Impaired Verbal Memory is a Significant Predictor of Early Cerebral-Cardiovascular Death, an 18-Year Follow-Up of a National Cohort

Background The mortality pattern of individuals with impaired verbal memory (IVM) has not yet been well described. We sought to describe the risk of all-causes, as well as specific causes of death associated with IVM. Method We used the data of 4151 nationally representative adults ≥60 years old who participated in the third National Health and Nutrition Examination Survey, 1988-1994, and completed one non-contextual (i.e., word list memory) and one contextual delayed-recall tests (i.e., short story recall). The participants were passively followed up through 31 December 2006. We determined the hazard ratio of death from all-causes and specific cause through Cox proportional hazard regression. Results Severe and moderate IVM were present in 268 (6.5%) and 495 (11.9%) participants at baseline survey, and 2550 deaths occurred by the end of 18-year follow-up (median = 12 years). The medians of survival time adjusted for all-causes death were 6.17(95%CI: 5.50, 6.92), 9.50 (8.92, 10.25), and 13.17 (12.75, 13.58) years, respectively for the individuals with severe, moderate, and no IVM. Severe IVM was significantly associated with death from cardio-cerebral vascular diseases [hazard ratio = 1.70, 95%CI = (1.36-2.12)], stroke [2.60 (1.69-3.99)], and Alzheimer's disease [3.50 (1.40-8.76)]. The shortened survival time of the participants with IVM was mainly driven by the deaths of cerebral-cardiovascular diseases, which accounted for almost half of all deaths. Conclusion The predictability of memory scores to early cerebral-cardiovascular deaths demonstrated that central challenge among individuals with cognitive impairment was cardiovascular diseases management.