Hae-Young Kim

Background: Although periodontitis has been linked to systemic conditions such as metabolic syndrome, longitudinal evidence remains limited. This nationwide population‐based cohort study aimed to investigate whether periodontal status is associated with the incidence of metabolic syndrome in Korea. Methods: We analyzed data obtained from the National Health Insurance Service‐Health Screening cohort (2009–2019), including 43,713 participants aged ≥40 years. Periodontitis was assessed using diagnostic codes and periodontal treatment procedures. Metabolic syndrome was defined based on the modified National Cholesterol Education Program Adult Treatment Panel III criteria. This longitudinal cohort analysis used Cox proportional hazards models to examine the association between baseline periodontitis and the subsequent incidence of metabolic syndrome. Results: During the 9‐year follow‐up, individuals with severe periodontitis had a significantly elevated risk of developing metabolic syndrome compared with those without periodontitis (hazard ratio [HR] = 1.10; 95% confidence interval [CI]: 1.05–1.16). The association was particularly pronounced in individuals aged 50–59 years (HR = 1.16; 95% CI: 1.10–1.23), indicating a significant interaction between age and periodontal severity ( p for interaction = 0.015). Conclusions: These findings suggest that oral inflammatory conditions may be implicated in metabolic health, with the strength of the association varying by age. Thus, age‐specific vulnerability should be considered when developing preventive strategies.

Objective: We aimed to develop and evaluate machine learning models to support population‐level risk stratification for dental caries in the permanent dentition of 12‐year‐old Korean children and to visualise the relative importance of key predictors. Methods: This cross‐sectional study analysed data from the 2021–2022 National Survey on Children's Oral Health, which included 18,671 children aged 12 years. Dental caries was assessed in accordance with the World Health Organisation Oral Health Surveys: Basic Methods. To predict the presence of dental caries, established caries risk factors were identified based on previous studies. Predictive models were constructed using regularised Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), XGBoost and CatBoost algorithms, and model performance was evaluated using the area under curve (AUC) and additional metrics. The importance of key predictors was visualised using SHAP values. Results: The study included 18,671 participants aged 12 years, of whom 51.6% were boys and 48.4% were girls; 92.9% resided in urban areas. Overall, 6.9% of participants had dental caries in permanent teeth. Ridge LR achieved the highest performance on the weighted test set (AUC: 0.724, 95% CI: 0.690–0.754). The most important predictors of dental caries, ranked by importance, included the absence of sealants, poorer self‐perceived oral health, unmet dental treatment needs, lack of preventive dental care within the past year, girls, gingival bleeding, lack of oral health education, absence of orthodontic appliances, non‐use of dental floss, and tooth pain or discomfort. Furthermore, dental caries was more likely among individuals with lower perceived household economic status, higher consumption of beverages and snacks, residence in rural areas and a lower frequency of tooth brushing. Conclusion: Findings underscore the importance of preventive oral behaviours; integrating predictive models into public health initiatives could help target interventions and improve oral health outcomes in children.

Background: Digital health technologies offer promising solutions for managing chronic pain and depression in older adults, yet low-income populations with limited digital literacy face substantial barriers to access. Community-based approaches that leverage existing care infrastructure may bridge this digital divide, but evidence remains limited on effective integration strategies for digitally excluded populations. Objective: We evaluated a multimodal digital health monitoring platform for this underserved population through 6-week living lab trials with 86 low-income Korean older adults (43 intervention and 43 age- and gender-matched controls) and 25 community caregivers. Methods: Participants were recruited offline from recipients of government-supported community care services. The platform integrated chatbot surveys, smartwatch monitoring, and motion sensors, generating personalized traffic-light alerts when health indicators deviated from individual baselines and triggering real-time caregiver notifications for persistent declines. Mobile apps for older adult users and caregivers, along with a centralized monitoring system for community center managers, were provided to the intervention group, while controls received standard community care. Pre- and posttest surveys assessed changes in pain-related functional limitations, depressive symptoms, sleep quality, and system usability through face-to-face assessments. Caregivers documented platform-triggered interventions. Results: Following attrition due to illness (n=3) and participation burden (n=4), we analyzed 1318 days of continuous monitoring data from 35 intervention participants (mean 37.7, SD 8.8 days per participant), with 77 participants (35 intervention and 42 control) completing pre-post assessments. Distinct patterns emerged in 24-hour heart rate profiles: participants experiencing higher-than-usual pain demonstrated elevated heart rates during early morning (5-8 AM) and late evening (10-11 PM) hours compared to their low-pain days. Multilevel modeling revealed significant within-person associations between digital biomarkers and symptoms. Heart rate variability (P=.02) and moderate physical activity (P=.05) were associated with same-day pain. Heart rate variability (P=.03) and long wake episodes during sleep (P=.03) predicted next-day pain. Shorter sleep duration (P=.03) and lower sleep efficiency (P=.05) were associated with same-day depressive symptoms. A significant group-by-time interaction was observed for pain-related functional limitations (P=.03): the intervention group maintained baseline levels while the comparison group reported increased functional limitations. Depressive symptoms, sleep quality, and platform usability did not show significant changes. Community caregivers successfully conducted 37 health decline–triggered interventions during regular service hours (25 hours per week), demonstrating integration of digital monitoring with existing care workflows. Conclusions: Integrating digital health monitoring within existing community care infrastructure may be a feasible approach to supporting vulnerable older adults. The intervention was associated with maintained pain-related functional limitations, while depressive symptoms and sleep quality showed no significant changes, possibly due to the 6-week duration. This feasibility study provides preliminary support for digital-human integration models to address health equity, though randomized controlled trials are needed to establish causal relationships and isolate active intervention components. Trial Registration: ClinicalTrials.gov NCT06270121; https://clinicaltrials.gov/study/NCT06270121

Introduction: The primary objective was to assess 1‐year clinical and radiographic outcomes of a newly designed implant with a machined long neck in the anterior region, including marginal bone loss, horizontal bone width, vertical bone height, and implant survival and success rates. The secondary objective was to evaluate the effect of guided bone regeneration on bone dimensional changes and implant survival and success rates. Materials and Methods: A total of 117 implants in 58 patients who received the novel implant in the anterior region were included. Marginal bone loss, horizontal bone width, and vertical bone height changes were assessed using panoramic radiographs and cone beam computed tomography (CBCT) scans obtained after prosthetic restoration and at the one‐year follow‐up. Implant survival and success rates, along with the influence of guided bone regeneration (GBR) on bone dimensional changes, were statistically analyzed. Results: Within the one‐year follow‐up period, implant survival and success rates were 97.43% and 91.45% at the implant level, and 94.83% and 82.76% at the patient level, respectively. Among the 114 surviving implants, the mean marginal bone loss was 0.46 ± 1.05 mm. In the subset of 32 implants with available CBCT data, horizontal bone width and vertical bone height changes were minimal in both buccal and palatal/lingual aspects. Subgroup analyses indicated that guided bone regeneration provided measurable advantages in preserving bone dimensions at specific sites, particularly in the mid‐buccal region and the palatal/lingual vertical height. However, at most other reference points, dimensional changes were comparable irrespective of whether GBR was performed. Conclusion: A novel implant featuring a machined long neck exhibited clinically acceptable outcomes with respect to implant survival, implant success, and bone dimensional stability over a one‐year follow‐up period.

Objective Research specifically targeting suicidal ideation (SI) in middle-aged populations remains limited. This study aimed to predict future and concurrent SI in middle-aged Korean adults by applying four machine learning (ML) models to a nationally representative longitudinal dataset.Methods We analyzed data from 8,992 individuals aged 40–64 years who participated in the Korea Welfare Panel Study from the 7th (2011) to the 18th (2022) waves. Four ML algorithms were employed to develop the predictive models. The SHapley Additive exPlanations method was applied to enhance explainability.Results Approximately half of the participants’ mean age was 49.3±8.2 years (range, 40–64 years) and 52.2% were male. The average annual SI rate between 2011 and 2022 was 2.8%±1.2%. Predictive performance for future SI was satisfactory, with area under the receiver operating characteristic curve (AUC) values of up to 0.806 (logistic regression, LR). Predictions for concurrent SI demonstrated AUC values of up to 0.907 (LR). Key predictors of future SI included subjective health status, satisfaction with family and spousal relationships, housing environment, and educational attainment. Concurrent SI was strongly associated with immediate stressors such as family violence and income dissatisfaction.Conclusion The ML models demonstrated good-to-excellent predictive performance for SI. These findings emphasize the importance of health, family, and socioeconomic factors, alongside mental health indicators in the prevention of SI among middle-aged adults. Building on these findings, tailored intervention strategies that comprehensively address multidimensional risk factors are essential for effective SI prevention.

Chronic low back pain (cLBP) is shaped by multiple lifestyle factors, yet models with practical behavioral cutoffs remain scarce. This study developed a machine learning model to identify key lifestyle factors and specific thresholds linked to cLBP in adults aged 50 and older, using data from 5607 participants in the Korean National Health and Nutrition Examination Survey. Machine learning algorithms were trained and validated, with performance assessed via AUROC and SHAP for interpretability. The logistic regression model performed best (AUROC = 0.721, 95% CI: 0.699–0.742). SHAP analysis revealed that higher cLBP risk was associated with older age, female gender, prolonged sitting (≥ 6 h/day), low walking frequency (< 4–5 times/week), infrequent strength training (< 1 time/week), moderate‐intensity work, elevated stress, and smoking over five packs lifetime. Diet also mattered: cLBP risk rose among those dining out less than ~2.2 times/week, consuming under 2.9 servings/day of protein, or with carbohydrate intake outside 55%–65% of total energy. These practical cutoffs can help clinicians identify high‐risk individuals through simple assessments, guiding tailored interventions in physical activity, diet, smoking cessation, and stress management to prevent cLBP.

BACKGROUND Digital health technologies offer promising solutions for managing chronic pain and depression in older adults, yet low-income populations with limited digital literacy face substantial barriers to access. Community-based approaches that leverage existing care infrastructure may bridge this digital divide, but evidence remains limited on effective integration strategies for digitally excluded populations. OBJECTIVE We evaluated a multimodal digital health monitoring platform for this underserved population through 6-week living lab trials with 86 low-income Korean older adults (43 intervention and 43 age- and gender-matched controls) and 25 community caregivers. METHODS Participants were recruited offline from recipients of government-supported community care services. The platform integrated chatbot surveys, smartwatch monitoring, and motion sensors, generating personalized traffic-light alerts when health indicators deviated from individual baselines and triggering real-time caregiver notifications for persistent declines. Mobile apps for older adult users and caregivers, along with a centralized monitoring system for community center managers, were provided to the intervention group, while controls received standard community care. Pre- and posttest surveys assessed changes in pain-related functional limitations, depressive symptoms, sleep quality, and system usability through face-to-face assessments. Caregivers documented platform-triggered interventions. RESULTS Following attrition due to illness (n=3) and participation burden (n=4), we analyzed 1318 days of continuous monitoring data from 35 intervention participants (mean 37.7, SD 8.8 days per participant), with 77 participants (35 intervention and 42 control) completing pre-post assessments. Distinct patterns emerged in 24-hour heart rate profiles: participants experiencing higher-than-usual pain demonstrated elevated heart rates during early morning (5-8 AM) and late evening (10-11 PM) hours compared to their low-pain days. Multilevel modeling revealed significant within-person associations between digital biomarkers and symptoms. Heart rate variability (P=.02) and moderate physical activity (P=.05) were associated with same-day pain. Heart rate variability (P=.03) and long wake episodes during sleep (P=.03) predicted next-day pain. Shorter sleep duration (P=.03) and lower sleep efficiency (P=.05) were associated with same-day depressive symptoms. A significant group-by-time interaction was observed for pain-related functional limitations (P=.03): the intervention group maintained baseline levels while the comparison group reported increased functional limitations. Depressive symptoms, sleep quality, and platform usability did not show significant changes. Community caregivers successfully conducted 37 health decline–triggered interventions during regular service hours (25 hours per week), demonstrating integration of digital monitoring with existing care workflows. CONCLUSIONS Integrating digital health monitoring within existing community care infrastructure may be a feasible approach to supporting vulnerable older adults. The intervention was associated with maintained pain-related functional limitations, while depressive symptoms and sleep quality showed no significant changes, possibly due to the 6-week duration. This feasibility study provides preliminary support for digital-human integration models to address health equity, though randomized controlled trials are needed to establish causal relationships and isolate active intervention components. CLINICALTRIAL ClinicalTrials.gov NCT06270121; https://clinicaltrials.gov/study/NCT06270121

Objective This study aimed to clarify how chronic diseases (CDs) contribute to depression and suicidal ideation (SI) prediction using machine learning (ML) techniques among the older adult population.Methods National representative data of 5,419 older adults from the Korea National Health and Nutrition Examination Survey conducted in 2013, 2015, 2017, and 2019 were used in this study. The number and type of CDs were incorporated into Models 1 and 2, respectively, using five ML methods.Results The average age of the participants was 72.7 years, with 43.2% males, 15.2% reporting depression, and 7.3% reporting SI. The number of CDs was correlated with increased depression and SI. The ML models showed moderate-to-good performance in the prediction of depression and SI. The area under the curve (AUC) values for Model 1 ranged from 0.729 to 0.772 for depression, and from 0.754 to 0.793 for SI. In Model 2, the AUC ranged from 0.704 to 0.768 for depression and from 0.750 to 0.785 for SI. More depression and SI were expected when the number of CDs was one or more and two or more, respectively. The top predictors of depression were osteoarthritis, myocardial infarction, diabetes, asthma, and stroke, whereas those predicting SI were stroke, hypertension, asthma, myocardial infarction, and rheumatoid arthritis.Conclusion The number and specific types of CDs predicted depression and SI among Korean older adults. These results may help enhance cooperation with clinicians treating CDs and promote the early detection and prevention of further SI and behaviors.

Background: This study investigated the association between lifestyle-related factors and the risk of cardiac and cerebrovascular disease in patients who underwent major abdominal surgery, using the Korean National Health Insurance Service (NHIS) data. Methods: We analyzed individuals aged >40 years who received health check-ups between 2009 and 2018. Among these data, we enrolled 108,261 patients who underwent major abdominal surgery and were followed-up to assess the development of cardiac or cerebrovascular disease for 1 year after surgery. Logistic regression was used to evaluate the association between lifestyle-related factors — body mass index, metabolic syndrome, alcohol consumption, cigarette smoking, and physical activity —and the outcomes. Results: Patients with metabolic syndrome and current smokers had a significantly higher risk of developing both cardiac and cerebrovascular disease within 1 year after major abdominal surgery. Ex-smokers and overweight patients had a higher risk of postoperative cerebrovascular disease. Alcohol consumption was associated with a reduced risk of cardiac disease and high levels of physical activity were linked to a reduced risk of cerebrovascular events. Conclusion: Our data suggest that lifestyle modification including weight control and cessation of smoking before undergoing major abdominal surgery may prevent postoperative cardiac and cerebrovascular risks.