Factors associated with recidivism: a retrospective analysis of state patients admitted to Sterkfontein hospital

Abstract

Aim This study examined common factors associated with recidivism amongst state patients at Sterkfontein Hospital. More specifically, demographic, clinical and criminological factors of a recidivist group were compared to a non-recidivist group with the intention of understanding to what extent these factors might determine the likelihood of re-offending. Method A retrospective case file review of 293 inpatients and a random selection of 120 outpatients was conducted. For the purpose of the study a patient was classified as a recidivist if an additional charge or act of violence was added to the file whilst the patient was on leave of absence. Of the inpatients only those who met the criteria for recidivism were included in the study. All 120 randomly selected outpatients were included. Demographic, clinical and criminological data were captured for all patients. Results Of the 293 inpatients 60 patients met the criteria for recidivism. Twenty out of 120 outpatients were classified as recidivists. Thus 80 recidivists were compared to 100 non-recidivists. Using logistic regression analysis the odds of recidivism in a patient with an index offence of assault was 8.4 (95% CI 1.6- 43.1) times of that who did not commit assault as an index offence. The odds of recidivism for patients with cannabis use was 2.8 (95% CI 1.3- 6.0) and for patients with in-ward sexual offence was 17.2 (95% CI 2.0-150). Conclusion This study suggests that state patients with comorbid substance use disorder and antisocial personality disorder are at higher risk for reoffending. Clinicians should also be aware of the potential risk amongst state patients with a charge of assault and patients who are found to be involved in in-ward adverse events. Important criminal history factors and certain clinical factors could however not be interpreted due to large amounts of missing data in patients’ files.