Robert G. Robinson, Ricardo E. Jorge, Jeffrey Long

This study re-examined patients from a 1-year randomized controlled double-blind trial of escitalopram, problem-solving therapy (PST), or placebo to prevent depression among patients less than 3 months after a stroke. The objective of the current study was to determine if preventive treatment would predict time to death over 8-10 years of follow-up. Based on the importance of depression in poststoke mortality and a previous study of this population at 18 months' follow-up showing that stopping escitalopram but not PST led to a significant increase in depression, the authors hypothesized that PST would be associated with the longest time to death. Of 129 eligible patients, 122 were contacted and 7 were lost to follow-up. Families and surviving patients were interviewed to determine current health status or the date and cause of death. Using the Weibull model of log time (years) to death, controlled for age, severity of physical illness, gender, severity of stroke, and history of depression after study entry, there was a significant independent effect of treatment. PST significantly and independently increased the time to mortality, whereas older age and major depression significantly and independently decreased the time to death. To the authors' knowledge, this is the first time a psychological antidepressant treatment administered for 1 year has been associated with increased survival among patients who have suffered a stroke.