OBJECTIVE

It is uncertain if the raised mortality in schizophrenia persists in later life. Register-based studies suggest that excess mortality continues, although at a lower level than in younger age groups. However, prospective follow-up studies of older schizophrenia samples are lacking.

METHODS

A cohort of 157 older patients (mean age at study entry: 68 years) diagnosed with schizophrenia or schizoaffective disorder in a psychiatric catchment area in Amsterdam, the Netherlands was studied. Standardized mortality rate (SMR) was estimated at a 5-year follow-up, in referral to the same age group in the general catchment area population. The impact on survival time of a range of independent demographic and clinical predictors was evaluated.

RESULTS

The cohort had an all-cause SMR of 1.89 (95% CI: 1.28-2.70). SMR was higher in men (2.60; 95% CI: 1.42-4.37) than in women (1.78; 95% CI: 1.02-2.90). All deaths were from natural causes. Reduced survival was associated with higher age (HR: 1.10; 95% CI: 1.05-1.16), male gender (HR: 3.94; 95% CI: 1.87-8.31), and having had one or more compulsory admissions in the past (HR: 2.61; 95% CI: 1.46-4.68). In contrast, no mortality associations were found with diagnosis (schizophrenia versus schizoaffective disorder), age at onset of the disorder, or current prescription of antipsychotics.

CONCLUSION

The excess mortality in schizophrenia continues into late life, affecting men more often than women. Given the poor insight into the underlying mechanisms of this disquieting finding, there is a need to identify modifiable clinical and social risk factors.