The relationship between homicide perpetration and mental illness is an emotive topic. It’s a devastating outcome, but in absolute terms it’s rare. The outcome with the most dramatic implications for public perception is homicide of a stranger, which is even rarer – around 1 per 14 million people per year (Nielssen et al, 2011). Despite this small overall contribution, there is a significant association between mental disorder and homicide in diagnostic groups such as psychotic illnesses (Whiting et al, 2022).
Overall, both in the UK and internationally, there was a fall in homicide rates from the mid-2000s into the mid-2010s. The current paper by Sandra Flynn and colleagues uses data from the National Confidential Inquiry into Suicide and Safety in Mental Health (NCISH) to answer two main questions:
- Did homicide by people with mental illness decrease along with total homicide in the UK?
- What is the pattern of sentencing for those with serious mental illness?
To understand the paper, a crash course in court process and sentencing is needed. When accused of an offence, there are several steps – arrest, charge, potential remand into custody (not yet convicted/sentenced), entering a plea, trial, conviction and sentencing. There are multiple ways that mental disorder can be relevant, and at each step there are options for transfer to hospital under the Mental Health Act for assessment and/or treatment.
In the current paper (Flynn et al, 2021), aside from the rates and proportions of homicide, two main parts of the process are considered. First is whether there was an alternative verdict to murder, e.g. manslaughter on grounds of diminished responsibility (where abnormal mental function substantially impaired ability to understand, judge or control conduct), or the rarer finding of not guilty by reason of insanity (NGRI). These alternative verdicts allow imposition of a hospital order, rather than prison sentence, which is the second aspect highlighted.
Homicide perpetration by those with mental illness is rare but devastating and perpetuates stigma.
Methods
The study is a consecutive case series of homicide offenders in England and Wales between 1997 – 2015 (including those convicted, offenders found unfit to plead [i.e. so unwell that unable to participate in court] and NGRI).
For these individuals, data was collated from:
- Psychiatric court reports (clinical history, mental state at time of offence, lifetime diagnosis of mental disorder)
- Police National Computer (previous convictions)
- If any contact with the main hospital/community mental health provider in their area, a questionnaire was sent to their psychiatrist (about history, treatment, final contact and views on prevention).
These data were used to examine patterns for five definitions of mental disorder: a schizophrenia diagnosis, abnormal mental state at time of the offence, patients in contact with mental health services, a diminished responsibility verdict, and an outcome of hospital disposal (hospital order rather than prison sentence). Mid-year population estimates were used to calculate age-standardised rates and proportions. Temporal and proportion trends were examined with Poisson models.
Results
Overall, there were 10,918 homicide offenders in the study period (91% male, median age 28). Psychiatric reports were obtained for 4,064 offenders (38%).
Results are presented in different combinations of time periods and definition groupings, so we’ll concentrate on the main findings here:
- General population homicide conviction rates fluctuated with an upward trend until a peak of 1.38 per 100,000 population in 2008. After this, convictions fell year on year to 0.78 in 2015.
- There was no significant fall in the number or proportion of perpetrators with mental disorder when defined by a diagnosis of schizophrenia, an abnormal mental state at the time of offence, or patient status.
- There was an overall fall in the number of homicides by those receiving a manslaughter verdict on grounds of diminished responsibility (IRR 0.98, 95% CI 0.96 to 0.99), or receiving a hospital disposal (IRR 0.98, 0.97 to 1.00).
The authors considered two time periods:
- 1997-2005: Rise in number of convictions by most definitions, significant increase in the number with schizophrenia, an abnormal mental state at time of offence, and patients; inversely, a fall in number and proportion of verdicts of manslaughter.
- 2006-2015: Fall in the number (but not proportion) of patients perpetrating homicide, but proportion with diminished responsibility verdicts and those with schizophrenia increased.
43% of patients with mental disorder had a secondary diagnosis of alcohol and/or drug misuse/dependence.
Of the 656 individuals with a schizophrenia diagnosis:
- 28% were found guilty of murder and 35% manslaughter (diminished responsibility)
- 35% received a prison sentence and 63% hospital disposal.
When using the definition of status as patients within 12-months of offence (n=1,143), the pattern is different:
- 47% were convicted of murder, 17% of manslaughter (diminished responsibility)
- 72% received a prison sentence, 26% hospital disposal.
There was a non-significant upward trend in the rate of people with schizophrenia receiving a prison disposal. There was a significant fall in those receiving a hospital order (IRR 0.98, 95% CI 0.96 to 0.99). A prison disposal was significantly more likely if there was substance misuse comorbidity, or a secondary diagnosis of ‘personality disorder’.
“Despite their complex needs, people with serious mental illness continue to be imprisoned.”
Conclusions
The authors summarise:
Over the 19-year period, there was a significant fall in the rate of homicide convictions in England and Wales since a peak in the mid-2000s, with a 43% fall between 2008 and 2015. Homicide by people with mental disorder has followed a similar pattern, but the fall has been less striking than in the general population. As a result, mental disorder is now becoming proportionately more important to homicide reduction.
The high proportion of prison disposals for those with mental disorder is highlighted, potentially explained by comorbid substance misuse.
“Addressing substance misuse comorbidity and maintaining engagement with services may help prevent patient homicide.”
Strengths and limitations
This is a nationwide cohort study that triangulates multiple data sources to draw some helpful observations, as part of a large and influential wider project. Findings are appropriately framed with reference to some caveats, including the reliance on pre-trial diagnosis (and that this is presented during trial), or prior contact with services. There are some further nuances not directly explained; for example, how any diagnostic discrepancies between psychiatric reports were dealt with.
Sending questionnaires to clinicians is a method worthy of some reflection, given that recalling aspects of treatment, history and views on prevention may be coloured by the serious untoward incident. In the event, this information was not reported in this paper.
Another part of the disposal puzzle is the “hybrid order” (Section 45A). These are prison sentences but with an initial hospital direction, and bring into play thorny issues such as partial culpability. Numbers are small, so would unlikely have a material impact on the findings, but they were on the rise during the study period (Delmage et al, 2015) and so information on where these fitted would have been interesting.
This nationwide cohort study triangulated multiple data sources to draw some helpful observations.
Implications for practice
For me, one main clinical implication is a simple, familiar refrain – we need to do better at supporting individuals with comorbid substance misuse and severe mental illness. Too often clinically these individuals fall between services (often provided by separate organisations), where substance misuse seems more an exclusion criterion than a flag of risk and need. It also reinforces that proportionate, individualised consideration of whether risk of violence is a clinical need should be a routine consideration in mental health services.
It is highlighted that only 63% of those with a schizophrenia diagnosis who perpetrated homicide received a hospital disposal. Is this inappropriately low? This is a question that can’t be answered by the current study. At an individual level, mental disorder is not a binary concept, and disposal is influenced by an array of factors including history, clinical factors, the offence particulars, treatment needs and optimal frameworks for risk management.
At a system level, pressure on secure hospital beds is high. This finite resource provides long-term care for those who have committed the gravest offences (and homicide recidivism in jurisdictions where these services are established is low [Golenkov et al, 2014]). However, the system also needs to retain capacity to accommodate in a timely fashion those acutely unwell individuals in prison who are in urgent need of treatment under the Mental Health Act. There are no easy answers, but even in my current role as a prison psychiatrist it would not be too controversial for me to champion the system-wide benefits of improving the provision of mental health care in prison (Piper et al, 2019).
Only 63% of homicide perpetrators with a schizophrenia diagnosis received a hospital disposal – is this too low?
Statement of interests
None.
Links
Primary paper
Flynn, S., Ibrahim, S., Kapur, N., Appleby, L., & Shaw, J. (2021). Mental disorder in people convicted of homicide: Long-term national trends in rates and court outcome. The British Journal of Psychiatry, 218(4), 210-216. doi:10.1192/bjp.2020.94
Other references
Delmage E, Exworthy T, Blackwood N (2015) The ‘Hybrid Order’: origins and usage, The Journal of Forensic Psychiatry & Psychology, 26:3, 325-336.
Golenkov A, Nielssen O, Large M. Systematic review and meta-analysis of homicide recidivism and schizophrenia. BMC Psychiatry. 2014 Feb 18;14:46. doi: 10.1186/1471-244X-14-46
Nielssen O, Bourget D, Laajasalo T, Liem M, Labelle A, Häkkänen-Nyholm H, Koenraadt F, Large MM. Homicide of strangers by people with a psychotic illness. Schizophr Bull. 2011 May;37(3):572-9. doi: 10.1093/schbul/sbp112. Epub 2009 Oct 12. PMID: 19822580; PMCID: PMC3080680.
Piper, M., Forrester, A., & Shaw, J. (2019). Prison healthcare services: The need for political courage. British Journal of Psychiatry, 215(4), 579-581
Whiting D, Gulati G, Geddes JR, Fazel S. Association of Schizophrenia Spectrum Disorders and Violence Perpetration in Adults and Adolescents From 15 Countries: A Systematic Review and Meta-analysis. JAMA Psychiatry.2022;79(2):120–132. doi:10.1001/jamapsychiatry.2021.3721. [PubMed abstract]
Photo credits
- Photo by Tingey Injury Law Firm on Unsplash
- Photo by John Cameron on Unsplash
- Photo by Hédi Benyounes on Unsplash
- Photo by Ussama Azam on Unsplash
- Photo by Shubham Dhage on Unsplash
- Photo by Claudio Schwarz on Unsplash
There are a number of problems with the NCISH report which in our view underestimates the impact of mental illness on homicide.
Firstly there has been a marked decline in the availability of psychiatric reports to court in such cases. In 1999 NCISH were able to obtain such reports in 70% of such cases, in 2016 they could obtain less than 50%, so are only reliably estimating the rate of mental disorder in only half the cases.
They acknowledged at the time this would lead to underreporting.
In Scandinavia where Psychiatric reports are routinely collected, a much higher rate of mental illness associated with homicide is recorded (see for example Fazel and Grann, Psych morbidity in homicide offenders Swedish population stud\y, 2004).
Also NCISH primarily focuses on mentally disordered offenders in touch with secondary services who have been convicted (so murder suicides are not included – as there’s no conviction) and fails to sufficiently include those in touch with only primary care or no service at all.
The recent London Violence Reduction Unit homicide study (Gdn 14 Nov 22) found much higher rates of mental illness associated with homicide than previously reported.