Physical multimorbidity in severe mental illness: incidence insights from a Queensland statewide cohort

People with schizophrenia-spectrum or bipolar-spectrum disorders die 10 to 20 years earlier than the general population, with much of this gap a result of premature mortality from chronic physical disease. Whilst the general population continues to see improved life expectancy, this mortality gap for people with severe mental illness has barely narrowed in decades. What has remained unclear is how early, and how broadly, the physical disease burden behind this mortality gap actually accumulates.

Our new study, just published in The Lancet Psychiatry, set out to answer that question. Rather than looking at individual disease pairs, we tracked systemic physical multimorbidity: the cumulative number of distinct organ systems, out of 11, affected by chronic physical disease over time.

Using linked statewide hospital, emergency department and mortality data from Queensland spanning 24 years (2000-2023), we followed 30,189 people with schizophrenia-spectrum or bipolar-spectrum disorder and compared them with 148,476 age- and sex-matched hospital comparators without these diagnoses.

The findings were stark. People with severe mental illness had more than double the relative incidence of physical multimorbidity affecting at least two systems (adjusted hazard ratio 2.25), rising to more than three times the incidence for five or more systems (hazard ratio 3.16). By 20 years, they had accrued 10-19 additional cases per 100 people across thresholds. Ten of eleven systems showed significantly elevated incidence, with endocrine disease the largest increase (hazard ratio 3.52). Only cancer incidence did not differ significantly; this finding has been commented on elsewhere, as other researchers have highlighted that cancers are likely to be underscreened and underdiagnosed in this population.

The most alarming pattern concerned how these rates were elevated in younger populations. The gap was widest in our youngest cohort, under 25, and widest of all in young women, who had over nine times the incidence of accumulating five or more affected systems compared with matched peers (hazard ratio 9.24). Physical multimorbidity in severe mental illness appears to start taking hold relatively quickly in the years following mental illness onset, rather than developing decades later.

This matters because our screening tools and models of care are mostly built around older populations. Our data suggest that lens is too narrow. Early, proactive, whole-of-system physical health screening deserves a place alongside early psychosis intervention, particularly for young people.

As with any hospital-based registry study, there are limits: we could not see primary care diagnoses, and our comparators are likely sicker than the general community, so true relative risks against the general population may be even higher than reported. But with a 24-year follow-up rare in this field, I think these findings strengthen the case for integrated physical and mental health care, especially soon after diagnosis.

The paper is open access and available on The Lancet Psychiatry website. 

Dr Sean Halstead
Psychiatry Registrar, Metro South Addiction and Mental Health Service, PhD Candidate, Medical School, Faculty of Medicine and Behavioural Sciences, University of Queensland

_Reference:

_{Halstead, S., Warren, N., Kisely, S., Arnautovska, U., Trott, M., Correll, C. U., Ebdrup, B. H., Heffernan, E., Høgnason Mohr, G., McCutcheon, R. A., Morley, K., O’Donoghue, B., Pillinger, T., Sara, G., Taipale, H., & Siskind, D. (2026). Incidence of systemic physical multimorbidity among people with schizophrenia-spectrum or bipolar-spectrum disorders compared with age- and sex-matched comparators in Queensland, Australia: a statewide, hospital cohort study. The Lancet Psychiatry, 13(9), 771–782. https://doi.org/10.1016/S2215-0366(26)00196-3.}