Neuropsychological and behavioral disturbance correlates of unawareness of memory impairment in dementia: a population-based study.

INTRODUCTION

This study investigated in a community sample associations of 2 different measures of unawareness of memory impairment in dementia with cognitive variables and behavioral and psychological symptoms of dementia.

METHOD

Design--cross-sectional, population-based survey. Settings--community samples (n = 15 022) from 3 world regions (Latin America, China, and India). Participants--829 people with dementia identified from standardized interviews and diagnostic algorithms. Measurements--unawareness of memory deficits was measured in 2 ways: comparison of participant subjective report with either objective performance on memory tests or informant report (IR). Associations were investigated using prevalence ratios and Poisson regressions. Differences in frequency of unawareness were explored with McNemar tests for each region and agreement between variables calculated with Cohen κ.

RESULTS

The unawareness variable based on comparison with performance was associated with cognitive variables, such as fluency (in China) and visuospatial impairments (Latin America and India), and behavioral symptoms, such as mania (Latin America) and hallucinations (China). The unawareness variable based on IR was associated only with behavioral symptoms, such as anxiety (in China and India) and hallucinations and mania (Latin America). Frequency of unawareness was significantly higher in the unawareness variable based on performance in India. Agreement between the 2 unawareness variables was moderate in Latin America and China but only slight in India.

CONCLUSION

Different ways of measuring unawareness are differentially associated with influencing factors. Informant-based unawareness measures may be more subjective and less sensitive than variables taking into account actual performance on cognitive tests.