Development and validation of the Vanderbilt Informant-Based Subjective Cognitive Decline Questionnaire (Vanderbilt I-SCD)
Objectives: Subjective cognitive decline (SCD), a self-reported concern about one’s memory or thinking, is emerging as a potential early marker of unhealthy brain aging, recognized by the recent operationalized definition of SCD. However, SCD has limitations, such as overreporting concerns. Informants (loved ones) can provide valuable insight to mitigate these limitations, but existing informant-based SCD (I-SCD) tools were developed prior to the recent operationalized definition and therefore fail to capture all relevant SCD criteria. In this study, we developed an I-SCD questionnaire and validated the questionnaire by comparing it with measures of objective cognition and biomarkers of Alzheimer’s disease (AD). Methods: The I-SCD questionnaire was created using data from 537 informants (62±13 years, 69% female) of 458 cognitively unimpaired (CU; 68±8 years, 55% female) and 79 cognitively impaired older adults (CI; 75±9 years, 51% female) from the Vanderbilt Memory and Alzheimer’s Center Participant Registry. Informants all completed an I-SCD protocol including the Informant-Everyday Cognition scale and the Informant-Cognitive Changes Questionnaire. Latent variable modeling was used for item selection. Participants in the validation study were drawn from an independent cohort, the Vanderbilt Memory & Aging Project, including 176 CU (73±7 years, 41% female) and 132 CI older adults (CI; 73±8 years, 44% female) and their loved ones. All completed the new I-SCD questionnaire, a self-SCD questionnaire, objective cognitive assessment, and brain MRI with a subset undergoing fasting lumbar puncture for acquisition of AD biomarkers (Aβ42 and tau). Area under the receiver operating characteristic (AUROC) curve was used to measure the utility of the I-SCD tool for diagnostic discrimination. Regressions adjusting for age, sex, education, race/ethnicity, depressed mood, and apolipoprotein ε4 (APOE-4) status related I-SCD score to objective cognition, self-SCD, and AD biomarkers. Results: The Vanderbilt I-SCD Questionnaire (Vanderbilt I-SCD) includes 25 items and fulfills all operationalized SCD criteria. Within the validation cohort, the total score of the I-SCD significantly discriminated between CU and CI (AUC = 0.802, CI = 0.741-0.863). Linear regressions indicated increasing total I-SCD score was significantly associated with greater self-SCD, worse objective cognition in the areas of memory, executive functioning, and language, and increasing levels of amyloidosis and tau deposition (all p-values <0.05). Regression analyses stratified by cognitive diagnosis suggested these findings are driven by the CI group, whereas total I-SCD score appears related to self-SCD only in the CU group (OR= 0.59, P = 0.012). Conclusions: Results indicate that the Vanderbilt I-SCD discriminates between diagnostic statuses (CU vs. CI). Vanderbilt I-SCD score was also strongly associated with multiple markers of unhealthy brain aging and AD pathology. This finding was most notable among participants in the prodromal phase of AD (mild cognitive impairment), which aligns with previous research highlighting that I-SCD is most useful in this phase compared to preclinical AD. Overall, findings support the utility of the Vanderbilt I-SCD as a useful screening tool or as a tool to monitor cognitive status in patients with prodromal AD.