Mortality in first-time psychosis: 25-year study sheds light on trajectories & at-risk groups

In patients with first-episode psychosis, the cumulative incidence of unnatural deaths (mostly suicide) rises sharply during the first 10 years, while deaths from natural causes begin to increase after the first decade, a 25-year population-based cohort study in Hong Kong has shown.

“Patients with first-episode psychosis face markedly increased risk of excess mortality, yet the long-term trends and key contributing factors remain insufficiently characterized,” wrote the researchers from the Department of Psychiatry at the University of Hong Kong (HKU).

Mortality patterns over 25 years

In the current population-based cohort study, the researchers followed 1,389 patients with first-episode psychosis for 25 years. A total of 137 deaths occurred: 102 from unnatural causes and 35 from natural causes (documented medical conditions). [Psychological Medicine 2025;55:e355]

The overall standardized mortality ratio (SMR) was 6.56. The highest all-cause SMR was 11.82 in the first 5 years, and the lowest all-cause SMR was 3.33 in the 21st–25th years.

Of note, the cumulative incidence of unnatural deaths rose sharply during the first decade and stabilized in the second, while natural deaths increased after the first decade.

Suicide was the leading cause of unnatural death (95 percent), while cardiovascular conditions were the leading cause of natural death (26 percent).

Who’s at higher risk of death?

Multivariate Cox regression analysis showed that male gender (hazard ratio [HR], 1.72; 95 percent confidence interval [CI], 1.20–2.45; p<0.01), low educational level (HR, 1.73; 95 percent CI, 1.07–2.79; p=0.03), poor social functioning (low Social and Occupational Functioning Assessment Scale [SOFAS] score in month 1; HR, 3.26; 95 percent CI, 1.29–8.23; p=0.01), and hospitalization in month 1 (HR, 1.66; 95 percent CI, 1.03–2.67; p=0.04) were significant risk factors associated with all-cause mortality in patients with first-episode psychosis. These factors were also associated with a higher risk of unnatural death (p<0.05 for all).

“In fact, male gender, low educational attainment, and poorer baseline social functioning have generally been considered as risk factors for poorer outcomes in people with psychosis,” explained the researchers. “Hospitalization at baseline likely reflects higher illness severity at initial presentation.”

Notably, large fluctuation in antipsychotic treatment during the first 10 years was associated with a higher risk of all-cause mortality (HR, 1.01; 95 percent CI, 1.00–1.03; p=0.01). “Fluctuations in antipsychotic use exacerbate cardiovascular risks — such as metabolic syndrome, QT-interval abnormalities, and diabetes-related complications — which likely contribute to increased natural-cause mortality,” explained the researchers. “These results highlight the importance of consistent antipsychotic use and thus illness stability in relation to long-term mortality.”

“The 25-year follow-up study of first-episode psychosis highlighted changes in long-term mortality pattern and thus the phase-specific needs of patients with first-episode psychosis,” commented the researchers. “Therefore, it is important to integrate physical care into mental health services, as well as provide stage-specific and individualized care for patients with first-episode psychosis to reduce long-term excess mortality.”