Course of cognition in mentally ill offenders and the implications for risk of violence: a 10-12-year follow-up study
Background: It is now well established that there are core cognitive impairments associated with a diagnosis of schizophrenia. In parallel with our increased understanding of these core deficits, our awareness that mentally ill offenders (MIOs) are at additional risk of cognitive impairment due to an increased rate of traumatic brain injury and substance abuse has also grown. Absent from the literature is evidence of whether these cognitive impairments change over longer periods of time in MIO’s and whether these changes, or baseline abilities, impact an individual’s risk of violence. Furthermore, the negative impact head injury has on an individuals’ cognitive, behavioural and psychological functioning is well documented. These changes can lead to an increased likelihood of violence and crime, yet there is currently a scarcity of knowledge regarding the prevalence of head injury within mentally ill offenders in Scotland and its association with risk-related outcomes. Aims: The aims of the present thesis were to; (a) Examine the course of cognition in N=49 mentally ill offenders who underwent neuropsychological assessment while in the State Hospital, Scotland in 2004-5, and assess whether baseline or change in cognition predicts violent incidents or risk at follow-up, and; (b) Examine the cross-sectional association between head injury, substance abuse and risk-related outcomes of all individuals within the forensic network in Scotland for whom data could be extracted (N=428). Hypotheses: (a) We hypothesized that processing speed, verbal comprehension, working memory, delayed verbal memory, delayed non-verbal memory, impulsivity, inattention and problem-solving would decline over a 10-year period, and that deficits in impulsivity, emotion recognition, working memory and delayed memory would predict patients’ risk-related outcomes in a sample of mentally ill offenders. (b) It was also hypothesized that the presence of head injury and/or substance abuse within patients would predict worse risk-related outcomes, namely: quantity of violent offences, risk of harm to self, risk of harm to others and severity of violent offences. Analysis: We conducted a series of repeated measures MANOVAs, MANCOVAs and hierarchical linear regressions in SPSS Statistics to test our hypotheses. Individuals with a primary or secondary diagnosis of a learning disability were excluded. Results: (a) Our results propose that cognitive abilities significantly change over time (F(1.51, 30.1) = 5.98, p = .011), but direction of change is ability dependent. We found that impulsivity (Effect Size (ES) = .253), inattention (ES = .233), working memory (ES = .288) and auditory delayed memory (ES= .268) worsen over time. Measures of impulsivity and working memory significantly predicted some, but not all, risk-related outcomes, however these effects became diluted once additional variables with shared variance were added into the predictive models. We did not find that traumatic brain injury, substance misuse or alcohol misuse significantly mediated change in cognition over time. (b) In the national cohort study, results suggested that head injury had a significant effect on HCR total scores, F(1,259) = 6.679, p = .010 (partial eta square = .025), violence during admission (χ2 = 5.545, p = .022) and violent offences at a .1 p-value only, F(1,259) = 3.495, p = .063 (partial eta square = .013). Drug misuse only had a significant impact on total violent offences, F(1,259) = 8.933, p = .003 (partial eta square = .033) and nothing else. Furthermore, the interaction between alcohol misuse and schizophrenia also only had impact on total violent offences, F(1, 259) = 7.516, p = .007 (partial eta square = .028). Head injury was not significantly associated with either historical or current self-harm, however alcohol misuse, drug misuse and schizophrenia were. Conclusions: Our results highlight the unstable nature of cognition in mentally ill offenders and the impact that head injury has on violence-related outcomes, over and above substance misuse and a diagnosis of schizophrenia. This has potentially renovating implications for clinical practice regarding risk management, assessment, and treatment planning.