Self- versus informant-report of cognitive decline in mild cognitive impairment: Concordance with cognitive and functional performance.

Debate continues regarding the use of self- versus informant-report to diagnose mild cognitive impairment (MCI) with studies reporting patients both overestimating and underestimating their abilities relative to informants. We assessed concordance of self- versus informant-report of cognitive decline with objective cognitive and functional performance in the participants of the preventing Alzheimer's dementia with cognitive remediation plus transcranial direct current stimulation in mild cognitive impairment and depression randomized controlled trial (PACt-MD).Three hundred six participants with MCI, and their informants, reported on cognitive decline; the participants also completed a comprehensive assessment of objective cognitive and functional performance. Based on the discrepancy between self- versus informant-report of cognitive decline, we grouped participants into categories of underestimators, congruent estimators, and overestimators.Informant- but not self-reported cognitive decline significantly correlated with objective cognitive performance. There were 68 underestimators, 94 congruent estimators, and 144 overestimators. Underestimators had significantly lower objective cognitive performance and functional performance than congruent estimators and overestimators. Cognitive performance significantly predicted functional performance in all three groups, and the relationship between cognitive and functional performance was moderated by the discrepancy between self- and informant-report.We showed a poor concordance among self-report of cognitive decline and both informant-report of cognitive decline and cognitive performance in patients with MCI. Our findings highlight clinical and research value in the assessment and consideration of degree of discrepancy between self- and informant-reports of cognitive decline in MCI. (PsycInfo Database Record (c) 2023 APA, all rights reserved).