Reducing misclassification of mild cognitive impairment based on base rate information from the Uniform data set

Area of Science:

• Neuroscience
• Psychometrics
• Gerontology

Background:

• Accurate diagnosis of mild cognitive impairment (MCI) is crucial for timely intervention and management.
• Traditional criteria for MCI can lead to misdiagnosis, impacting patient care and research.
• The Uniform Data Set Czech version (UDS-CZ 2.0) offers a standardized tool for cognitive assessment.

Purpose of the Study:

• To define and validate psychometric criteria for characterizing "possible" and "probable" cognitive deficits.
• To reduce the misdiagnosis rate of mild cognitive impairment (MCI) using the UDS-CZ 2.0.
• To develop a computational tool for improved MCI classification.

Main Methods:

• Computed the prevalence of low scores on 14 subtests of UDS-CZ 2.0 in healthy older adults.
• Validated criteria for possible and probable cognitive impairment in a sample of amnestic MCI patients.
• Assessed misclassification rates and false positive rates in clinical and healthy control samples.

Main Results:

• Psychometrically derived criteria demonstrated low misclassification rates for possible cognitive impairment (66-76% correct classification in MCI patients, 2-8% false positives in controls).
• Similar accuracy was observed for probable cognitive impairment classification.
• The UDS-CZ 2.0 psychometric approach proved effective in minimizing misdiagnosis compared to traditional MCI criteria.

Conclusions:

• The proposed psychometric approach using UDS-CZ 2.0 provides a reliable method for diagnosing cognitive deficits.
• This approach significantly reduces misdiagnosis rates in mild cognitive impairment.
• A computational tool based on these criteria can aid clinicians in accurate MCI assessment.