Abstract
BACKGROUND: Limited research has explored the preceding sequences of medical conditions leading to mild cognitive impairment (MCI), particularly patterns that may signal progression to dementia or reversion to normal cognition. OBJECTIVES: Our study aims to analyze common sequences of chronic conditions preceding an MCI diagnosis and examine differences between retention in MCI or progression towards dementia and reversion to normal cognition, using hypergraph clustering, a network analysis approach. METHODS: We categorize participants into two groups (i) M2P: stay or progressed to dementia, or (ii) M2N: reversion to normal within 5 years after the first onset of MCI. Among 414 participants, 210 are males and the mean age is 80.8 years. We performed network analysis to obtain categories and sequences of chronic conditions. We applied hypergraph spectral clustering to characterize participants with similar sequences. RESULTS RESULTS: We identify and validate the generic key indicators (e.g., chronic kidney disease), highlight the sex-specific potential indicators (e.g., arthritis) for MCI reversal, and open new research directions for identifying potential disparities among men and women. CONCLUSION: We categorized the chronic conditions preceding MCI diagnosis and discovered unique sequences suggesting MCI reversion among men and women to facilitate future research.