Long-Term Violent Reoffending Following Forensic Psychiatric Treatment: Comparing Forensic Psychiatric Examinees and General Offender Controls

Abstract

Background: Long-term violent re-offending in forensic psychiatric (FP) patients vs. non-FP offenders is largely unknown. Methods: We studied rates and facets of long-term violent reoffending among 1,062 violent forensic psychiatric examinees (FPE) consecutively undergoing pre-trial, forensic psychiatric examination (FPE) in Denmark during 1980-1992. Altogether, 392 were sentenced to FP treatment (FPE+T); the remaining 670 examinees received ordinary non-FP sanctions (FPE-T). FPE+T were compared to 392 contemporary matched violent general offenders (GEN) without FPE or other psychiatric contacts and sentenced to ordinary non-FP sanctions. FPE data were linked to population-based registers with sociodemographic, psychiatric, and crime information, and we estimated relative risks controlling for birth year, sex, educational and marital status, and previous violent crime. Results: During follow-up (mean = 18.0-19.5 years), FPE+T and GEN had any violent recidivism rates of 43% vs. 29% [adjusted hazard ratio (aHR) = 1.5; 95% CI, 1.1-1.9], respectively. Corresponding findings for severe violence (21% vs. 14%; aHR = 1.3; 95% CI, 0.9-1.9) and recurrent violence (3+ violent convictions; 16% vs. 6%; adjusted odds ratio [aOR] = 2.5; 95% CI, 1.5-4.4) also suggested weakly to moderately increased risks in FPE+T, albeit non-significantly for the former. Comparing FPE+T to FPE-T suggested decreased risk of any violence (43% vs. 51%; aHR = 0.8; 95% CI, 0.6-1.1), severe (21% vs. 34%; aHR = 0.6; 95% CI, 0.4-0.8), and recurrent violence [16% vs. 22%; adjusted odds ratio (aOR) = 0.7; 95% CI, 0.5-1.0] in FP patients, though non-significantly for any violence and recurrent violence. Among all FPE examinees, violent reoffending was independently predicted by male sex, younger age, pre-index violent crime, personality disorder (vs. schizophrenia spectrum and other psychiatric disorder), substance use disorder, and 5+ hospital admissions. Conclusion: FPE examinees, untreated followed by treated, reoffend violently more often than GENs. Similar trends are suggested also for severe and recurrent violence suggesting a need for continua of services for FPE examinees, independently of medico-legal status (i.e., sentencing to treatment or not).

Keywords: facets of violence; forensic psychiatric evaluation; forensic psychiatric patients; long-term follow-up; violent reoffending risk.

Figure 1

Proportion not reconvicted for any violent reoffending as a function of time at risk among violent offenders who underwent pre-trial forensic psychiatric evaluation (FPE) in Denmark 1980–1992. We compared 392 FPE examinees consequently treated (+T) in forensic psychiatry (FPE+T), 392 matched violent general offender controls (GEN), and 670 FPE examinees with ordinary sanctions (e.g., prison) treated as usual outside forensic psychiatry (FPE-T). Numbers below graphs represent the remaining number of non-censored subjects that had not recidivated at the beginning of each 4-year interval indicated on the X-axis.