Randy A. Sansone, MD; Justin S. Leung, BA; and Michael W. Wiederman, PhD
Dr. Sansone is a Professor in Psychiatry and Internal Medicine at Wright State University School of Medicine in Dayton, Ohio, and Director of Psychiatry Education at Kettering Medical Center in Kettering, Ohio; Mr. Leung is a medical student at Wright State University School of Medicine in Dayton, Ohio; and Dr. Wiederman is a Professor in Psychology in the Department of Human Relations at Columbia College in Columbia, South Carolina.

The association between aggression/violence and mental disorders continues to be controversial. For example, a number of studies have reported associations between violence and mental disorders,[1,2] and violence risk appears to be empirically higher in individuals with psychosis,[3–6] including schizophrenia,[7,8] as well as among those with personality disorders and substance abuse.[7] However, Stuart[9] emphasized that, “…mental disorders are neither necessary, nor sufficient causes of violence.” In this study, using a consecutive sample of internal medicine outpatients, we further examined the controversial associations between aggression and mental disorders using a unique methodology.

Participants in this study were men and women, ages 18 years or older, who were being seen at an internal medicine outpatient clinic staffed predominantly by resident providers. The recruiter excluded individuals with symptoms of a severity that would preclude the candidate’s ability to successfully complete a survey (n=62, mostly language difficulties).

At the outset, 480 individuals were approached, and 369 agreed to participate (participation rate of 76.9%). Of these, 338 completed the relevant study measures, 230 (68.0%) women, 106 (31.4%) men, and 2 (0.6%) who did not indicate their gender.

Participants ranged in age from 18 to 87 years (median [M]=49.99, standard deviation [SD]=15.58), and most were White/Caucasian (86.0%) followed by African-American (9.3%). All but 7.5 percent had at least graduated high school whereas 29.4 percent had earned a four-year college degree or higher.

During clinic hours, one of the authors (JSL) positioned himself in the lobby of the outpatient clinic, approached consecutive incoming patients, and informally assessed exclusion criteria. With potential candidates, the recruiter reviewed the focus of the project and then invited each to participate by completing a six-page survey. Participants were asked to place completed surveys into sealed envelopes and then into a collection box in the lobby.

The survey consisted of three sections, the first being a demographic query. The second section explored participants’ past mental health history: 1) “Have you ever seen by a psychiatrist?” 2) “Have you ever been hospitalized in a psychiatric hospital?” 3) “Have you ever been in counseling?” and 4) “Have you ever been on medication for your nerves?” The third section contained an author-developed 21-item questionnaire designed to explore lifetime aggressive behaviors (labeled, “Aggressive Behavior Questionnaire,” or ABQ; see appendix). With yes/no response options, the total number of affirmative responses is considered a general measure of the different forms of aggressive behavior in which the respondent has engaged.

This project was reviewed and exempted by two institutional review boards. Completion of the survey was assumed to function as implied consent and was explained as such for the patients on the cover page.

Of the 338 respondents, 137 (40.5%) had been seen a psychiatrist, 50 (14.8%) had been hospitalized in a psychiatric hospital, 173 (51.2%) had been in counseling, and 149 (44.1%) had been on medication for their nerves. Scores on the ABQ ranged from 0 to 17 (M=2.56, SD=3.07), with 32.8 percent of respondents denying having engaged in any of the aggressive behaviors.

Scores on the ABQ are presented in Table 1 as a function of whether respondents reported each form of mental healthcare utilization. Across all forms of utilization, respondents who reported mental healthcare utilization had higher scores on the measure of aggressive behavior compared to their peers who did not report mental healthcare utilization.

To investigate the unique predictive power of each in explaining ABQ scores, we performed a multiple regression analysis in which all four forms of mental healthcare utilization were simultaneously entered as predictors of ABQ scores. As expected, the overall equation was statistically significant: R=0.40, F(4,331)=15.82, p<0.001. Importantly, only two forms of mental healthcare utilization were independently predictive of ABQ scores after controlling for the effects of the remaining predictor variables: having seen a psychiatrist (Beta=0.24, t=3.49, p<0.001) and having ever been on medications for nerves (Beta=0.20, t=3.27, p<0.001). Both having been in counseling (Beta=0.03, t=0.44, p<0.67) and having been hospitalized in a psychiatric hospital (Beta=0.01, t=0.17, p<0.88) were not uniquely related to scores on the ABQ. In this study, univariate analyses indicated that all mental-healthcare-utilization variables were significantly related to scores on the measure of lifetime aggressive behaviors. In multivariate analyses, only two variables remained uniquely predictive: being seen by a psychiatrist and having been on medications for “nerves.” While we did not examine psychiatric comorbidity, various factors may moderate these findings, such as substance abuse and/or personality pathology, both of which have previously been associated with violence and aggression. This study has a number of potential limitations. First, all data are self-report in nature. Second, we did not undertake any assessment of psychiatric diagnoses; therefore, we do not know the mediating and moderating roles of specific forms of mental disorders. Third, the ABQ is an unvalidated measure of aggression. Fourth, we cannot infer a causal relationship between mental disorder and violence based upon the current methodology (i.e., a cross-sectional approach). Despite these potential limitations, the sample size is reasonable, the study population of consecutive primary care patients is atypical in this area of study, and the use of the ABQ is a novel measure. Findings underscore relationships between aggression and past mental healthcare utilization. References 1. Brennan PA, Mednick SA, Hodgins S. Major mental disorders and criminal violence in a Danish birth cohort. Arch Gen Psychiatry. 2000;57:494–500. 2. Swanson JW. Mental disorder, substance abuse, and community violence: an epidemiological approach. In: Monahan J, Steadman HJ (eds). Violence and Mental Disorder: Developments in Risk Assessment. Chicago: University of Chicago Press; 1994:101–136. 3. Douglas KS, Guy LS, Hart SD. Psychosis as a risk factor for violence to others: a meta-analysis. Psychol Bull. 2009;135:679–706. 4. Fazel S, Gulati G, Linsell L, et al. Schizophrenia and violence: systematic review and meta-analysis. PLoS Med. 2009;6:e1000120. 5. Fazel S, Yu R. Psychotic disorders and repeat offending: systematic review and meta-analysis. Schizophr Bulletin. 2011;37:800–810. 6. Large MM, Nielssen O. Violence in first-episode psychosis: a systematic review and meta-analysis. Schizophr Res. 2011;125:209–220. 7. Gillies D, O’Brien L. Interpersonal violence and mental illness: a literature review. Contemp Nurse. 2006;21(2):277–286. 8. Tiihonen J, Isohanni M, Rsnen P, et al. Specific major mental disorders and criminality: a 26-year prospective study of the 1996 Northern Finland Birth Cohort. Am J Psychiatry. 1997;154:840–845. 9. Stuart H. Violence and mental illness: an overview. World Psychiatry. 2003;2:121–124.