Risk Factors for Natural Cause Mortality in Schizophrenia

JAMA Netw Open. 2024 Sep 3;7(9):e2432401. doi: 10.1001/jamanetworkopen.2024.32401.

Abstract

Importance: Schizophrenia is associated with premature mortality from mostly natural causes. Decreased cognitive functioning has been identified as a determinant of mortality in the general population. However, there have been few prospective studies of this issue in persons with schizophrenia.

Objective: To examine whether lower cognitive functioning is a risk factor for natural cause mortality in schizophrenia.

Design, setting, and participants: This prospective cohort study included persons with schizophrenia or schizoaffective disorder enrolled between February 1, 1999, and December 31, 2022, at a nonprofit psychiatric system in Baltimore, Maryland. Participants were evaluated using the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) and other clinical measures.

Exposure: Natural cause mortality.

Main outcomes and measures: Associations of cognitive function, obesity, tobacco smoking, and medical conditions with natural cause mortality were evaluated using Cox proportional hazards regression models.

Results: Of the 844 participants enrolled (mean [SD] age, 39.6 [12.1] years; 533 male [63.2%]), 158 (18.7%) died of natural causes during a median follow-up of 14.4 years (range, 7.0 days to 23.9 years). The most significant factor associated with mortality was lower cognitive functioning as measured by the RBANS (Cox coefficient, -0.04; 95% CI, -0.05 to -0.03; z = -5.72; adjusted P < .001). Additional factors independently associated with mortality included the diagnosis of an autoimmune disorder (hazard ratio [HR], 2.86; 95% CI, 1.83-4.47; z = 4.62; adjusted P < .001), tobacco smoking (HR, 2.26; 95% CI, 1.55-3.30; z = 4.23; adjusted P < .001), diagnosis of chronic obstructive pulmonary disease (HR, 3.31; 95% CI, 1.69-6.49; z = 3.48; adjusted P = .006), body mass index as a continuous variable (HR, 1.06; 95% CI, 1.02-1.09; z = 3.30; adjusted P = .01), diagnosis of a cardiac rhythm disorder (HR, 2.56; 95% CI, 1.40-4.69; z = 3.06; adjusted P = .02), and being divorced or separated (HR, 1.80; 95% CI, 1.22-2.65; z = 2.97; adjusted P = .02). An RBANS score below the 50th percentile displayed a joint association with being a smoker, having an elevated body mass index, and having a diagnosis of an autoimmune or a cardiac rhythm disorder.

Conclusions and relevance: In this prospective cohort study, lower cognitive functioning was a risk factor for natural cause mortality in schizophrenia. Efforts should be directed at methods to improve cognitive functioning, particularly among individuals with additional risk factors.

MeSH terms

  • Adult
  • Baltimore / epidemiology
  • Cause of Death
  • Cognitive Dysfunction / epidemiology
  • Cognitive Dysfunction / mortality
  • Female
  • Humans
  • Male
  • Middle Aged
  • Neuropsychological Tests / statistics & numerical data
  • Proportional Hazards Models
  • Prospective Studies
  • Psychotic Disorders / complications
  • Psychotic Disorders / epidemiology
  • Psychotic Disorders / mortality
  • Risk Factors
  • Schizophrenia* / complications
  • Schizophrenia* / epidemiology
  • Schizophrenia* / mortality