Background:Impairments in cognition and instrumental activities of daily living (iADL) increase the risk of developing dementia; therefore, it is vital to identify cognitive predictors of potential functional decline. It is unclear how aware older people and people with dementia (PwD) are of their iADL abilities, and there are concerns over the reliability of iADL ratings. This thesis will explore predictors of iADL and the accuracy of iADL ratings in healthy older people and PwD.Method:A meta-analysis synthesised results of studies investigating the association between executive function and activities of daily living in PwD. Four empirical studies built on the meta-analysis by investigating predictors of iADL in healthy older people and PwD. Rating accuracy was investigated by comparing self-ratings with informant ratingsand by comparing self-ratings and informant ratings with performance on an objective iADL measure.
Results:The Trail Making Test 4 predicted iADL in healthy older people whereas everyday memory predicted iADL in PwD. The Addenbrooke’s Cognitive Examination-Revised(ACE-R) predicted iADL in both healthy older people and PwD. Healthy older people underestimated their functional ability whereas their informants were more accurate. PwD were able to accurately rate their iADL ability whereas informant ratingsunderestimatedthe functional ability of PwD.Conclusions:The ACE-R may potentially help to identify older people at risk of developing functional dependence and to predict functional decline in PwD. Caution should be used when interpreting iADL ratingsmade by healthy older people as they may underestimate their ability. The finding that PwD were more able to accurately rate their iADL ability compared with their informants has important implications for the use of self-ratings and informant ratings of iADL in the dementia diagnostic process. The findings of the thesis have profound implications for the reliability of ratings of iADL made by healthy older people and PwD.