Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Hyeri Shin has a total of 83 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Predicting Care Needs in Community-Dwelling Older Adults Using Explainable Machine Learning and a Multidimensional Approach Integrating Health, Social, and Environmental Factors: Cross-Sectional Study

    by Hyeri Shin

    Background: Rapid population aging and a worsening shortage of care workers necessitate the identification of older adults who require proactive interventions. Although machine learning (ML) has been increasingly applied in gerontology, existing studies have predominantly focused on social isolation, loneliness, depression, falls, and frailty in isolation rather than on the integrated construct of care needs. Objective: This study aimed to develop and interpret an explainable ML model that identifies care needs in community-dwelling Korean older adults. Beyond physical health indicators such as disease and functional status, this study adopted a comprehensive approach that included mental health, cognitive function, health behaviors, and socioenvironmental determinants, such as social participation, social support, and the housing environment, to present an integrated model encompassing both health and social care needs. Methods: Data were obtained from the 2023 Korea Senior Survey, a nationally representative sample of 10,078 community-dwelling adults aged 60 years and older. The data were split 70:30 into training (n=7054) and held-out test (n=3024) sets. Seven algorithms were compared (logistic regression, decision tree, support vector machine, random forest, gradient-boosted decision trees, extreme gradient boosting, and light gradient boosting machine) using stratified 5-fold cross-validation on the training set. Discrimination was assessed by the area under the receiver operating characteristic curve (AUC). Model interpretability used Shapley additive explanations with bootstrap stability assessment across folds. Results: Model A (excluding activities of daily living or instrumental activities of daily living [IADL]) achieved good discrimination (AUC 0.892, 95% CI 0.866‐0.918), adequate calibration (calibration slope=0.826), and positive clinical net benefit, demonstrating that upstream factors alone can identify older adults with care needs without relying on functional status. Shapley additive explanations analysis identified age, nutritional risk, employment status, depressive symptoms, self-rated health, cognitive function, household income, and home modification as the leading predictors, with high rank stability across cross-validation folds. Model B (including activities of daily living or IADL) yielded a higher AUC (0.976, 95% CI 0.962‐0.989), but this reflected the near-tautological relationship between IADL and self-reported care needs rather than genuine upstream predictive value. Using an objective composite outcome yielded equivalent discrimination (AUC 0.892), supporting robustness to the outcome definition. Conclusions: Explainable ML models offer high predictive accuracy and strong interpretability for identifying care needs among older adults. Care needs in community-dwelling Korean older adults can be identified with good discrimination, calibration, and clinical net benefit using multidimensional nonfunctional factors alone. By highlighting the significant roles of health, social, and environmental factors, this study provides empirical evidence to support evidence-based decision-making for the Long-Term Care Insurance system and integrated community care policies.

  • Open Access

    Digital Social Engagement and Frailty Among Korean Older Adults: The Sequential Mediating Role of Social Isolation and Self-Efficacy (Preprint)

    by Heejeong Yoon, Jungwan Lee, Siwoo Ban, Donghyun Lee, Hyeri Shin, Youngsun Kim

    BACKGROUND Digital social engagement is increasingly recognized as a modifiable resource for healthy aging. However, the psychosocial mechanisms linking digital engagement to physical frailty remain insufficiently understood. OBJECTIVE This study examined whether social isolation and self-efficacy sequentially mediated the association between digital social engagement and frailty among Korean older adults. METHODS Data were drawn from the 2024 AgeTech Research Institute panel survey of Korean older adults (N = 435). Serial mediation analysis was conducted using PROCESS macro Model 6, adjusting for sociodemographic covariates. RESULTS Higher digital social engagement was associated with lower frailty. Two indirect pathways were statistically significant: an independent pathway through self-efficacy and a sequential pathway through social isolation and self-efficacy. The independent mediating effect of social isolation was not statistically significant. Self-efficacy showed the strongest mediating effect in the association between digital social engagement and frailty. CONCLUSIONS Digital social engagement may support healthy aging not only by strengthening social connections but also by enhancing psychological resources. Self-efficacy appears to be a key mechanism linking digital social engagement to lower frailty vulnerability. Interventions that promote meaningful digital social engagement and strengthen self-efficacy may contribute to frailty prevention among older adults.

  • AgeTech Adoption as a Moderator of the Relationship Between Job Stress and Turnover Intention Among Long-Term Care Workers in South Korea

    by Nam Hwa Kim, Hye Ri Shin, Yeo Min Han, Jae Hyeok Choi, Young Sun Kim

    This cross-sectional study examined whether AgeTech adoption moderates the association between job stress and turnover intention among long-term care workers in South Korea. A survey was conducted among care workers employed in 15 long-term care facilities and hospitals, and data from 217 respondents were analyzed. The findings showed that job stress was significantly associated with greater turnover intention among care workers. In addition, higher levels of AgeTech adoption weakened this positive relationship. The results provide empirical evidence that AgeTech functions as a buffering resource in the caregiving environment. Effective AgeTech adoption may serve as both a practical measure and a policy strategy to alleviate workforce shortages and enhance long-term care sustainability.

  • Open Access

    Predicting Care Needs Among Older Adults Using Explainable Machine Learning: A Multidimensional Approach Integrating Health, Social, and Environmental Factors (Preprint)

    by Hyeri ‍sHin

    BACKGROUND Rapid population aging and a worsening shortage of care workers necessitates the identification of older adults who require proactive interventions. Although machine learning (ML) has been increasingly applied in gerontology, existing studies have predominantly focused on social isolation. OBJECTIVE This study aims to develop a robust predictive model for care needs using ML. Beyond physical health indicators such as disease and functional status, this study adopted a comprehensive approach including mental health, cognitive function, health behaviors, and socio–environmental determinants—such as social participation, social support, and housing environment—to present an integrated model encompassing both health and social care needs. METHODS Data were obtained from the 2023 Korea Senior Survey, a nationally representative sample of adults aged 60 and older. A set of candidate predictors was initially screened using the Light Gradient Boosting Machine (LightGBM) algorithm. Subsequently, we trained and compared 7 predictive models—logistic regression, decision tree, support vector machine, random forest, gradient boosting decision tree, XGBoost, and LightGBM. Model performance was evaluated using accuracy, sensitivity, specificity, F1 score, area under the receiver operating characteristic curve (AUC), and precision-recall AUC. Model interpretability was enhanced using Shapley Additive Explanations (SHAP) to quantify the contributions of individual predictors. RESULTS Among the evaluated models, XGBoost demonstrated the highest predictive performance. SHAP analysis revealed that limitations in instrumental activities of daily living (IADL) were the most influential predictors. Age, nutritional risk, depression, subjective health, work, cognitive function, home modification, and income emerged as the top predictors when functional status was excluded. These findings indicate that care needs are driven not only by functional impairment but also by distinct sociodemographic factors. CONCLUSIONS Explainable ML models offer high predictive accuracy and strong interpretability for identifying care needs among older adults. By highlighting the significant roles of health, social, and environmental factors, this study provides empirical evidence to support evidence-based decision-making for the Long-Term Care Insurance system and integrated community care policies.

  • Open Access

    Effects of Digital Literacy on Older Adults’ Intention to Use Digital Healthcare: A Serial Multiple Mediation Analysis of Technostress and Technology Self-Efficacy

    by Jungwan Lee, Hyeri Shin, Yeomin Han, Donghyun Lee, Youngsun Kim

    Objective: This study aims to examine the association between digital literacy and older adults’ intention to use digital healthcare services by testing the serial multiple mediating roles of technostress and technology self-efficacy. Methods: Data were drawn from the 2024 Korean Older Adults Technology Acceptance Panel Survey conducted with 435 adults aged 65 and over. Participants completed validated measures of digital literacy, technostress, technology self-efficacy, and intention to use digital healthcare. Serial multiple mediation analysis was performed using PROCESS Model 6 with demographic controls. Results: Digital literacy significantly reduced technostress and increased technology self-efficacy. A sequential mediating pathway was confirmed: higher digital literacy lowered technostress, which subsequently enhanced technology self-efficacy. Although digital literacy did not directly predict intention to use digital healthcare, a significant indirect effect was observed through technology self-efficacy, supporting the proposed serial multiple mediation model. Conclusions: Digital literacy indirectly shapes older adults’ willingness to use digital healthcare by alleviating technostress and strengthening technology self-efficacy. These results highlight the importance of targeted digital literacy initiatives to promote meaningful digital health engagement among older adults, particularly in rapidly aging and digitalized societies.

  • Open Access

    Technostress in older healthcare workers: Latent profiles and implications for care robot acceptance

    by Heejeong Yoon, Jungwan Lee, Namhwa Kim, Siwoo Ban, Hyeri Shin, Youngsun Kim

    Background: The adoption of care technologies in long-term care settings is ongoing; while the aging of the care workforce continues, particularly in South Korea. This study aims to identify technostress types among older care workers in Korea, including their influencing factors, and examine differences in acceptance of transfer-assistive care robots. Methods: The study analyzed data from a 2023 survey on 336 care workers aged 50 years and above. Latent profile analysis was conducted on five technostress dimensions: overload, invasion, complexity, privacy, and inclusion. The Three-Step method and the Bolck–Croon–Hagenaars command were used to examine the antecedents of profile membership and differences in technology acceptance variables, including self-efficacy, anxiety, attitude, perceived ease of use, perceived usefulness, and intention to use robots. Results: Four technostress profiles were identified: low technostress, high complexity and inclusion, moderate technostress, and high technostress with low inclusion. Age, education, income, job position, type of facility, work experience, and digital competence were all significantly associated with technostress profile membership. The low technostress group demonstrated the most favorable outcomes across all technology acceptance variables, whereas the high complexity and inclusion group exhibited the least favorable outcomes, characterized by the highest levels of anxiety and lowest levels of self-efficacy, perceived usefulness, and behavioral intention to use. Conclusions: Technostress profiles significantly influenced older healthcare workers’ acceptance of transfer-assistive robots. Tailored interventions to decrease technostress—particularly related to complexity and inclusion pressures—may enhance technology adoption and improve care delivery in healthcare settings.


  • Open Access

    An Integrated Analysis of Factors Influencing Acceptance of Care Robots Among Older Korean Adults

    by Hee Jeong Yoon, Si Woo Ban, Yeo Min Han, Hye Ri Shin, Young Sun Kim, Won Chul Shin, Seung Don Yoo, Ji Ho Park

    Background: As populations rapidly age, care robots have been proposed as a promising solution, supporting independent living and alleviating care burdens. However, acceptance of care robots among older adults remains limited. This study examined the relative contributions of demographic, health-related, digital competence, and technology-related psychological factors to care robot acceptance among older adults in South Korea. Methods: A cross-sectional survey was conducted with 506 community-dwelling older adults. Hierarchical multiple regression analyses were used to identify predictors of care robot acceptance, sequentially entering demographic characteristics, health-related factors, digital competence, and technology-related psychological variables. Results: Demographic and health-related factors were initially associated with care robot acceptance, but their effects diminished after accounting for digital competence and psychological variables. In the final model, technology-related psychological factors—particularly technology use self-efficacy and technology enthusiasm—were the strongest predictors of acceptance, while most demographic and health variables became non-significant, with the exception of instrumental activities of daily living (IADLs). Conclusions: These findings indicate that psychological readiness and confidence in technology use outweigh demographic or health characteristics in shaping older adults’ acceptance of care robots. Interventions and design strategies that enhance self-efficacy, foster positive engagement, and support functional independence may be critical for promoting the effective and sustainable adoption of care robots in aging societies.


  • Open Access

    Development of the Dietary Practices and Food Safety Literacy Scale for Older Adults

    by Ye-Rin Lee, Gi-Moon Nam, Young-Sun Kim, Hye-Ri Shin, Yoo-Kyung Park, Ji-Hye Mun, Su-Hyeun Cho, Hee-Sook Lim

    Background/Objectives: This study aimed to develop and validate a Dietary Practices and Food Safety Literacy Scale comprehensively assessing competencies among Korean older adults in healthy dietary practices, hygiene, and food safety. Methods: Item development was informed by a literature review, national dietary guidelines, and existing literacy frameworks. Content validity was reviewed by a 10-member expert panel using the Delphi method. Construct validity was tested using exploratory factor analyses (EFA) and confirmatory factor analyses (CFA), and reliability was assessed through Cronbach’s α, composite reliability (CR), and average variance extracted (AVE). Results: EFA identified three factors—management, decision-making, and moderation competencies—comprising 13 items. Internal consistency was acceptable (α = 0.69–0.83), and CFA supported the three-factor structure (CFI = 0.919, RMSEA = 0.087). CR values exceeded 0.70, and AVE were close to or exceeded the recommended threshold. Conclusions: The scale demonstrates sound psychometric properties and provides a practical tool for identifying competency gaps in Dietary Practices and Food Safety Literacy. Its application may guide tailored health education and community-based interventions to promote healthy aging and support public health strategies in aging societies. By translating health information literacy into measurable, behavior-oriented domains, this tool bridges the gap between theoretical constructs and practical assessment. It can be incorporated into routine health monitoring, enabling policymakers and practitioners to design evidence-based interventions that enhance older adults’ dietary self-management and food safety awareness.


  • Open Access

    Unsupervised Frailty Intervention by Digitalized Exercise Coaching: A Feasibility Study

    by Changseok Lee, Jehun Lee, Heeyoung Jeong, Haeram Lee, Eunah Wang, Gyungyoon Baek, Hyeri Shin, Seongjun Yoon

    Multi-component interventions have been demonstrated to be an effective method for the prevention of frailty. Nevertheless, they have not yet been widely adopted in practice due to considerable resource and labor demands associated with their administration. To overcome the limitations, this study aimed to determine the feasibility of unsupervised intervention based on digital therapy. A mobile application has been developed to deliver multi-component exercise coaching to frail older adults. A total of 30 participants (with a mean age of 72.10 ± 4.54 years) were recruited from two community centers and used the mobile application for 12 weeks without supervision. Prior to utilizing the mobile application, each participant received an initial education. Outcomes of the Short Physical Performance Battery (SPPB), the Korean version of the Fatigue, Resistance, Ambulation, Illnesses, and Loss of weight (K-FRAIL) scale, activities of daily living (ADL), instrumental activities of daily living (IADL), self-efficacy, and depression, were evaluated before and after using the mobile application. Significant improvements in the mean SPPB score (9.6 ± 1.7 to 11.7 ± 0.5) and depression (3.23 ± 3.08 to 2.00 ± 2.11) were observed. The total adherence rate of all participants was 86.1%. No statistically significant differences were observed in the remaining outcomes. These findings suggest that the unsupervised intervention could be a viable option for older adults.