Brief report Can lipid analysis help identify repeatedly violent patients after discharge from acute psychiatry? John Olav Roaldset a,b,c,n , Pål Hartvig c , Stål Bjørkly c,d a Psychiatric Department, Ålesund Hospital, Møre & Romsdal Health Trust, 6026 Ålesund, Norway b Faculty of Medicine, The Norwegian University of Science and Technology, 7489 Trondheim, Norway c Centre for Research and Education in Forensic Psychiatry, Oslo University Hospital, Bygg 7, Gaustad, 0320 Oslo, Norway d Institute of Health and Social Sciences, Molde University College, Norway, Box 2110, 6402 Molde, Norway article info Article history: Received 25 October 2012 Received in revised form 30 March 2013 Accepted 5 July 2013 Keywords: Prospective V-RISK-10 Bio-psychosocial abstract In this study of 196 patients discharged from an acute psychiatric ward,11 patients (6%) were identified as exhibiting repeated violence and having frequent readmissions. Compared with non-violent patients and those with only one violent post-discharge episode, repeatedly violent patients were significantly characterised by male gender, higher rates of previous threats of violence, lack of empathy, more severe violence during follow-up, and lower high-density lipoprotein levels. & 2013 Elsevier Ireland Ltd. All rights reserved. 1. Introduction Recent reviews of studies in psychiatric facilities have shown that a small group of patients generate a majority of the violence that occurs in these facilities (Bowers et al., 2011). However, research on patients with repeated violent episodes after dis- charge (PRVEs) is rare. Monahan found that 28% of admitted psychiatric patients committed a violent act during their first year after discharge from acute psychiatric wards and that 12% were involved in two or more violent acts (Monahan et al., 2001). An investigation of patients with violent episodes prior to hospitalisa- tion showed that 20% of the patients behaved violently during the first year after discharge from a short-term psychiatric ward and that 3% of the patients exhibited repeated violence (Amore et al., 2012). Despite some contradictory results, low lipid levels have been linked to violence (Hillbrand and Spitz, 1999) and impulsive aggression (Conklin and Stanford, 2008; Troisi, 2011). However, in our literature search, we found no lipid studies that discriminated between patients with one violent post-discharge episode (POVEs) and PRVEs. Based on data from a prospective study of acute psychiatric patients after discharge (Roaldset et al., 2011a, 2011b), the present investigation aimed to identify possible bio- psycho-social predictors or markers of PRVEs. 2. Methods All 489 acutely admitted patients during a 1-year period (2006–2007) in one psychiatric hospital in Norway were included. Baseline variables (demographic and psychosocial variables, lipids, platelet-serotonin, and Violence Risk Screening 10 (V- RISK-10)) were prospectively compared with the occurrence of violent threats and acts recorded during the first year after discharge. A fasting blood sample was obtained within 3 days after admission to measure blood lipids (total cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), and triglycerides) and platelet-serotonin. We obtained follow-up data for 196 patients (77% of the 254 patients who consented to blood testing and 40% of the total sample of 489 patients) for the first year after discharge. The physician, psychiatrist, or psychol- ogist in charge performed a V-RISK-10 screening at discharge (Hartvig et al., 2007). The V-RISK-10 items were as follows: (1) violent acts; (2) violent threats; (3) substance abuse; (4) major mental illness (psychosis); (5) personality disorder; (6) lack of insight; (7) suspiciousness; (8) lack of empathy; (9) unrealistic planning; and (10) exposure to future stressful situations. The outcome measures were incidents of violent threats and violent acts. A violent threat was operationally defined as a verbal or non-verbal communica- tion conveying a clear intention to inflict physical injury upon another person, and violent acts were defined as the actual infliction of intended bodily injury upon another person (Dean et al., 2006; Monahan et al., 2001). Information concerning violent behaviour after discharge was obtained from four sources: (i) inquiries by staff from the wards and outpatient clinics, who asked patients about violent behaviour; (ii) hospital and outpatient medical records; (iii) criminal records; and (iv) information from collaterals, if available. The data were analysed using SPSS, version 16.0. Analyses of variance (ANOVAs) and t-tests were used for continuous variables, chi-squared tests for categorical variables, Pearson' s r for correlations, logistic regression analyses for estimation of odds ratios and multiple testing, and the area under the curve of the receiver operating characteristic (ROC-AUC) for effect sizes of predictive accuracy. More detailed methodological information can be found in previous publications (Roaldset et al., 2011a, 2011b). The project was approved by the Regional Committee for Medical Research Ethics and the Norwegian Ministry of Health and Care. Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/psychres Psychiatry Research 0165-1781/$ - see front matter & 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.psychres.2013.07.010 n Corresponding author at: Psychiatric Department, Ålesund Hospital, Møre & Romsdal Health Trust, 6026 Ålesund, Norway. Tel.: +47 70106550. E-mail address: johnolr@gmail.com (J.O. Roaldset). Please cite this article as: Roaldset, J.O., et al., Can lipid analysis help identify repeatedly violent patients after discharge from acute psychiatry? Psychiatry Research (2013), http://dx.doi.org/10.1016/j.psychres.2013.07.010i Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎