community samples47–51
or studies of primary care populations drawn from health maintenance organizations4,52
or community-based clinics,9,15
but are similar to other studies drawn from inner-city, public hospitals or academic medical centers.53–55
Those study populations have much greater racial diversity and lower educational levels than ours. The high rates of violence and disability and the low incomes and health function in our mostly white, well-educated patients highlight the challenges faced by academic medical centers and their patients, even in less racially diverse cities.
Our secondary analyses further the literature by showing that the association between IPV or CV and physical symptoms persists even after adjustment for depression and PTSD. This finding supports the notion that depression and PTSD should be seen as important comorbid illnesses that influence the health of violence survivors, but do not fully explain physical symptoms.
One can speculate as to why or how violence increases the risk of depression or physical symptoms. Only 3% of women in our study had experienced CV alone. Perpetrators of family violence classically berate their victims, isolate them socially, control their activities, and alternate between extreme displays of love and abuse.56,57
Cognitive-behavioral theory attributes depression to maladaptive thoughts and behaviors.58,59
A victim may internalize negative beliefs about herself, which can then lead to depressive thoughts. The social isolation and control may cause her to adopt behaviors that also increase risk for depression. In our study, the number of different forms of violence was most associated with current risk of depression. It may be that experiencing violence from multiple different perpetrators and in multiple different situations may further reinforce negative beliefs that had been instilled in a victim by an abusive family member or partner.
Less is known as to why violence is associated with chronic physical symptoms, but one can imagine that the nature of abuse may contribute to the development of somatic complaints. Women are likely to have been told by their abuser that they are “crazy” or that the problem is “all in [their] head.” Moreover, survivors often feel that providers and others blame them for choosing to stay in an abusive relationship.56
An abuse survivor may be more hesitant to accept a mental health diagnosis, may interpret it as proof that her provider thinks “there is something wrong” with her for getting into an abusive relationship, or may be more likely to experience mental distress as physical symptoms.
Alternatively, it is possible that violence causes lasting changes in the neuroendocrine system that chemically mediate an increase in symptoms. Though evidence is emerging for neuroendocrine abnormalities in patients with chronic mental or physical distress,60–67
little is known about the mechanisms linking violence to such changes.
Studies that only assess one form of violence may make inferences that are confounded by participants’ other violent experiences. Similarly, studies that lump all violence together in a dichotomous fashion may be missing the importance of multiple victimizations. Researchers interested in the relationship between violence and health need to consider the cumulative effect of many different forms of violence.
Clinicians need to be aware that a patient may not simply be a survivor of one form of violence. Most efforts to train clinicians focus on a specific form of violence.56,68–70
However, clinicians need to take into account the complexity of women's experiences and needs. Conversely, when seeing a patient with depressive symptoms or multiple physical complaints, clinicians should consider the possibility that a lifetime of violence may be diminishing her health and should be aware of the repetitive nature of violence and the increased risk of future victimization.
Further research is needed to understand the relationship between violence and health, to test the effectiveness of interventions meant to decrease the impact of violence, and to explore ways to prevent future violence.