Objective: To assess cognitive performance as a predictor of noncompletion of cardiac rehabilitation (CR) using a standardized verbal memory test.
Methods: This was a prospective cohort study of consecutive patients with coronary artery disease (n = 131) entering 1-year outpatient CR between April 2007 and May 2009. Verbal memory performance was assessed using the California Verbal Learning Test, Second Edition. Attendance at weekly CR sessions was recorded, and completion or noncompletion was determined according to comprehensive CR criteria. Depression was diagnosed according to DSM-IV criteria as a possible confounder.
Results: Verbal memory performance at entry into CR differed significantly (F(1,130) = 7.80, p = .006) between noncompleters and completers (mean [SD] cumulative California Verbal Learning Test, Second Edition, score, -1.15 [2.59] versus 0.47 [3.12]) in analysis of covariance controlling for pertinent clinical confounders. Better verbal memory performance predicted a reduced risk of noncompletion (hazard ratio [HR] = 0.86, 95% confidence interval [CI] = 0.77-0.96, p = .009) in time-to-event analysis adjusted for depression (HR = 2.62, 95% CI = 1.33-5.17, p = .006) and smoking history (HR = 2.03, 95% CI = 0.98-4.22, p = .06). A post hoc analysis suggested that better verbal memory performance predicted a reduced risk of noncompletion for medical reasons (HR = 0.83, 95% CI = 0.70-0.99, p = .03).
Conclusions: Poorer verbal memory performance was associated with an increased risk of noncompletion of CR among participants with coronary artery disease. Further studies exploring practical methods for screening and targeted support might improve rehabilitation outcomes.