Longitudinal Associations of Mental Disorders with Physical Diseases and Mortality among 2.3 Million New Zealand Citizens

Abstract

Importance: Excess risk of physical disease and mortality has been observed among individuals with psychiatric conditions, suggesting that ameliorating mental disorders might also be associated with ameliorating the later onset of physical disability and early mortality. However, the temporal association between mental disorders and physical diseases remains unclear, as many studies have relied on retrospective recall, used cross-sectional designs or prospective designs with limited follow-up periods, or given inadequate consideration to preexisting physical illnesses. Objective: To examine whether mental disorders are associated with subsequent physical diseases and mortality across 3 decades of observation. Design, Setting, and Participants: This population-based cohort study used data from the New Zealand Integrated Data Infrastructure, a collection of nationwide administrative data sources linked at the individual level, to identify mental disorders, physical diseases, and deaths recorded between July 1, 1988, and June 30, 2018, in the population of New Zealand. All individuals born in New Zealand between January 1, 1928, and December 31, 1978, who resided in the country at any time during the 30-year observation period were included in the analysis. Data were analyzed from July 2019 to November 2020. Exposures: Nationwide administrative records of mental disorder diagnoses made in public hospitals. Main Outcomes and Measures: Chronic physical disease diagnoses made in public hospitals, deaths, and health care use. Results: The study population comprised 2349897 individuals (1191981 men [50.7%]; age range at baseline, 10-60 years). Individuals with a mental disorder developed subsequent physical diseases at younger ages (hazard ratio [HR], 2.33; 95% CI, 2.30-2.36) and died at younger ages (HR, 3.80; 95% CI, 3.72-3.89) than those without a mental disorder. These associations remained across sex and age and after accounting for preexisting physical diseases. Associations were observed across different types of mental disorders and self-harm behavior (relative risks, 1.78-2.43; P