Background: The prevalence of depression in acutely medically ill elderly inpatients is high. Depression in the elderly is associated with increased mortality.
Method: The association between mortality at 18-month follow-up and depression, depressive symptoms and demographic variables at the outset in a cohort of acutely medically ill elderly inpatients was examined.
Results: The mortality at 18-month follow-up was 47%. Depression, depressive symptoms and demographic variables were not associated with mortality.
Conclusion: An important explanation of this absence of association between mortality and depression may be an artifact due to patient selection designed to resemble normal clinical practice.