Mortality among first-admission patients with psychosis

Abstract 

Objective

This study compared 5- and 10-year survival and absolute and relative mortality rates among first-admission patients with 1 of 4 psychotic disorders: schizophrenia/schizoaffective disorder, bipolar disorder, major depression, and other nonorganic psychoses.

Method

The authors conducted a prospective 10-year follow-up of subjects first admitted with a diagnosis of nonorganic psychosis to any of 12 hospitals in Suffolk County, New York, during the period 1989 to 1995. Information on their death status since study entry was ascertained from the Social Security Death Index and the National Death Index. Survival analyses were conducted using the Kaplan-Meier product-limit estimator.

Results

There were no significant differences in survival rates among the 4 diagnostic groups at 5-year (range, 96.3%-97.8%) or 10-year (range, 90.2%-97.8%) follow-up. Absolute mortality over the study period ranged from 2.8% of bipolars to 6.7% of those with major depression. About 60% of deaths among schizophrenic/schizoaffective subjects were due to unnatural causes, whereas for the other 3 groups, deaths were more evenly split between natural and unnatural causes. Suicides comprised most deaths from unnatural causes, most of which occurred during the 2- to 5-year follow-up period. Deaths due to natural causes tended to be related to lifestyle factors.

Conclusions

First-admission patients with psychosis experience similar patterns of mortality risk over the first 10 years after index admission regardless of underlying diagnosis. Causes of death (both natural and unnatural) were potentially preventable with more intensive medical and psychiatric follow-up and intervention. Many deaths from unnatural causes occurred during or shortly after discharge from an inpatient or residential treatment setting, highlighting this period as one needing close scrutiny by treating clinicians.