The association between subjective and objective cognitive functioning from a transdiagnostic perspective: An umbrella review and meta-analysis

Abstract

The relationship between subjective (self-reported) and objective (performance-based) cognitive functioning has significant clinical implications across neuropsychiatric syndromes. We performed an umbrella review examining literature on the association between subjective and objective cognition from a transdiagnostic perspective. Eligible studies were full reports of review papers examining the relationship between subjective and objective cognition in humans across the lifespan. Risk of bias was evaluated using a modified version of the AMSTAR rating scale. Of 7329 initial reports screened, 50 reviews of mostly cross-sectional data were included. Review size ranged from 4 to 488 studies and demographic factors aside from age were rarely reported. Risk of bias was mixed. A narrative synthesis across more than 20 neuropsychiatric and neuromedical conditions found an inconsistent relationship between subjective and objective cognition, with effect sizes ranging from null to weak. Mental health factors such as depression and anxiety were more reliably and strongly related to subjective cognition than were objective cognitive test scores. In a second-order meta-analysis (10 reviews; N = 92,606), the pooled correlation between overall objective and subjective cognition was 0.14 (95 % CI [0.08, 0.21]; p = 0.001, Q(9) = 179.50, τ = 0.09, I² = 95.00, <2 % variance shared). Subjective and objective cognition are largely non-overlapping constructs, reflecting unique and complementary aspects of overall cognitive functioning regardless of the specific population investigated or assessment method used. Researchers and clinicians should not expect to find concordance between concurrent self-reported cognition and neuropsychological testing, and simple discrepancies between subjective and objective cognition are so common as to be diagnostically unhelpful.