Traceability and comparability through crosswalks with the NeuroMET Memory Metric

Area of Science:

• Neuroscience
• Psychometrics
• Gerontology

Background:

• Accurate memory assessment is crucial for Alzheimer's disease (AD) diagnosis, progression monitoring, and therapy evaluation.
• Current neuropsychological tests lack standardization and metrological quality assurance.
• Developing improved memory metrics can enhance clinical utility and reduce patient burden.

Purpose of the Study:

• To link items from different legacy short-term memory tests using psychometric "crosswalks".
• To develop a composite metric, the NeuroMET Memory Metric (NMM), for standardized memory assessment.
• To create conversion tables for comparing NMM results with legacy tests and the Mini Mental State Examination (MMSE).

Main Methods:

• Data collected from European EMPIR NeuroMET and SmartAge studies (n=360).
• A 57-item bank (NMM) developed from legacy tests (Corsi Block, Digit Span, RAVLT, CERAD, MMSE).
• Rasch analysis (RUMM2030) used to create crosswalks and conversion tables.

Main Results:

• The NMM demonstrated direct comparability between different legacy memory tests.
• Measurement uncertainties for NMM estimates were smaller than individual legacy tests across the full memory span.
• Higher measurement uncertainties for NMM were observed in individuals with very low memory ability (raw score ≤ 19) compared to the MMSE.

Conclusions:

• The NMM provides a practical tool for clinicians and researchers to standardize memory measurements.
• Conversion tables enable traceability, reliable comparisons, and comparability between different legacy tests.
• The NMM enhances metrological quality assurance in memory assessment for individuals with cognitive decline.