Concordance of Mini-Mental State Examination, Montreal Cognitive Assessment and Parkinson Neuropsychometric Dementia Assessment in the classification of cognitive performance in Parkinson's disease

Background: Cognitive impairment (CI) is frequently observed in Parkinson's disease (PD) and negatively influences the patient's and carer's quality of life. As a first step, assessment of CI is often accomplished by using screening instruments (level I diagnosis). Three commonly used instruments are the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), and Parkinson Neuropsychometric Dementia Assessment (PANDA). Because different preferences regarding test selection exist between clinics, this study aims to provide evidence about the concordance of these tests. It also converts total test scores of the three instruments to assist clinical practice. Methods: Between January and December 2018, 96 patients with idiopathic PD were examined at the University Hospital of Cologne, Germany. Comparability of MMSE, MoCA, and PANDA scores was investigated by calculating correlations, classification agreements, and percentile ranks. Additionally, we converted test scores among the three screening instruments by implementing the equipercentile equating method and log-linear smoothing. Results: The MMSE classified 26%, the PANDA 32.3% and the MoCA 54.2% of PD patients as having CI. The screening instruments' concordance in classifying cognition into normal cognition versus CI was 75% (AC1 = 0.62) for MMSE and PANDA, 63.5% (AC1 = 0.28) for MoCA and PANDA, and 57.3% (AC1 = 0.24) for MMSE and MoCA. The provided conversion table enables a quick and easy transformation of the three screening instruments within PD diagnostics. Conclusion: These results contribute to a better understanding of the screenings' utility and concordance in a population of PD patients. Additionally, communication between clinics may be enhanced.