Premium
Modifiable risk factor possession patterns of dementia in elderly with mild cognitive impairment: A 4‐year longitudinal study
Author(s) -
Katayama Osamu,
Lee Sangyoon,
Bae Seongryu,
Makino Keitaro,
Shinkai Yohei,
Chiba Ippei,
Harada Kenji,
Shimada Hiroyuki
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.040736
Subject(s) - dementia , gerontology , medicine , logistic regression , longitudinal study , psychosocial , risk factor , latent class model , cognition , demography , psychology , psychiatry , statistics , mathematics , disease , pathology , sociology
Background We clarified modifiable risk factor patterns of dementia in elderly patients with mild cognitive impairment (MCI) and determined the relationship between retention patterns and reversion from MCI over a 4‐year follow‐up. Method Participants were 5104 older adults who completed the National Center for Geriatrics and Gerontology‐Study of Geriatric Syndromes (NCGG‐SGS) examinations in August 2011–February 2012; 3095 (60.6%) completed the second‐wave cognitive examination in August 2015–August 2016. Among the longitudinal data, missing values were imputed for 789 community‐dwelling people who were aged ≥65 years (mean age, 72.0±5.4 years; 368 men and 421 women) with MCI and classified them as “reverters” and “nonreverters.” Longitudinal analysis was performed after missing values resulting from dropout were imputed. Nine modifiable risk factors at baseline—namely, less education, hearing loss, hypertension, obesity, smoking, depression, physical inactivity, social isolation, and diabetes—were classified in latent class analysis. In subsequent binomial logistic regression analysis, reversion status was the dependent variable (the nonreverters served as a reference group) and cluster membership the independent variable. Result Of the 461 participants for whom data were not imputed, 234 (50.8%) reverted from MCI to normal cognitive status. Reversion rate among the 789 participants for whom we imputed data was 30.9%. The modifiable risk factors among participants with MCI were classified into five patterns: low‐risk, psychosocial, health behavior, educational, and smoking factors. According to logistic regression analysis, people with low‐risk factors were more likely to revert from MCI to normal cognitive status than were patients with the other factors ( P <.05). Conclusion We formulated five patterns of modifiable risk factors among elderly people with MCI. Participants with low‐risk factors were more likely to recover from MCI than participants with the other factors. These results may provide useful information for designing interventions to prevent cognitive decline and dementia in individuals with MCI.