Abstract
We examined interpersonal violence victimization and prior interpersonal violence exposure as predictors of suicide risk, namely, suicide ideation and suicide attempt, in a sample of 207 Indian college students. Results of regression analyses indicated that interpersonal violence victimization, but not prior interpersonal violence exposure, was a unique predictor of suicide risk, independent of age and sex. Finally, we found support for a significant Interpersonal Violence Victimization × Past Interpersonal Violence Exposure interaction effect consistent with the notion that the association between interpersonal violence victimization and suicide risk is potentiated by the presence of prior interpersonal violence exposure among Indian students.
Violence, including interpersonal violence or IPV (i.e. intentional use of physical force, verbal threats, or insults between family members or intimate partners), represents a worldwide problem, especially within developing countries (European Institute for Crime Prevention and Control and HEUNI, 2010; United Nations Economic and Social Commission for Asia and the Pacific, 2013). For example, IPV has become a pressing social problem in India (National Crime Records Bureau, 2013). Indeed, incidence of IPV nearly doubled in India between 1998 and 2006 (Simister and Mehta, 2010), and despite preventive legislation enacted by the Indian parliament in 2006, a recent study found rates of IPV as high as 31 percent among Indian adults (Kimuna et al., 2012). In turn, IPV victimization has been found to be associated with a wide range of negative psychological outcomes (e.g. stress, depression, and anxiety; Katz and Gurtovenko, 2015; Próspero and Kim, 2009), including heightened suicide risk (Cavanaugh et al., 2011; Randle and Graham, 2011). For example, findings from recent studies of college students have shown that IPV victimization is associated with greater suicide risk (e.g. Chang et al., 2015; Muyan and Chang, 2016). This positive association has been found in predicting both suicide ideation (e.g. Yanqiu et al., 2011) and suicide attempt (e.g. Seedat et al., 2005). Yet despite the growing problem of IPV in India, the relationship between IPV victimization and suicide risk in Indian adults remains unclear.
IPV victimization and suicide risk in Indian adults: Does past exposure to IPV make a difference?
As in many parts of the world (Fleischmann and De Leo, 2014), suicide among young adults represents a major public health problem in India (Vijaykumar, 2007). Indeed, the rate of suicide in India has been found to be comparable to those often found in other countries (e.g. the USA; Radhakrishnan and Andrade, 2012). Unfortunately, despite findings supporting the role of IPV victimization as a reliable predictor of suicide risk in culturally diverse adult populations (Stein et al., 2010), studies examining the association between IPV victimization and suicide risk (e.g. suicide ideation, suicide attempt) in young adult Indians have been limited. For example, in a study of adolescents and young adults from India, Pillai et al. (2009) found that the presence (or absence) of recent IPV was a significant predictor of suicide risk. Unfortunately, suicide risk in that study was simply coded as present (or absent) if there was any sign of suicidal thinking or attempts. That is, given that suicide risk varies across a spectrum (Carbajal et al., 2017), it would have been more useful to examine for degree of suicide risk, rather than simply the presence of suicide risk. Moreover, little is known about the potential impact prior exposure to IPV may have on suicide risk among Indians, on the relationship between IPV victimization and suicide risk among Indian adults, or on both. Notably, findings from some studies have indicated that prior exposure to IPV (e.g. witnessing a parent become a victim of IPV) is positively associated with poor psychological adjustment in adults (e.g. trauma symptoms; Evans et al., 2014). Thus, for example, any positive association that might be identified between IPV victimization and suicide risk among young adult Indians (e.g. college students) might itself be determined by the presence (or absence) of prior exposure to IPV. To date, however, no study has yet examined the potential impact of prior IPV exposure among young adult Indians. Accordingly, it would be important to examine the role of IPV victimization and prior IPV exposure in predicting suicide risk among young adult Indians.
Purpose of the present study
Given these considerations, we conducted the present study to (a) determine whether IPV victimization and prior IPV exposure are significant predictors of suicide risk (i.e. suicide ideation and suicide attempt) in Indian college students, and (b) determine whether any association found between IPV victimization and suicide risk in Indian college students might be moderated by prior IPV exposure, as would be indicated by a significant IPV victimization × past IPV exposure interaction effect.
Consistent with the reliable pattern found in studies of both European and Western adult populations (e.g. Cavanaugh et al., 2011; Chang et al., 2015; Randle and Graham, 2011), we expected to find IPV victimization to emerge as a significant positive predictor of suicide risk in Indian college students. In addition, consistent with some findings pointing to the potential role of prior IPV exposure as a positive predictor of suicide risk in adults (e.g. Evans et al., 2014), we predicted that prior IPV exposure would also be a significant predictor of suicide risk in Indian students. Finally, consistent with the possibility that the presence (or absence) of prior IPV exposure might moderate any association found between IPV victimization and suicide risk in students, we expected to find support for a significant IPV victimization × past IPV exposure interaction effect even after accounting for the potential effects of IPV victimization and prior IPV exposure in Indian students.
Method
Participants
A total of 207 Indian college students (87 males and 120 females) were recruited from a university in Northern Karnataka, India. Participants’ ages ranged from 20 to 27 years with a mean of 22.5 (SD = 1.36).
Measures
IPV victimization
To assess for IPV victimization, we used the Hurt, Insulted, Threatened with harm, and Screamed at them scale (HITS; Sherin et al., 1998). The HITS is a four-item self-report measure that assesses for the experience of IPV (e.g. ‘How often does your partner physically hurt you?’, ‘How often does your partner insult or talk down to you?’). Participants were asked to respond to each item using a 5-point Likert-type scale, ranging from 1 (never) to 5 (frequently). Evidence for the construct validity of the HITS scale has been reported in Sherin et al. (1998). Internal reliability of the HITS scale in the present sample was found to be good (α = .84). Higher scores of the HITS scale indicate greater IPV victimization.
Prior IPV exposure
To assess for prior IPV exposure, we used a single-item scale to assess for the presence or absence of a history of witnessing IPV (i.e. ‘Have you ever witnessed interpersonal violence among family members?’). Participants were asked to respond to the item with either 0 (no) or 1 (yes). A score of zero indicates no prior IPV exposure, whereas a score of one indicates prior IPV exposure.
Suicide risk
To assess for suicide ideation and suicide attempt, we used two items from the Suicidal Behaviors Questionnaire-Revised (SBQ-R; Osman et al., 2001). Suicide ideation was assessed by a question from the SBQ-R that tapped into thoughts about self-harm (i.e. ‘How often have you thought about killing yourself in the past year?’). Suicide attempt was assessed by a question from the SBQ-R that tapped into potential for making a suicide attempt (i.e. ‘How likely is it that you will attempt suicide someday?’). Participants were asked to respond to these items across a 5- or 7-point Likert-type scale, ranging from 0 or 1 (never) to 5 (very often) or 6 (very likely). Evidence for the construct validity of the SBQ-R has been reported in Osman et al. (2001). Higher scores on the SBQ-R items indicate greater risk for suicide ideation or suicide attempt.
Procedure
Approval for the study was obtained from the Institutional Review Board prior to data collection. Participants were solicited from various psychology classes (e.g. Introduction to Psychology, Social Psychology) in exchange for extra credit for their participation. Participants who provided written informed consent to participate in the present study were given a paper-and-pencil survey with some basic demographic questions (e.g. age and sex) and the present study measures. The survey was administered in small groups shortly after the end of class. No personal identifying information was collected. Information regarding local campus counseling resources for addressing mental health concerns was provided to participants at the end of the study.
Results
Correlations, means, and standard deviations for all study variables in our sample of Indian college students are presented in Table 1. As the table shows, all the variables were significantly correlated in the expected manner. As predicted, IPV victimization was positively and significantly associated with both suicide ideation and suicide attempt (rs = .33 and .30, respectively). Likewise, prior IPV exposure was also positively and significantly associated with both suicide ideation and suicide attempt (rs = .20 and .21, respectively). Notably, prior IPV exposure was positively associated with IPV victimization (r = .22).
Correlations, means, and standard deviations for all study measures in Indian college students.
N = 207.
p < .05; **p < .01; ***p ⩽ .001.
Examining IPV victimization and prior IPV exposure as unique predictors of suicide risk in Indian students
To determine the unique role of IPV victimization and prior IPV exposure as predictors of suicide risk in Indian college students, we conducted a pair of hierarchical regression analyses in predicting suicide ideation and suicide attempt. For each of the analyses, we controlled for demographic variables (i.e. age and sex) in Step 1. IPV victimization and prior IPV exposure were entered as a set in Step 2. Finally, we entered the multiplicative IPV victimization × past IPV exposure term in Step 3. To determine whether any of the predictors accounted for a small, medium, or large amount of variance in suicide risk, we used Cohen’s (1977) convention for small (f2 = .02), medium (f2 = .15), and large effects (f2 = .35) as a general guide.
Results for predicting suicide ideation and suicide attempt are presented in Table 2. In predicting suicide ideation, the set of demographic variables was found to account for a small (f2 = .03) but significant 3.0 percent of the variance. Within the predictor set, age was the only significant predictor (β = .19, p < .05). When IPV victimization and prior IPV exposure were entered as a set in Step 2, they were found to account for a small (f2 = .14) but significant 12.3 percent of additional unique variance in suicide ideation. Within the predictor set, however, IPV victimization was the only significant predictor (β = .32, p < .001). Finally, when the IPV victimization × past IPV exposure interaction term was entered in the last step, it was found to account for a small (f2 = .04) but significant 4.0 percent of additional unique variance in suicide ideation (β = .46, p ⩽ .001). Overall, the total prediction model was found to account for a medium (f2 = .25) 19.7 percent of the variance in suicide ideation, F(5, 201) = 9.54, p < .001.
Results of hierarchical regression analyses showing amount of variance in suicide ideation and suicide attempt accounted for by IPV victimization and prior IPV exposure in Indian college students.
N = 207.
p < .05; **p < .01; ***p ⩽ .001.
To visually inspect the manner in which IPV victimization and prior IPV exposure interacted with each other in predicting suicide ideation in Indian college students, we plotted the regression of suicide ideation on IPV victimization at low and high levels (±1 SD below and above the mean [2.91 and 9.32], respectively) among students with and without prior IPV exposure (see Figure 1). As the figure shows, the results of plotting this interaction are consistent with the notion that the negative influence of IPV victimization on suicide ideation is amplified, or potentiated, by prior IPV exposure.

Suicide ideation at low versus high IPV for students with and without past IPV exposure.
In predicting suicide attempt, demographic variables as a set were found to account for a small (f2 = .03) but significant 2.9 percent of the variance in suicide attempt. Within the predictor set, age was again the only significant predictor (β = .15, p < .05). When IPV victimization and prior IPV exposure were entered as a set in Step 2, they were found to account for a small (f2 = .10) but significant 9.2 percent of additional unique variance in suicide attempt. Within the predictor set, experiencing IPV was again the only significant predictor (β = .27, p < .001). Finally, when the IPV victimization × past IPV exposure interaction term was entered, it was found to account for a small (f2 = .04) but significant 3.6 percent of additional unique variance in suicide attempt (β = .41, p < .01). Overall, the total prediction model was found to account for a medium (f2 = .19) 15.7 percent of the variance in suicide attempt, F(5, 201) = 7.24, p < .001.
Again, to visually inspect the manner in which IPV victimization and prior IPV exposure interacted with each other in predicting suicide attempt in Indian college students, we plotted the regression of suicide attempt on IPV victimization at low and high levels (±1 SD below and above the mean, respectively) among students with and without prior IPV exposure (see Figure 2). As this figure shows, the results of plotting this interaction are consistent with the notion that the negative influence of IPV victimization on suicide attempt is potentiated by prior IPV exposure.

Suicide attempt at low versus high IPV for students with and without past IPV exposure.
Discussion
We conducted the present study to address a major limitation in the extant literature on the impact of IPV around the world, namely, the lack of any empirical studies examining for the relationship between IPV and suicide risk in an Indian population, specifically, in Indian college students. In that regard, consistent with past research findings based on other cultural groups (e.g. Cavanaugh et al., 2011; Chang et al., 2015; Evans et al., 2014; Muyan and Chang, 2016; Seedat et al., 2005), our correlational findings indicated that both IPV victimization and prior IPV exposure were positively associated with both suicide ideation and suicide attempt among Indian students.
Notably, when we conducted hierarchical regression analyses examining the relative contributions of IPV victimization and prior IPV exposure as predictors of suicide risk, independent of age and sex, we found that only IPV victimization consistently emerged as a unique predictor of both suicide ideation and suicide attempt among Indian students. Interestingly, however, in predicting both suicide risk outcomes, we found support for an interaction effect, consistent with the notion that the positive association between IPV victimization and suicide risk is potentiated by the presence of prior IPV exposure. Thus, for example, suicide ideation and suicide attempt were found to be highest among Indian students who indicated both IPV victimization and prior IPV exposure. Taken together, the present regression findings point to at least two important implications for understanding the relationship between IPV and suicide risk in young adult Indians. First, these findings indicate that although IPV victimization and prior IPV exposure are both associated with heightened suicide risk in Indian college students, IPV victimization plays a more robust role, compared to prior IPV exposure, in predicting suicide risk in Indian students. This may be due to the fact that IPV victimization, compared to prior IPV exposure, has a more direct and immediate impact on one’s mental health. Indeed, consistent with this pattern, findings from a global multi-site study of IPV involving more than 100,000 adults from 21 different countries, including India, indicated a stronger association between IPV victimization and suicide risk than between prior IPV exposure (e.g. witnessing IPV in the past) and suicide risk (Stein et al., 2010).
Second, despite the limited unique role that prior IPV exposure – compared to IPV victimization – may have on suicide risk, our interaction findings point to one important way in which prior IPV exposure might remain an important and central component of any thorough assessment of suicide risk among Indian students. Specifically, our findings indicate that prior IPV exposure (e.g. witnessing parents physically fighting) significantly and consistently potentiates suicide risk among Indian students who have, compared to have not, been victims of IPV. Accordingly, it would be important for counselors to consider prior IPV exposure among Indian students who have become recent victims of IPV and may be at heightened risk for suicide.
Beyond these implications, our findings also point to the potential value of working within the academic community to both raise awareness of the critical problem of IPV on college campuses (e.g. providing education linking IPV victimization to increased suicide risk among students; Ackerson et al., 2008) and to facilitate efforts to reduce IPV victimization among students (e.g. providing a range of diverse campus-wide services to ensure the emotional/physical safety of students on campus; Menning and Holtzman, 2015; Nguyen-Feng et al., 2015; Stewart, 2014). As our findings suggest, emphasis on suicide risk assessment and management should especially target those students who have experienced, and continue to experience, interpersonally stressful situations in and outside the classroom setting. In that regard, social workers will need to not only be mindful of the role of stigma associated with poor mental health, but also focus on increasing sources of social support (e.g. teachers, counselors, and family members) for those at risk of dying by suicide (Joe and Niedermeier, 2008). Beyond these proximal initiatives, our findings also point to the potential value of developing stronger and more effective distal programs that might help reduce exposure to IPV within Indian society in general (e.g. programs focused on reducing family/domestic violence; Dave, 2013).
Some limitations of the present study
Despite the importance of our findings, some limitations to the present study are worth mentioning. First, we focused our study on the potential impact of IPV on young adult Indian college students. Therefore, for example, it would be interesting to determine whether a similar pattern of findings would emerge had we studied an older Indian adult sample. For instance, Vijaykumar (2007) found that the highest rates of suicide risk among Indians were found among middle-aged adults. Second, and relatedly, it would be important to determine whether IPV victimization and prior IPV exposure remain important correlates of suicide risk in a clinical sample of Indian adults (e.g. actively suicidal Indian adults). Third, given that IPV has been associated with other important negative affective conditions, it would be useful to determine whether the present prediction model is useful in predicting other important outcomes (e.g. depression, anxiety) in Indian adults. Fourth, given that single-item measures were used for some of our study variables, it would also be useful to determine whether different findings might emerge when multi-item measures of our variables are used (e.g. suicidal ideation; Chang and Chang, 2016). Fifth, given the cross-sectional nature of the present study, it would be important to expand on the present findings to determine whether IPV victimization and prior IPV exposure have prospective associations with suicide risk across time. Finally, given that the scope of the present study was limited to IPV, it is important to consider the present findings in the context of the plethora of variables (e.g. social support, depressed mood, poverty, etc.) that have also been linked to suicide risk in adults. Accordingly, more complicated models of suicide risk in adults that consider the interplay of micro, mezzo, and macro factors will need to be considered in future research.
Concluding thoughts
Given the lack of studies examining the potential negative psychological impact of IPV among Indian adults, we conducted the present study to examine whether IPV victimization and prior IPV exposure were associated with greater suicide risk (i.e. suicide ideation and suicide attempt) in young adult Indians. Consistent with past findings obtained in studies of other cultural groups, findings from the present study indicated a positive association between IPV victimization and suicide risk, and between prior IPV exposure and suicide risk. Although IPV victimization, compared to prior IPV exposure, emerged as a consistent unique predictor of suicide risk, we found evidence for a significant IPV victimization × prior IPV exposure interaction effect. That is, we were able to show that the association found between IPV victimization and suicide risk in Indian students was potentiated by the presence of prior IPV exposure. Overall, these initial findings not only highlight the importance of studying the impact of IPV on suicide risk among young adult Indians, but they also underscore a need for researchers to appreciate the importance of studying IPV in developing countries where violence remains a central social problem.
Footnotes
Acknowledgements
We would like to thank members of the Perfectionism and Optimism-Pessimism (POP) Lab for their helpful feedback on earlier versions of the present work. In addition, the first author would like to acknowledge Tae Myung-Sook and Chang Suk-Choon for their encouragement and support throughout this project.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
