That people with schizophrenia and bipolar disorder die earlier of physical illness is not news. What has been missing is a description of how that burden actually accumulates — whether it is one bad condition or many, and when it starts. This Queensland cohort used statewide hospital registries from 2000 to 2023, matching 30,189 people with schizophrenia-spectrum or bipolar disorder to 148,476 hospital comparators by age and sex at 1:4. Median age at entry was 32. The outcome was systemic physical multimorbidity — the number of organ systems affected by chronic physical disease.
The hazard ratios climb with each system added: 1.85 (95% CI 1.81 to 1.88) for at least one affected system, 2.25 for two, 2.65 for three, 2.93 for four, and 3.16 (3.01 to 3.30) for five or more. That gradient is the finding. It is not that these patients get one extra illness; it is that the more systems you count, the wider the gap becomes.
The absolute numbers make it concrete. At 20 years there were 18.43 additional cases per 100 people with disease in at least two systems, and 11.66 additional per 100 with five or more. Roughly one in five people with severe mental illness acquires multi-system chronic disease that a matched person without it would not have.
And it starts young. Age-stratified analysis showed the largest relative differences in the youngest groups, peaking at an adjusted hazard ratio of 9.24 (7.30 to 11.71) in women under 25 with at least five affected systems. A woman in her early twenties with a first psychotic episode is not decades away from her physical health problem; she is at the point where the divergence begins.
The implication for practice is about when, not whether. Physical health monitoring in severe mental illness is typically framed as something that becomes relevant with age and with cumulative antipsychotic exposure. This says the divergence is steepest in people under 25, which is exactly when services are focused on the psychosis and least focused on the metabolic panel. In Indian practice, where early intervention services are scarce and physical health monitoring often falls between the psychiatrist and a general practitioner the patient does not have, that argues for the psychiatric service owning it rather than referring it.
- Start physical health monitoring at first presentation, not after years of treatment.
- Prioritise patients under 25, where the relative divergence from the general population is steepest.
- Count systems, not conditions — the gap widens with each additional organ system involved.
- Assume the psychiatric service owns physical monitoring unless a named other clinician has accepted it.
- Expect roughly one in five to acquire multi-system disease attributable to the excess risk over 20 years.
The statistics, in plain English
Subdistribution hazard ratios are used here because death competes with the outcome: a person who dies cannot go on to accumulate another affected system, and ignoring that would understate the burden. The Fine-Gray method accounts for it. Two limits are worth holding. The comparison group was other hospital patients, not the general population — that controls for healthcare contact but means the comparators are not healthy people, so the true gap versus the general population is probably larger. And ascertainment depends on hospital coding, so people with severe mental illness, who are admitted more often, have more opportunity for a physical diagnosis to be recorded.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for psychiatry, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free