Excess COVID-19 deaths in severe mental illness persisted after vaccination began

Design

Retrospective whole-country linked cohort

Population

13,463,945 people in England with COVID-19, 160,190 with severe mental illness

Primary outcome

All-cause and COVID-19 mortality by pandemic stage

Effect

All-cause aHR 1.56 (1.53 to 1.59); 1.27 in stage 1 vs 1.55 after roll-out

This whole-country English cohort linked primary care records for everyone who had COVID-19 between January 2020 and June 2023: 13,463,945 people, of whom 160,190 (1.19%) had severe mental illness (schizophrenia, schizoaffective disorder, bipolar disorder or other psychotic illness).

People with severe mental illness had higher all-cause mortality after infection (adjusted HR 1.56, 95% CI 1.53 to 1.59). The gap was smaller in the first wave (aHR 1.27) and larger during and after vaccine roll-out (aHR 1.56 and 1.55). Vaccine uptake started higher in this group (73.2% vs 67.4% during roll-out) but by the end fewer were fully vaccinated (79.4% vs 87.2%). Excess COVID-19-specific mortality was partly attenuated by vaccination but persisted (aHR 1.25).

The finding cuts against the assumption that a well-run vaccine programme closes a health gap by itself. Early prioritisation worked; follow-through on later doses did not. For services, that points at boosters and routine immunisation as part of the physical-health review, not something left to general practice alone.

• Check vaccination status, including boosters, at every physical-health review for patients with psychosis or bipolar disorder.
• Offer or arrange vaccination on site where the service can, rather than relying on a separate primary care visit.
• Treat respiratory infection in this group as higher risk and lower the threshold for review.
• Flag patients who missed later doses; initial uptake in this group was high, but completion fell behind.

Why it matters

Prioritising people with severe mental illness for the first dose did not stop their mortality gap widening afterwards.

Don't overread it

This is observational; residual confounding by physical comorbidity and deprivation is likely.

The statistics, in plain English

An adjusted hazard ratio of 1.56 means deaths after infection occurred at about 1.5 times the rate in people with severe mental illness, after accounting for age, sex, ethnicity and other recorded factors. The intervals are narrow because the cohort is enormous — that makes the estimate precise, not necessarily free of bias.

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.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.