Echoes of Violence: A Study Makes a Connection Between Recent Abuse and Mortality

Interview with Kathleen M. Weber, R.N. on the study, "The Effect of Gender-Based Violence (GBV) on Mortality: A Longitudinal Study of U.S. Women with and at Risk for HIV," at the 19th International AIDS Conference, July 2012 in Washington, D.C.

Enid Vázquez: Your findings were clear and fascinating, and I wanted to talk with you about the experiences and inspirations for the study, but let me start by asking if this is the first time there was a correlation found between abuse and survival?

Kathleen Weber, Cook County Health and Hospital Systems: This is the first time we saw a correlation between recent abuse and mortality. Even in this analysis, when we looked at just history of abuse alone without looking at current abuse separately, there was no correlation. And part of that is that 78% of our women have a history of abuse, so a sample of those who have never been abused is quite small. But the women who had experienced current abuse in addition to having a history of abuse were the women with whom we saw that increased mortality risk. It's been a long time and we really needed to have enough "dust," unfortunately, to occur to be able to actually look at this and then to control for all the other things that are associated with death like smoking, substance abuse, and depression. So in this model, all of those things are actually controlled for -- you need a longer period of time and a greater number of deaths to be able to do that, statistically. This is really the first time we've seen this in the WIHS data.

EV: In terms of the inspirations for the study, do you think they relate to critical consciousness as raised by Gwendolyn Kelso, for example, and why women stay in bad relationships after you give them referrals for help?

KM: When you look at the WIHS study and you look at the graphs, the abuse is definitely decreasing over time, which is a good thing. For women in the WIHS there was a reduction over the 17-year period in which this was looked at. It is telling us that women are accessing some services because every six months [when the women are seen], when they answer "yes" about being in an abusive relationship, they do get a referral. I think in the process of asking women regularly about abuse raises their awareness that maybe this isn't the experience other people are having -- otherwise they wouldn't see me here [at a referral site]. It brings the discussion to the table, not just in the WIHS study but in the provider relationship.

So we're thinking if over time you keep asking people about their experience, they start thinking about their experience, and then slowly seek access to available services or look for opportunities to get services. In a Cook County-related study in JAMA [Journal of the American Medical Association] in August, where they looked at whether referrals actually reduced violence, they didn't find any substantive randomized, controlled clinical trials, but we know from working with women that it takes a long time. You can't just say, "Here's a postcard and the address of a place where you can get help." It might take years and years of a relationship with a service provider and really concrete support to get women the help they need. You know, a postcard alone is better than nothing, but it's probably not the most effective approach to get women out of an abusive relationship, right?

EV: Some people don't get that. What are some other approaches?

KW: You asked about Gwen's research. She's looking at our Chicago WIHS in terms of racial and sexual discrimination. Sometimes, as you get a sense of what's happening and you become more involved socio-politically, things start to change for women as they become more empowered. Part of the study is an offshoot of a bigger study Mardge Cohen [M.D., CORE Center in Chicago, where WIHS is conducted] is conducting in collaboration with the Boston [University] group looking at gender roles and this concept of self-silencing, what women do in terms of a tendency to silence themselves to be in a relationship or caring for their children at the expense of taking care of themselves. If those are the expectations you feel you're supposed to fulfill as a woman, as a mother, as a partner, you can see where that might be setting yourself up for situations where you're more likely to be abused or in an abusive relationship and not able to get out. So that's been some of the work that's been happening here locally. It's starting to look more at women, gender roles, and particular cultures, what the norms might be, or the expectations. It gets complicated when women are single and they have HIV and the stigma of HIV might also play a part in that relationship.

EV: I was thinking your group started this study because there's a clear relationship between past abuse and addiction, HIV infection, and other concerns.

KM: Mardge had looked at the connection [of abuse, substance use, and race] to antiviral therapy in a study she did. In terms of WIHS, you ask questions and you get a sense of how much abuse has happened. In collaboration with the Boston University group, we started this study to have women do these biographical narratives. When the women started to tell their stories in an unscripted way, it was just incredible how many women had horrifying experiences as early as childhood and then beyond. We felt this was something to be aware of, and if you're not working with women closely you don't have an appreciation of how difficult these women's lives really are.

For us, on some level, people laugh. It's like, "Duh, stress is a bad thing. Abuse is a bad thing." But nobody ever really believes you until you show it.

EV: It needs to be documented.

KM: I don't think any one of us felt that abuse doesn't impact people and women's health outcomes. Of course it does. But it's a whole different thing when you can document it. People always want to see the evidence, because if there's no clear evidence that there's a direct link or what the link is and through what mechanism, you don't know how to intervene. You don't know what's going to be most effective.

So we really look at this as a first step in documenting that there is a clear association with something as severe as mortality. It cannot get any more severe in terms of a health outcome than that, right? So if you back up from mortality, now we're starting to look at earlier health outcomes. Do you see this manifest with respect to cardiovascular disease in HIV-infected women? Diabetes? Is this affecting other co-morbidities much earlier? But as a first step for us, if you can see it with mortality, than you're going to be able to find it elsewhere too. Mortality is a profound outcome. If you go for something so profound as your starting point and you find it there, then you can bet that it's starting much earlier than that.

Something's driving this mortality. This isn't because women were beaten up. This is not because women were thrown down the stairs or thrown in front of trains. This isn't homicide or suicide. Those things contribute a little bit to the findings, but they don't explain everything that we're finding as a relationship between current abuse and mortality. There's some biologic change that occurs as a result of being in an abusive relationship.

EV: I worked on an issue with women from the West Side, in the Project WISH study at the University of Illinois, and it was astounding to me to hear the same story of abuse over and over again from each woman. You talked about women not being allowed to go to their medical visits by their [male] partner. Talk about some of these influences and inspirations you had for the study.

KM: In the WIHS, in the section where we ask about violence, that is one of the questions. Does your partner prevent you from using the phone? From contacting your friends? From going to your necessary medical visits? Those are questions that are asked. For me, thinking that anybody would be able to stop you from going to your medical visit in this country, in this day and age, is really baffling. You're sitting at home and you know you need your HIV care, you need your HIV medications, and someone is actually preventing you from doing that or from seeing friends. Sure, there's some level of lack of social support and external contact if someone is really controlling you to the point that he is preventing you from leaving your home, preventing you from doing the things you like to do. I guess in many ways you can imagine that the lack of social support and activity with people who care about you would have some impact, the direct effect of not being able to go to your research facility or care appointment. I think in Rwanda [where Dr. Cohen is working with HIV-positive women] Mardge talked about some examples of this in which the women were HIV-infected and the husbands were HIV-infected, but they didn't want to be seen going into the HIV clinic, so they weren't going to the clinic for their own medication and they were taking their wives' medications. Those different levels and types of abuse really drove us to look at the data to see if we can see in the data what we see when women are in front of us actually telling their stories. You can tell how horrible this makes them feel. I think we always thought this finding would come out. It's just that so many of the ways in which we were handling the data before [weren't right for our study] -- for example, with depression -- as people have more abuse, they become more abused, and it's a cyclical thing. So the new approach we were able to take with this collaborate from the University of North Carolina really helped us look at the data and use some models that we didn't have available to us before.

EV: You're talking about the marginal structural models? I didn't discuss that aspect in my article.

KM: It is a mechanism where we can look at longitudinal data and you can make more sense out of all the contributing factors and then how those individual factors change over time. If you were just looking at it cross-sectionally, one time point, you probably wouldn't be able to appreciate the findings as much. It wouldn't have shown up as glaringly.

EV: What message would you like to give to women, to men, to any group?

KM: For me, I just feel that the most important message is really that abuse does have a profound effect on women. If it manifests itself in an outcome as terminal as mortality, it's really something that deserves attention. I just really don't feel like enough attention has been placed on the impact of violence on the health of women. I feel that you see a lot more attention now on violence against girls and women, and I think it's really becoming a topic to think about. I think it has to take place in every venue. It has to be a community discussion. It has to be a health care provider question as they come in for care, for every type of contact. It has to become more a discussion that we have openly to increase awareness. I think there's a lot more awareness in this country than there used to be, but we have a long way to go. And then you think about what it's going to take to really change and reduce the amount of violence that women and children really experience. Our women in the study have a really high rate of childhood violence exposure as well, directly and indirectly. So I just think it's good to begin to have some data to have dialogue about the effects of violence on health.

EV: Are the effects of violence the same on men?

KM: Absolutely! There is literature [research] describing the association between being abused and subsequently being more likely to abuse others. Men are most definitely not exempt from the violence epidemic. If we begin to ask, certainly we will find (and others have already reported) that abuse is as common among men as it is among women and that the impact of abuse is just as devastating.

While special attention needs to be paid to the most vulnerable groups, those least likely to protect and advocate for themselves, the message really is ... we need to collectively work toward ending violence of all kinds against all individuals by speaking out and adopting a zero tolerance approach in all settings (public and private, domestic and international, etc).

Got a comment on this article? Write to us at