• anxiety;
  • depression;
  • cognitive decline



The aim of the study was to empirically investigate the hypothesis that anxiety in the elderly, secondary to loss of memory, predicts future cognitive decline.


The participants were 137 elderly subjects with no depression or cognitive impairment from a community geriatric assessment unit, 45% with anxiety. In addition to demographic characteristics, cognitive status was assessed using the Mini Mental State Examination; depression was assessed by Tucker's short Interviewer-Assisted Depression Rating Scale; anxiety by Sinoff's Short Anxiety Screening Test and Activities of Daily Living function by Shah's modified Barthel's Index.


At follow-up 37 persons had dropped out, leaving 100 participants for final analysis. Mean re-examination time was 3.2 years with no group differences. The mean MMSE and modified Barthel scores decreased significantly more in those with anxiety. A relative risk of 3.96 for developing future cognitive impairment was found. Regression analysis showed that only anxiety was a significant predictor of cognitive decline. By path analysis, a more parsimonious model showed anxiety to have both a direct and an indirect effect on predicting future cognitive decline, and that the effect of loss of memory on cognitive decline was via anxiety.


Anxiety is inter-related and inseparable with loss of memory and its presence is a strong predictor for future cognitive decline, directly or indirectly via depression. It appears that loss of memory is the initial problem with consequent development of anxiety. Therefore, anxiety, like depression, is probably an early predictor of future cognitive decline and even possible future cognitive impairment. Copyright © 2003 John Wiley & Sons, Ltd.