Robust normative standards for the California Verbal Learning Test (CVLT) ages 60–89: A tool for early detection of memory impairment

Abstract

Objective: To detect cognitive “impairment,” neuropsychologists rely on normative data to compare patient performance to “normal” peers. However, the true normality of normative samples may be called into question given the high prevalence of preclinical proteinopathies amongst clinically normal older adults. Given its common use in memory clinics, we aimed to develop a robust California Verbal Learning Test (CVLT) normative standard reflecting only the most cognitively stable sample of older adults available.

Method: Two hundred and twenty-eight older adults (mean age = 69.9, range = 60–89, 91% White, mean education = 17.6 years) who were clinically normal at baseline and demonstrated clinical stability on longitudinal assessment completed the CVLT at baseline. We applied a standardized algorithm to convert raw scores into normalized scaled scores and then regressed on age, sex, and education using fractional polynomial modeling.

Results: There were significant main effects of age and sex across CVLT metrics, but not education. Means and standard deviations were higher and less variable in our robust normative data than the data used to create the CVLT-II and CVLT-3 normative standards.

Conclusions: These norms set a higher standard for what should be considered “normal” in the spectrum of age-related memory changes and may help clinicians identify patients with memory and potential neurodegenerative changes in the earliest stages, further optimizing clinical management and clinical trial stratification. As with any standard, these robust norms are only appropriately utilized with patients that closely match the demographic profile of the individuals represented in the sample used for this study.

Keywords:

• Norms
• older adults
• demographics
• cognitive aging
• neuropsychology

Acknowledgments

This study was supported by NIH-NIA grants K23AG058752 (PI: Casaletto), L30AG057123 (PI: Casaletto),1R01AG032289 (PI: Joel Kramer), and R01AG048234 (PI: Joel Kramer). Our work was also supported by Larry L. Hillblom Network Grant (PI: Joel Kramer; 2014-A-004-NET) and Fellowship Grant (PI: Casaletto; 2017-A-004-FEL). We also thank Abraham Sprague for his help in creating the normative tables.

Disclosure statement

No potential conflict of interest was reported by the authors.