Mayo normative studies: A conditional normative model for longitudinal change on the Auditory Verbal Learning Test and preliminary validation in preclinical Alzheimer's disease

Abstract

Introduction: The aim of this study was to develop a conditional normative model for Rey's Auditory Verbal Learning Test (AVLT) that accounts for practice effects.

Methods: In our normative sample, robust conditional norms were derived from 1001 cognitively unimpaired (CU) adults ages 50 to 89 who completed the AVLT up to eight times. Linear mixed-effects models adjusted for baseline performance, prior test exposures, time, demographics, and interaction terms. In our preliminary validation, mean performance on conditional and typical normative scores across two to four completed follow-up tests in preclinical Alzheimer's disease participants at baseline with positive amyloid and tau positron emission (n = 27 CU amyloid [A]+tau[T]+) was compared to biomarker negative individuals (n = 269 CU A-T-).

Results: AVLT performance using typical norms did not differ across A+T+ and A-T- groups. Conditional norms z-scores were lower in the A+T+ relative to the A-T- group for 30-minute recall (P = .033) and sum of trials (P = .030).

Discussion: Conditional normative methods that account for practice effects show promise for identifying longitudinal cognitive decline.

Keywords: Alzheimer's disease; Rey Auditory Verbal Learning Test; amyloid; biomarker; memory; mild cognitive impairment; neuropsychology; practice effects; reliable change index (RCI); robust normative data; standardized regression‐based change scores (SRB); tau; transitional cognitive decline; validity.

FIGURE 1

Flowchart showing the overall study design and methods. Part 1 (top row, red boxes) addresses study aim 1 to develop robust conditional normative data; this is our training sample. Part 2 (bottom row, blue boxes) addresses study aim 2 to assess clinical utility for preclinical Alzheimer's disease; this is our validation sample. A, amyloid; AD, Alzheimer's disease; AVLT, Auditory Verbal Learning Test; CU, cognitively unimpaired; LMM, linear mixed model; MOANS, Mayo's Older Americans Normative Studies; T, tau

FIGURE 2

Robust normative sample Auditory Verbal Learning Test (AVLT) raw scores. Raw score estimates over test numbers 2 through 8 are displayed for males and females for the AVLT delayed recall and sum of trials. Test number 1 (baseline) is not depicted as it is one of the model predictors and not a model output. A, Performance trajectories at the 25th percentile, (B) at the 50th percentile, and (C) at the 75th percentile

FIGURE 3

Comparative performance trajectories in validation sample participants for 30‐minute recall. Example of performance trajectories across four tests for AVLT 30‐minute delayed recall. A,B, Performance trajectories for the entire validation sample depicted by biomarker group using the traditional cross‐sectional MOANS norms applied at each follow‐up (A) and the conditional normative model at second through fourth follow‐up (B). C,D, Two individual participants, with corresponding raw and numeric normative scores reported in (E). The first participant is a 64‐year‐old female with 16 years of education who is A–T– and remains CU across all follow‐up tests (orange solid line). Conditional norms suggest this individual demonstrated a higher than typical improvement in performance at test number 2 and the trajectory remained above average at test numbers 3 and 4. The second participant is a 65‐year‐old female with 14 years of education who is A+T+ and is CU at tests 1 and 2, but is diagnosed with MCI at tests 3 and 4 (blue dashed line) per consensus conference. Despite a subtle improvement in raw score at test number 2, the conditional norm suggests that this individual's ability to benefit from practice was subtly low relative to similar peers. At test number 3, despite an identical raw score as test number 1, the conditional norms indicate this individual's trajectory over time is deviating from what is typical and this corresponds with the consensus diagnosis of MCI at this test number. By test number 4, performance is clearly abnormal per conditional norms that consider the number of test exposures (failure to benefit from practice), baseline performance, age, education, sex, and time since baseline despite a MOANS score that is only subtly low and in a range many label as within normal limits (equivalent to a z of –1) and a raw score only 1 point lower than the baseline test number. A, amyloid; AVLT, Auditory Verbal Learning Test; CU, cognitively unimpaired; MCI, mild cognitive impairment; MOANS, Mayo's Older Americans Normative Studies; T, tau