A brief questionnaire of self‐reported cognitive decline predicts MMSE and study partner QDRS: findings from the Davis Memory and Aging Cohort at Mass General Brigham

Abstract

AbstractBackgroundA brief self‐reported questionnaire to efficiently capture subjective cognitive decline, a stage considered to be between cognitively unimpaired and mild cognitive impairment, could enhance clinical research and practice. We hypothesized that a 3‐question measure of self‐reported cognitive decline is associated with worse cognitive performance and increased functional changes reported by a study partner (SP).MethodThe Davis Memory and Aging Cohort (MAC) at Mass General Brigham recruits individuals 45 years or older with a broad range of cognitive symptoms. Participants answered the following three questions adapted from the NIA‐AA research framework: Q1. “Have you experienced a change in your memory in the last 1‐3 years?” Q2. “Has this been a persistent change over the last 6 months?” Q3. “Are you concerned about this change?” scored on a Likert scale (0‐4), with high scores representing greater decline. Responses to Q2 and Q3 were set to 0 if Q1 = 0. Participants completed the Mini‐Mental State Examination (MMSE; cognitive performance), and SPs completed the Quick Dementia Rating System (QDRS; functional changes). We tested the association between the total BCSS score (Q1+Q2+Q3; predictor) and MMSE or QDRS total scores (outcome) using linear regression controlling for age, sex, and years of education. We then tested these associations in a subgroup without significant SP concern (QDRS<1.5).ResultThere were 399 MAC participants with complete data (mean age = 67.2±8.3, 64% female, Table 1). Based on QDRS cut‐offs, there were 248 cognitively unimpaired participants, 118 with mild cognitive impairment, and 33 with dementia. The total score was associated with lower MMSE (beta = ‐0.30, p<0.001) and higher QDRS (beta = 0.51, p<0.001). Q1 score by itself was associated with lower MMSE (beta = 0.98, p<0.001) and higher QDRS (beta = 1.2, p<0.001) (Figure 1). In a subset with QDRS<1.5 (n = 248), both the total score (beta = ‐0.22, p = 0.002) and Q1 score (beta = ‐0.62, p<0.001) were associated with lower MMSE.ConclusionSelf‐reported cognitive decline was associated with worse global cognition, even among those whose SPs reported no significant functional concerns. Our results suggest that the brief self‐reported questionnaire could be a useful screening tool in research and clinic that efficiently probes cognitive functioning.