Differential functioning of mini-mental test items according to disease.

Area of Science:

• Psychometrics
• Neuropsychological Assessment

Background:

• Measurement invariance is crucial for comparing cognitive levels across diverse patient groups.
• Differential Item Functioning (DIF) indicates a test has different measurement properties for different populations.
• Lack of DIF is a prerequisite for valid cross-group comparisons.

Purpose of the Study:

• To assess the measurement invariance of the Mini-Mental State Examination (MMSE) across different patient groups.
• To determine if the MMSE exhibits DIF in patients with Parkinson's disease, Alzheimer's-type dementia, and normal subjects.

Main Methods:

• The study employed the Rasch Model, an advanced psychometric technique, to analyze MMSE data.
• A total of 400 subjects were included: Parkinson's disease patients, Alzheimer's patients, and normal controls.
• Differential Item Functioning (DIF) analysis was performed to detect item bias.

Main Results:

• Significant differences were found between the groups.
• The MMSE demonstrated DIF for Parkinson's disease patients, suggesting an underestimation of their cognitive function.
• Eleven out of 30 MMSE items exhibited DIF, challenging measurement invariance.

Conclusions:

• The MMSE is not measurement invariant across the studied groups.
• Score comparisons between Parkinson's disease patients, Alzheimer's patients, and normal subjects using the MMSE are not statistically justified.
• The findings highlight the need for caution when interpreting MMSE scores in diverse clinical populations.