Re: Bell-McGinty et al. Standard measures of executive function in predicting instrumental activities of daily living in older adults

✍ Scribed by Pascale Barberger-Gateau; Colette Fabrigoule

• Publisher: John Wiley and Sons
• Year: 2003
• Tongue: English
• Weight: 32 KB
• Volume: 18
• Category: Article
• ISSN: 0885-6230
• DOI: 10.1002/gps.825

✦ Synopsis

Re: Bell-McGinty et al. Standard measures of executive function in predicting instrumental activities of daily living in older adults Dear Editor

We were very interested by the paper by Bell-McGinty et al. (2002) which shows a cross-sectional association between a lower performance on selected tests of executive function and IADL functioning. However, the authors are not completely right when they state that 'this study is the only one, to date, to examine the differential contribution of specific, commonly used neuropsychological tests of executive function in predicting functional ability in a geriatric population'.

Indeed, we conducted a survey in 1792 nondemented community dwellers aged 66 and over, included in the PAQUID (Personnes Age ´es QUID) study, to examine the relationships between performance on seven commonly used neuropsychological tests and selfreported performance on four selected Instrumental Activities of Daily Living (IADL) (Barberger-Gateau et al., 1999b). These four IADLs were selected because they were highly predictive of the onset of a dementia in the three to five subsequent years (Barberger-Gateau et al., 1999a). Although these neuropsychological tests were not specific of executive functions, as most tests assess in fact several areas of cognitive functioning, most of them loaded on the first factor of a Principal Component Analysis which was interpreted as the most controlled aspect of the tests in a previous research (Fabrigoule et al., 1998). The four selected IADLs also loaded on this factor, indicating that impairment in these four activities was strongly associated with an impairment in controlled processes. Our analyses were adjusted for several confounders, including age, education, and visual and hearing impairments (Barberger-Gateau et al., 1999b). These results show that the early IADL limitations observed at the preclinical stage of dementia could be explained