Article Text

Download PDFPDF

Violent victimisation is more common among people with schizophrenia-spectrum disorders than among the general population
  1. Robert A Schug
  1. California State University, Long Beach, Los Angeles, USA

Statistics from


Question: Do rates of recorded violent and non-violent victimisation differ among people with schizophrenia-spectrum disorders compared with the general population and have recorded rates changed over a 30-year period?

Population: 4168 adults diagnosed with schizophrenia-spectrum disorders (cases) were matched by age and year of birth to 4641 adults without schizophrenia-spectrum disorders randomly selected from the community (controls). Participants were aged 17–65 years. Cases were identified from a state-wide public mental health register comprising all people first diagnosed with a schizophrenia illness in the years 1975, 1980, 1985, 1990, 1995, 200 and 2005. Community controls were gathered from a representative electoral roll.

Setting: Victoria, Australia; 1975–2005.

Assessment: Recorded diagnoses were made by psychiatrists at the time of patient discharge from community, outpatient and inpatient services and coded using International Classification of Diseases (ICD) criteria as any diagnosis of schizophrenia, schizoaffective disorder, paraphrenia, shared psychotic disorder, delusional disorder or unspecified non-organic psychosis. Diagnoses were only included if they were supported in 75% of subsequent diagnoses if there was a clear diagnostic progression indicative of schizophrenia-spectrum disorders. Rates of victimisation over 30 years (1975–2005) were also compared between a subsample of schizophrenia spectrum cases and matched community controls.

Outcomes: Lifetime violent (any offence involving physical contact or harm to the victim including sexual violence) and non-violent victimisation. Incidents were identified from individual records on a state-wide police criminal records database containing details of all known offences and victimisation incidents. Rates of victimisation in each of the cohort years of diagnosis were calculated using data linkage.


Design: Case–control study.

Main results

People with schizophrenia-spectrum disorders were significantly more likely to have a record of violent victimisation compared to community controls (10.1% among cases vs 6.6% among controls; adjusted OR (AOR) 1.42, 95% CI 1.19 to 1.70). However, people with schizophrenia-spectrum disorders were less likely to have record of a non-violent victimisation or of victimisation overall (non-violent victimisation 25% among cases vs 36.8% among controls, AOR 0.45, 95% CI 0.41 to 0.50; victimisation overall 28.7% among cases vs 39.1% among controls, AOR 0.5, 95% CI 0.45 to 0.56). Sexually violent victimisation was significantly higher among people with schizophrenia-spectrum disorders compared to community controls (1.7% among cases vs 0.3% among controls; OR 2.77, 95% CI 1.76 to 4.36; whether this figure was adjusted was not reported). From 1995 to 2005, rates of overall recorded victimisation remained relatively constant among community controls (range 40–45%), while there was an increase in recorded rates among people with schizophrenia-spectrum disorders from 15.3% among those diagnosed in 1975 up to 37.4% in those diagnosed in 2005.


Compared with the general population, people with schizophrenia-spectrum disorders are more likely to have experienced officially recorded violent victimisation, but less likely to have experienced officially recorded non-violent victimisation.


The study used a broad definition of schizophrenia-spectrum disorders which encompassed the subtypes of schizophrenia as well as other chronic primary psychotic disorders. The study assessed victimisation across the person's lifetime; some recorded events may have preceded diagnosis.

Abstracted from



The literature on schizophrenia and criminal and violent offending remains somewhat transfixed in its infancy, perhaps due to hesitancy by researchers to move beyond the reification of this relationship (which has been supported and replicated across decades of studies) and embrace a second generation of systematic examinations in its aetiological underpinnings. Research linking schizophrenia to criminal victimisation currently lies in an even more embryonic state, therefore difficult to understand given that this line of inquiry should not be hindered by concerns of amplified negative stigmatisation of schizophrenia, which, arguably, fuels timidity and impedes progress in the former area of research.

The literature in this area has been based on self-report data, and the limitations to this methodology are well stated. Additional issues include the unique threats schizophrenia may pose to the validity of self-report data (ie, neuropsychological effects on recall and delusional systems on the interpretation of events); and the fact that the majority of these studies are from Australia, which may not be generalisable to other areas. Researchers worldwide are to be commended; especially in the present study, which demonstrates how mental health and criminal justice data management resources may be used to address this very important issue. The methodology of their case-linkage study is sound, and their results present an important next step in elucidating the schizophrenia–victimisation relationship.

Some concerns might include the fact that it is unclear how the victimisation data here was coded in the LEAP database (eg, how would a mutual combative event be coded?). Also, given the heterogeneity of schizophrenia, it would be interesting to know if specific symptoms of schizophrenia render one more susceptible to violence from others. Furthermore, the present study does not address the chronological sequencing of illness onset and crime victimisation.

Prospective research studies combining official record and self-report methodologies may provide the best empirical picture of the schizophrenia–victimisation relationship: what is likely a complex interplay between offending and victimisation, schizophrenia symptoms, substance use, and predisposing psychosocial and environmental factors to offending and victimisation. The authors’ speculations about results reflecting deinstitutionalisation are appropriate but should be interpreted cautiously.

View Abstract


  • Sources of funding: The Australian Research Council.


  • Competing interests None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.