AUTHOR=He Jiawei , Wang Weijie , Wang Shiwei , Guo Minhua , Song Zhenyan , Cheng Shaowu 

TITLE=Taking precautions in advance: a lower level of activities of daily living may be associated with a higher likelihood of memory-related diseases

JOURNAL=Frontiers in Public Health

VOLUME=Volume 11 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2023.1293134

DOI=10.3389/fpubh.2023.1293134

ISSN=2296-2565

ABSTRACT=Memory-related diseases (MDs) pose a global healthcare challenge, requiring early detection for effective intervention. This study explores the potential of Activities of Daily Living (ADL) as a clinical diagnostic indicator for MDs. Using data from the 2018 national baseline survey of the China Health and Retirement Longitudinal Study (CHARLS), involving 10,062 Chinese individuals aged 45 or older, we assessed ADL using the Barthel Index (BI) and correlated it with MDs. Statistical analysis, complemented by machine learning algorithms (SVM, DT, and LR), elucidated the relationship between ADL and MDs.

Background:
MDs are a public health concern, necessitating early intervention. Identifying clinical signs for MDs is imperative. ADL has potential as a marker. This study rigorously analyzes clinical data and validates machine learning algorithms for ADL as an MD indicator.

Methods:
Data from the 2018 CHARLS survey (10,062 individuals, aged 45+) were used. ADL was assessed using BI; MDs were determined through health reports. Statistical analysis used SPSS 25.0, and machine learning algorithms (SVM, DT, LR) in Python 3.10.2.

Results:
Population analysis showed the average BI for MD individuals (70.88) was lower than controls (87.77). Pearson's correlation revealed a negative association (r = -0.188, p < 0.001) between ADL and MDs. Adjusting for covariates, the negative relationship remained (B = -0.002, β = -0.142, t = -14.393, p = 0.000).

Machine learning confirmed ADL's predictive accuracy for MDs: SVM-AUC = 0.69, DT-AUC = 0.715, LR-AUC = 0.7. BI enhanced predictive abilities, with DT showing superior performance.

Conclusion:
This study establishes a negative correlation between ADL and MDs. Results validate ADL as an MD indicator, particularly when coupled with the Barthel Index. Lower ADL levels associate with a higher likelihood of memory-related diseases, emphasizing ADL's clinical relevance in early detection.