Objectives: This study aimed to identify factors associated with life satisfaction by developing machine learning (ML) models to predict life satisfaction in older adults living alone. Methods: Data were extracted from 3112 older adults participating in the 2020 Korea Senior Survey. We employed 5 ML models to classify the life satisfaction of older adults living alone: logistic Lasso regression, decision tree-based classification and regression tree (CART), C5.0, random forest, and extreme gradient boost (XGBoost). The variables used as predictors included demographics, health status, functional abilities, environmental factors, and activity participation. The performance of these ML models was evaluated based on accuracy, precision, recall, F1-score, and area under the curve (AUC). Additionally, we assessed the significance of variable importance as indicated by the final classification models. Results: Out of the 1411 older adults living alone, 45.3% expressed satisfaction with their lives. The XGBoost model surpassed the performance of other models, achieving an F1-score of 0.72 and an AUC of 0.75. According to the XGBoost model, the five most important variables influencing life satisfaction were overall community satisfaction, self-rated health, opportunities to interact with neighbors, proximity to a child, and satisfaction with residence. Conclusions Overall satisfaction with the community environment emerged as the most significant predictor of life satisfaction among older adults living alone. These findings indicate that enhancing the supportiveness of the community environment could improve life satisfaction for this demographic.

Objectives: This study aimed to identify factors associated with life satisfaction by developing machine learning (ML) models to predict life satisfaction in older adults living alone. Methods: Data were extracted from 3112 older adults participating in the 2020 Korea Senior Survey. We employed 5 ML models to classify the life satisfaction of older adults living alone: logistic Lasso regression, decision tree-based classification and regression tree (CART), C5.0, random forest, and extreme gradient boost (XGBoost). The variables used as predictors included demographics, health status, functional abilities, environmental factors, and activity participation. The performance of these ML models was evaluated based on accuracy, precision, recall, F1-score, and area under the curve (AUC). Additionally, we assessed the significance of variable importance as indicated by the final classification models. Results: Out of the 1411 older adults living alone, 45.3% expressed satisfaction with their lives. The XGBoost model surpassed the performance of other models, achieving an F1-score of 0.72 and an AUC of 0.75. According to the XGBoost model, the five most important variables influencing life satisfaction were overall community satisfaction, self-rated health, opportunities to interact with neighbors, proximity to a child, and satisfaction with residence. Conclusions Overall satisfaction with the community environment emerged as the most significant predictor of life satisfaction among older adults living alone. These findings indicate that enhancing the supportiveness of the community environment could improve life satisfaction for this demographic.

Objectives: This study aimed to identify factors associated with life satisfaction by developing machine learning (ML) models to predict life satisfaction in older adults living alone. Methods: Data were extracted from 3112 older adults participating in the 2020 Korea Senior Survey. We employed 5 ML models to classify the life satisfaction of older adults living alone: logistic Lasso regression, decision tree-based classification and regression tree (CART), C5.0, random forest, and extreme gradient boost (XGBoost). The variables used as predictors included demographics, health status, functional abilities, environmental factors, and activity participation. The performance of these ML models was evaluated based on accuracy, precision, recall, F1-score, and area under the curve (AUC). Additionally, we assessed the significance of variable importance as indicated by the final classification models. Results: Out of the 1411 older adults living alone, 45.3% expressed satisfaction with their lives. The XGBoost model surpassed the performance of other models, achieving an F1-score of 0.72 and an AUC of 0.75. According to the XGBoost model, the five most important variables influencing life satisfaction were overall community satisfaction, self-rated health, opportunities to interact with neighbors, proximity to a child, and satisfaction with residence. Conclusions Overall satisfaction with the community environment emerged as the most significant predictor of life satisfaction among older adults living alone. These findings indicate that enhancing the supportiveness of the community environment could improve life satisfaction for this demographic.

OBJECTIVE: To evaluate the utility of using the Personal Assistance Services classification system (PAS-CS) that examines individuals with disabilities for services and government funding. To this end, this study also tests for significant differences in PAS-CS scores across disability grades and disability types. METHODS: A retrospective analysis was conducted using the 2014 National Survey on People with Disabilities (NSPD) data set. We selected patients with three types of disabilities (physical disabilities, brain lesions, and visual impairments). We compared the average PAS-CS scores of patients with different disability types and grades using general linear models with multiple comparisons. RESULTS: A total of 4,810 patients were included in the analysis. Patients with brain lesions had the highest average PAS-CS scores in activities of daily living (ADL) and instrumental activities of daily living (IADL) domains. Patients with visual impairments had the highest average scores in ‘Disease-specific disability’ and ‘Social-environment’ domains. For patients with physical disabilities and visual impairments, no PAS-CS domains were significantly different between patients with disability grade III and those with disability grade IV (p>0.05). CONCLUSION: The PAS-CS scores of disability grades were not equivalent among individuals with different disability types. The Korean Ministry of Health and Welfare currently only considers certain disability grades for PAS preeligibility, as a result disregarding the characteristics of different disability types. Thus, the current PAS-CS requires modifications.
Activities of Daily Living ; Brain ; Classification* ; Dataset ; Disability Evaluation ; Disabled Persons ; Financial Management ; Humans ; Linear Models ; Public Assistance ; Republic of Korea ; Retrospective Studies ; Social Welfare ; Vision Disorders

OBJECTIVES: Few longitudinal studies have explored age-related differences in the relationship between lifestyle factors and cognitive decline. This study investigated lifestyle factors at baseline that slow the longitudinal rate of cognitive decline in young-old (55-64 years), middle-old (65-74 years), and old-old (75+ years) individuals. METHODS: We conducted an 11-year follow-up that included 6,189 older adults from the Korean Longitudinal Study of Aging, which is a cohort study of community-dwelling older Koreans. Lifestyle factors, including physical activity, social activity (SA), smoking, and alcohol consumption were assessed at baseline. Cognitive function was measured at 2-year intervals over 11 years. Latent growth modeling and multi-group analysis were performed. RESULTS: The influence of lifestyle factors on the rate of cognitive decline differed by age. Smoking at baseline (-0.05; 95% confidence interval [CI], -0.11 to -0.00, per study wave) accelerated cognitive decline in young-old individuals, whereas frequent participation in SA at baseline (0.02; 95% CI, 0.01 to 0.03, per study wave) decelerated cognitive decline in middle-old individuals. None of the lifestyle factors in this study decelerated cognitive decline in old-old individuals. CONCLUSIONS Cognitive strategies based on modifiable lifestyle factors such as smoking cessation in young-old individuals and frequent SA participation in middle-old age individuals may have great potential for preventing cognitive decline. Because the influence of lifestyle factors varied by age group, age-specific approaches are recommended to promote cognitive health.

OBJECTIVE: To describe the longitudinal characteristics of unintentional fall accidents using a representative population-based sample of Korean adults. METHODS: We examined data from the Korean Community Health Survey from 2008 to 2013. Univariate analysis and multivariable logistic regression were used to identify the characteristics of fall accidents in adults. RESULTS: Between 2008 and 2013, the incidence rate of fall accidents requiring medical treatment increased from 1,248 to 3,423 per 100,000 people (p<0.001), while the proportion of indoor fall accidents decreased from 38.12% to 23.16% (p<0.001). Females had more annual fall accidents than males (p<0.001). The major reason for fall accidents was slippery floors (33.7% in 2011 and 36.3% in 2013). Between 2008 and 2010, variables associated with higher fall accident risk included specific months (August and September), old age, female gender, current drinker, current smoker, diabetes, osteoarthritis, osteoporosis, and depression. A high level of education and living with a partner were negatively associated with fall accident risk. In 2013, people experiencing more than 1 fall accident felt more fear of falling than those having no fall accidents (odds ratio [OR] for 1 fall, 2.12; 95% confidence interval [CI], 2.04–2.12; OR for more than 2 falls, 2.97; 95% CI, 2.83–3.10). CONCLUSION: The occurrence of fall accidents has consistently increased in Korea from 2008 to 2013. Future intervention studies are needed to reduce the increasing incidence rates of fall accidents in community dwelling adults.
Accidental Falls ; Adult* ; Depression ; Education ; Female ; Health Surveys* ; Humans ; Incidence ; Independent Living* ; Korea ; Life Style ; Logistic Models ; Male ; Osteoarthritis ; Osteoporosis

OBJECTIVE: To compare the risk of depressive symptoms in adults with physical disabilities between rehabilitation-receivers and non-receivers. METHODS: A total of 3,568 adults with physical disabilities were retrieved from the 2014 Korean National Survey on People with Disabilities database. Four covariate adjustment methods (a multivariable regression model, inverse probability of treatment weighting [IPTW] adjusted for normalized weight, IPTW with stabilized weight, and greedy algorithm with 1:1 propensity score matching) were used to estimate the odds of having depressive symptoms. The dependent variable was depressive symptoms and the independent variable was the use of rehabilitation services. Baseline covariates were 19 demographic variables and 10 chronic condition variables. RESULTS: The four covariate adjustment methods revealed that adults with physical disabilities receiving rehabilitation services had a higher risk of depressive symptoms than those who did not receive these services (adjusted odds ratio, 1.191–1.294). CONCLUSION: Our findings suggest that adults with physical disabilities receiving rehabilitation services have higher risk of developing depressive symptoms. Therefore, rehabilitation professionals need to pay attention to depressive symptoms and establish therapeutic strategies that can reduce such risk in rehabilitation settings.
Adult ; Depression ; Disabled Persons ; Humans ; Odds Ratio ; Propensity Score ; Rehabilitation

OBJECTIVE: To investigate the psychometric properties of the activities of daily living (ADL) instrument used in the analysis of Korean Longitudinal Study of Ageing (KLoSA) dataset. METHODS: A retrospective study was carried out involving 2006 KLoSA records of community-dwelling adults diagnosed with stroke. The ADL instrument used for the analysis of KLoSA included 17 items, which were analyzed using Rasch modeling to develop a robust outcome measure. The unidimensionality of the ADL instrument was examined based on confirmatory factor analysis with a one-factor model. Item-level psychometric analysis of the ADL instrument included fit statistics, internal consistency, precision, and the item difficulty hierarchy. RESULTS: The study sample included a total of 201 community-dwelling adults (1.5% of the Korean population with an age over 45 years; mean age=70.0 years, SD=9.7) having a history of stroke. The ADL instrument demonstrated unidimensional construct. Two misfit items, money management (mean square [MnSq]=1.56, standardized Z-statistics [ZSTD]=2.3) and phone use (MnSq=1.78, ZSTD=2.3) were removed from the analysis. The remaining 15 items demonstrated good item fit, high internal consistency (person reliability=0.91), and good precision (person strata=3.48). The instrument precisely estimated person measures within a wide range of theta (−4.75 logits <θ< 3.97 logits) and a reliability of 0.9, with a conceptual hierarchy of item difficulty. CONCLUSION: The findings indicate that the 15 ADL items met Rasch expectations of unidimensionality and demonstrated good psychometric properties. It is proposed that the validated ADL instrument can be used as a primary outcome measure for assessing longitudinal disability trajectories in the Korean adult population and can be employed for comparative analysis of international disability across national aging studies.
Activities of Daily Living ; Adult* ; Aging ; Dataset ; Humans ; Longitudinal Studies* ; Outcome Assessment (Health Care) ; Psychometrics* ; Reproducibility of Results ; Retrospective Studies ; Stroke* ; Surveys and Questionnaires

Background: Impaired executive function is common in older adults. This study examined the causal relationship between the use of information and communication technology (ICT) and executive function in older adults over time.Method: This study performed a secondary analysis of data from four waves (2016–2019) of the National Health and Aging Trends Study. A fixed-effect analysis was conducted to examine the effects of ICT on the executive function of older adults without dementia aged ≥65 years. This study analyzed data from 3,334 respondents. Results: We observed significant positive effects of ICT use on executive function over time (standardized β=0.043–0.045; 95% confidence interval, 0.001–0.043; p<0.05). Conclusion The current findings support the use of ICT as a protective approach to prevent decline in executive function in community-dwelling older adults.

Background: Impaired executive function is common in older adults. This study examined the causal relationship between the use of information and communication technology (ICT) and executive function in older adults over time.Method: This study performed a secondary analysis of data from four waves (2016–2019) of the National Health and Aging Trends Study. A fixed-effect analysis was conducted to examine the effects of ICT on the executive function of older adults without dementia aged ≥65 years. This study analyzed data from 3,334 respondents. Results: We observed significant positive effects of ICT use on executive function over time (standardized β=0.043–0.045; 95% confidence interval, 0.001–0.043; p<0.05). Conclusion The current findings support the use of ICT as a protective approach to prevent decline in executive function in community-dwelling older adults.