Lowering Barriers to Health Risk Assessments in Promoting Personalized Health Management.

This study investigates the feasibility of accurately predicting adverse health events without relying on costly data acquisition methods, such as laboratory tests, in the era of shifting healthcare paradigms towards community-based health promotion and personalized preventive healthcare through individual health risk assessments (HRAs). We assessed the incremental predictive value of four categories of predictor variables-demographic, lifestyle and family history, personal health device, and laboratory data-organized by data acquisition costs in the prediction of the risks of mortality and five chronic diseases. Machine learning methodologies were employed to develop risk prediction models, assess their predictive performance, and determine feature importance. Using data from the National Sample Cohort of the Korean National Health Insurance Service (NHIS), which includes eligibility, medical check-up, healthcare utilization, and mortality data from 2002 to 2019, our study involved 425,148 NHIS members who underwent medical check-ups between 2009 and 2012. Models using demographic, lifestyle, family history, and personal health device data, with or without laboratory data, showed comparable performance. A feature importance analysis in models excluding laboratory data highlighted modifiable lifestyle factors, which are a superior set of variables for developing health guidelines. Our findings support the practicality of precise HRAs using demographic, lifestyle, family history, and personal health device data. This approach addresses HRA barriers, particularly for healthy individuals, by eliminating the need for costly and inconvenient laboratory data collection, advancing accessible preventive health management strategies.

The risk of fracture in age-related macular degeneration according to the presence of visual disability: a nationwide cohort study.

PURPOSE: To evaluate the association between age-related macular degeneration (AMD) with or without visual disability (VD) and the risk of fracture using the National Health Insurance data in South Korea. METHODS: In total, 3,894,702 individuals who had taken part in health-screening programs between January 1, 2009, and December 31, 2009, were included in the cohort and followed until December 31, 2019. The participants with VD, which could be related to the severity of AMD, were defined as those with a loss of vision or visual field defect as certified by the Korean government's Ministry of Health and Welfare. The hazard ratio was calculated for groups (control and AMD with/without VD) using the multivariable-adjusted cox regression analysis. RESULTS: In total, 466,890 participants (11.99%) were diagnosed with fractures during the study period. An increased risk of fracture was observed in individuals with AMD compared with the control (adjusted hazard ratio (aHR), 1.09, 95% confidence interval (CI), 1.06-1.11). Furthermore, among the AMD individuals, an increased risk of fracture was prominent in individuals with VD (aHR 1.17, 95% CI 1.08-1.27) than those without VD (aHR 1.08, 95% CI 1.06-1.11) compared with the reference group (control). CONCLUSIONS: AMD was associated with an increased risk of fracture even without VD. Prevention for fracture should be considered in AMD patients, especially when accompanied by VD.

Development of machine learning algorithms for scaling-up antibiotic stewardship.

BACKGROUND: Antibiotic stewardship programs (ASP) aim to reduce inappropriate use of antibiotics, but their labor-intensive nature impedes their wide adoption. The present study introduces explainable machine learning (ML) models designed to prioritize inpatients who would benefit most from stewardship interventions. METHODS: A cohort of inpatients who received systemic antibiotics and were monitored by a multidisciplinary ASP team at a tertiary hospital in the Republic of Korea was assembled. Data encompassing over 130,000 patient-days and comprising more than 160 features from multiple domains, including prescription records, laboratory, microbiology results, and patient conditions was collected.Outcome labels were generated using medication administration history: discontinuation, switching from intravenous to oral medication (IV to PO), and early or late de-escalation. The models were trained using Extreme Gradient Boosting (XGB) and light Gradient Boosting Machine (LGBM), with SHapley Additive exPlanations (SHAP) analysis used to explain the model's predictions. RESULTS: The models demonstrated strong discrimination when evaluated on a hold-out test set(AUROC - IV to PO: 0.81, Early de-escalation: 0.78, Late de-escalation: 0.72, Discontinue: 0.80). The models identified 41%, 16%, 22%, and 17% more cases requiring discontinuation, IV to PO, early and late de-escalation, respectively, compared to the conventional length of therapy strategy, given that the same number of patients were reviewed by the ASP team. The SHAP results explain how each model makes their predictions, highlighting a unique set of important features that are well-aligned with the clinical intuitions of the ASP team. CONCLUSIONS: The models are expected to improve the efficiency of ASP activities by prioritizing cases that would benefit from different types of ASP interventions along with detailed explanations.

Supervised machine learning model to predict mortality in patients undergoing venovenous extracorporeal membrane oxygenation from a nationwide multicentre registry.

BACKGROUND: Existing models have performed poorly when predicting mortality for patients undergoing venovenous extracorporeal membrane oxygenation (VV-ECMO). This study aimed to develop and validate a machine learning (ML)-based prediction model to predict 90-day mortality in patients undergoing VV-ECMO. METHODS: This study included 368 patients with acute respiratory failure undergoing VV-ECMO from 16 tertiary hospitals across South Korea between 2012 and 2015. The primary outcome was the 90-day mortality after ECMO initiation. The inputs included all available features (n=51) and those from the electronic health record (EHR) systems without preprocessing (n=40). The discriminatory strengths of ML models were evaluated in both internal and external validation sets. The models were compared with conventional models, such as respiratory ECMO survival prediction (RESP) and predicting death for severe acute respiratory distress syndrome on VV-ECMO (PRESERVE). RESULTS: Extreme gradient boosting (XGB) (areas under the receiver operating characteristic curve, AUROC 0.82, 95% CI (0.73 to 0.89)) and light gradient boosting (AUROC 0.81 (95% CI 0.71 to 0.88)) models achieved the highest performance using EHR's and all other available features. The developed models had higher AUROCs (95% CI 0.76 to 0.82) than those of RESP (AUROC 0.66 (95% CI 0.56 to 0.76)) and PRESERVE (AUROC 0.71 (95% CI 0.61 to 0.81)). Additionally, we achieved an AUROC (0.75) for 90-day mortality in external validation in the case of the XGB model, which was higher than that of RESP (0.70) and PRESERVE (0.67) in the same validation dataset. CONCLUSIONS: ML prediction models outperformed previous mortality risk models. This model may be used to identify patients who are unlikely to benefit from VV-ECMO therapy during patient selection.

Use of the Smart Excretion Care System Linked to Electronic Medical Records to Alleviate Nursing Burden and Enhance Patient Convenience: Mixed Methods Study.

BACKGROUND: The surge in older demographics has inevitably resulted in a heightened demand for health care, and a shortage of nursing staff is impending. Consequently, there is a growing demand for the development of nursing robots to assist patients with urinary and bowel elimination. However, no study has examined nurses' opinions of smart devices that provide integrated nursing for patients' urinary and bowel elimination needs. OBJECTIVE: This study aimed to evaluate the feasibility of the Smart Excretion Care System tethered to electronic medical records in a tertiary hospital and community care setting and discuss the anticipated reductions in the burden of nursing care. METHODS: Focus group interviews were conducted using the Consolidated Criteria for Reporting Qualitative Research (COREQ) guidelines. The interviews were conducted in March 2021 and involved 67 nurses who had worked at Seoul National University Bundang Hospital for more than 1 year and had experience in assisting patients with excretion care. Data were collected using purposive and snowball sampling methods. RESULTS: A total of four themes relevant to the Smart Excretion Care System were found: (1) expected reductions in the burden of nursing care, (2) applicable indications (by departments and diseases), (3) preferred features/functions, and (4) expected benefits of using the Smart Excretion Care System in clinical facilities. Nurses from comprehensive nursing care wards had the highest burden when it came to excretion care. It was a common opinion that the Smart Excretion Care System would be very useful in intensive care units and should be applied first to patients with stroke or dementia. CONCLUSIONS: Excretion care is one of the most burdensome tasks for nurses, increasing their workload. The development of the Smart Excretion Care System as a digital health intervention could help improve nurses' work efficiency, reduce their burden, and extend to caregivers and guardians.

Nurses' Burden of Elimination Care: Sequential Explanatory Mixed-Methods Design.

Background: Inpatients commonly experience problems with elimination due to incontinence, urinary retentions, and complications with indwelling catheters. Although elimination care (EC) is an important nursing area, few studies explore the burden of EC on nurses. Aim: To identify the burden on EC by analyzing nurses' opinions using sequential explanatory mixed method. Methods: This research was conducted using a sequential explanatory mixed-methods design. A total of 59 nurses at a tertiary hospital in South Korea participated in the study from January 1 to March 31, 2022. For quantitative analysis, information about number of delays of work due to EC, required time for serving bedpan or diaper changes, percentage of EC per shift, and percentage of patients who need EC was collected through a survey. For qualitative analysis, focus group interviews were conducted to identify factors that put a burden on EC. Important themes were derived by analyzing nurses' opinions on EC. Results: For nurses in intensive care units, general wards, and integrated nursing care wards, the number of work delays due to EC was 3.6 ± 1.5, 2.3 ± 1.2, and 4.8 ± 2.4 (p<0.01), respectively. The mean percentage of EC work out of total nursing tasks per shift was 36.2 ± 19.0, 29.3 ± 14.4, and 43.8 ± 14.1 (p=0.02), respectively. The mean percentage of patients requiring EC out of patients a nurse cares was 85.4 ± 16.6, 41.3 ± 26.1, and 58.8 ± 21.9 (p<0.01), respectively. Following qualitative analysis, four themes related to nurses' EC burden were derived: physical burden, frequent care needs, delay of other jobs due to EC, and complications. Among them, frequent care needs were found to be the primary factor requiring consideration to reduce nurses' burden. Conclusion: This research found that EC is one of the most burdensome tasks that nurses want to avoid. To alleviate their burden, effective EC protocol or smart medical devices assisting with EC should be developed.

The effects of special nursing units in nursing homes on healthcare utilization and cost: A case-control study using propensity score matching.

BACKGROUND: Most nursing homes in South Korea lack professional nursing services, resulting in transporting residents to hospitals for mild health problems and nursing treatment needs. While the number of nursing homes has increased, the number of registered nurses working in nursing homes has declined. In 2019, the Korean Ministry of Health and Welfare and the National Health Insurance Service launched the Special Nursing Units in Nursing Homes, a pilot nurse-led model, to resolve the lack of health and nursing services in nursing homes by mandating registered nurses' minimum staffing levels and protecting their scope of practice. OBJECTIVE: This study explored the effects of the Special Nursing Units model in Nursing Homes on healthcare utilization and cost among nursing home residents. DESIGN: A comparative effectiveness research design using propensity score matching. SETTING(S): Eighteen nursing homes were selected based on the region and number of beds. PARTICIPANTS: There were 323 matched-pairs of residents from the case and control groups. METHODS: Nursing homes with more than 30 beds were recruited nationwide, with 18 nursing homes being selected based on the region and number of beds. The case group included 323 older adults receiving professional nursing services by registered nurses under the Special Nursing Units model in Nursing Homes for more than six months consecutively in 18 nursing homes between April and December 2019. We matched control participants using propensity score matching with health insurance and long-term care data. We analyzed the differences in healthcare utilization and cost changes between the case and control groups using generalized estimating equations. RESULTS: The groups were not statistically different in baseline demographic or health-related characteristics. There were 26 (8.1%) and 30 (9.3%) deaths in the case and control groups, respectively, during the six months of the model, which was not statistically different (p = .576). The case group showed statistically significant decreases in healthcare utilization and costs, including hospitalization frequency (p = .008), length of stay (p = .002), and hospitalization costs (p = .003); outpatient visit frequency (p = .003) and costs (p < .001); and home healthcare frequency (p < .001) and cost (p < .001) than the control group. CONCLUSIONS: Professional nursing services by registered nurses under the Special Nursing Units model in Nursing Homes decreased healthcare utilization and costs. A nurse-led model in nursing homes, which includes mandating the minimum staffing levels of registered nurses and protecting their scope of practice, promises to improve resident health outcomes.

Changes in smoking, alcohol consumption, and the risk of Parkinson's disease.

Objective: There have been no studies on the association between changes in smoking and alcohol consumption or combined changes in smoking and alcohol consumption frequencies and PD risk. To assess the influence of changes in smoking and alcohol consumption on the risk of Parkinson's disease (PD). Methods: National Health Insurance Service (NHIS) database between January 2009 to December 2011 was analyzed. A total of 3,931,741 patients were included. Study participants were followed up for the incidence of PD until December 2017. Results: Compared to the sustained non-smokers, sustained light smokers (adjusted hazard ratio [aHR] 0.80, 95% confidence interval [CI] 0.75-0.85), sustained moderate smokers (aHR 0.54, 95% CI 0.47-0.61), and sustained heavy smokers (aHR 0.49, 95% CI 0.44-0.55) had a lower risk of PD. Compared to those who sustained non-drinking, sustained light drinkers (aHR 0.85 95% CI 0.89-0.91), sustained moderate drinkers (aHR 0.68, 95% CI 0.60-0.78), and sustained heavy drinkers (aHR 0.77, 95% CI 0.68-0.87) showed decreased risk of PD. Among non-drinkers, those who started drinking to a light level were at decreased risk of PD (aHR 0.84, 95% CI 0.77-0.91). Among non-smoking and non-drinking participants, those who initiated smoking only (aHR 0.78, 95% CI 0.70-0.86), drinking only (aHR 0.77, 95% CI 0.68-0.87), and both smoking and drinking (aHR 0.69, 95% CI 0.58-0.82) showed decreased risk of PD. Conclusion: Smoking is associated with decreased risk of PD with a dose-response relationship. Alcohol consumption at a light level may also be associated with decreased risk of PD. Further studies are warranted to find the possible mechanisms for the protective effects of smoking and drinking on PD, which may present insights into the etiology of PD.

The risk of dementia in multiple sclerosis and neuromyelitis optica spectrum disorder.

Introduction: Cognitive impairment is a common feature of multiple sclerosis (MS) and neuromyelitis optica spectrum disorder (NMOSD). However, there is a lack of population-based study of dementia risk in these disorders. In the present study, the risk of dementia in MS and NMOSD patients in Republic of Korea was estimated. Methods: Data analyzed in this study were obtained from the Korean National Health Insurance Service (KNHIS) database between January 2010 and December 2017. The study included 1,347 MS patients and 1,460 NMOSD patients ≥40 years of age who had not been diagnosed with dementia within 1 year prior to the index date. Matched controls were selected based on age, sex, and the presence of hypertension, diabetes mellitus, or dyslipidemia. Results: In MS and NMOSD patients, the risk of developing any dementia [adjusted hazard ratio (aHR) = 2.34; 95% confidence interval (CI) = 1.84-2.96 and aHR = 2.19; 95% CI = 1.61-3.00, respectively], Alzheimer's disease [AD; aHR = 2.23; 95% confidence interval (CI) = 1.70-2.91 and aHR = 1.99; 95% CI = 1.38-2.88, respectively], and vascular dementia (aHR = 3.75; 95% CI = 1.91-7.35 and aHR = 3.21; 95% CI = 1.47-7.02, respectively) was higher compared with the matched controls. NMOSD patients had a lower risk of any dementia and AD compared with MS patients after adjusting for age, sex, income, hypertension, diabetes, and dyslipidemia (aHR = 0.67 and 0.62). Conclusion: The risk of dementia increased in MS and NMOSD patients and dementia risk was higher in MS than in NMOSD.

The Risk of Type 2 Diabetes Mellitus according to Changes in Obesity Status in Late Middle-Aged Adults: A Nationwide Cohort Study of Korea.

BACKGRUOUND: Although obesity is a well-known risk factor of type 2 diabetes mellitus (T2DM), there is scant data on discriminating the contribution of previous obesity and recent weight gain on developing T2DM. METHODS: We analyzed the Korean National Health Insurance Service-Health Screening Cohort data from 2002 to 2015 where Korean residents underwent biennial health checkups. Participants were classified into four groups according to their obesity status (body mass index [BMI] ≥25 kg/m2) before and after turning 50 years old: maintaining normal (MN), becoming obese (BO), becoming normal (BN), and maintaining obese (MO). Cox proportional hazards regression model was used to estimate the risk of T2DM factoring in the covariates age, sex, BMI, presence of impaired fasting glucose or hypertension, family history of diabetes, and smoking status. RESULTS: A total of 118,438 participants (mean age, 52.5±1.1 years; men, 45.2%) were prospectively evaluated for incident T2DM. A total of 7,339 (6.2%) participants were diagnosed with T2DM during a follow-up period of 4.8±2.6 years. Incidence rates of T2DM per 1,000 person-year were 9.20 in MN, 14.81 in BO, 14.42 in BN, 21.38 in MO. After factoring in covariates, participants in the groups BN (adjusted hazard ratio [aHR], 1.15; 95% confidence interval [CI], 1.04 to 1.27) and MO (aHR, 1.14; 95% CI, 1.06 to 1.24) were at increased risk of developing T2DM compared to MN, whereas BO (hazard ratio, 1.06; 95% CI, 0.96 to 1.17) was not. CONCLUSION: Having been obese before 50 years old increased the risk of developing T2DM in the future, but becoming obese after 50 did not. Therefore, it is important to maintain normal weight from early adulthood to prevent future metabolic perturbations.

Association of Dexmedetomidine With New-Onset Atrial Fibrillation in Patients With Critical Illness.

Importance: Dexmedetomidine is a widely used sedative in the intensive care unit (ICU) and has unique properties that may be associated with reduced occurrence of new-onset atrial fibrillation (NOAF). Objective: To investigate whether the use of dexmedetomidine is associated with the incidence of NOAF in patients with critical illness. Design, Setting, and Participants: This propensity score-matched cohort study was conducted using the Medical Information Mart for Intensive Care-IV database, which includes records of patients admitted to the ICU at Beth Israel Deaconess Medical Center in Boston dating 2008 through 2019. Included patients were those aged 18 years or older and hospitalized in the ICU. Data were analyzed from March through May 2022. Exposure: Patients were divided into 2 groups according to dexmedetomidine exposure: those who received dexmedetomidine within 48 hours after ICU admission (dexmedetomidine group) and those who never received dexmedetomidine (no dexmedetomidine group). Main Outcomes and Measures: The primary outcome was the occurrence of NOAF within 7 days of ICU admission, as defined by the nurse-recorded rhythm status. Secondary outcomes were ICU length of stay, hospital length of stay, and in-hospital mortality. Results: This study included 22 237 patients before matching (mean [SD] age, 65.9 [16.7] years; 12 350 male patients [55.5%]). After 1:3 propensity score matching, the cohort included 8015 patients (mean [SD] age, 61.0 [17.1] years; 5240 males [65.4%]), among whom 2106 and 5909 patients were in the dexmedetomidine and no dexmedetomidine groups, respectively. Use of dexmedetomidine was associated with a decreased risk of NOAF (371 patients [17.6%] vs 1323 patients [22.4%]; hazard ratio, 0.80; 95% CI, 0.71-0.90). Although patients in the dexmedetomidine group had longer median (IQR) length of stays in the ICU (4.0 [2.7-6.9] days vs 3.5 [2.5-5.9] days; P < .001) and hospital (10.0 [6.6-16.3] days vs 8.8 [5.9-14.0] days; P < .001), dexmedetomidine was associated with decreased risk of in-hospital mortality (132 deaths [6.3%] vs 758 deaths [12.8%]; hazard ratio, 0.43; 95% CI, 0.36-0.52). Conclusions and Relevance: This study found that dexmedetomidine was associated with decreased risk of NOAF in patients with critical illness, suggesting that it may be necessary and warranted to evaluate this association in future clinical trials.

Risk of fracture in neuromyelitis optica spectrum disorder and multiple sclerosis: a nationwide cohort study in South Korea.

PURPOSE: Interest in fractures in patients with multiple sclerosis (MS) and neuromyelitis optica spectrum disorder (NMOSD) has considerably increased in the last decade. However, few studies have compared the incidence of fractures between patients with MS and NMOSD using a nationwide database. This study aimed to evaluate the differences in the risk of fracture between patients with NMOSD and MS compared to that in healthy controls using cohort data from a Korean nationwide database. METHODS: In this retrospective cohort study, data from the National Health Insurance Service (NHIS) database from January 2010 to December 2017 were analyzed. A total of 1,217/1,329 patients with MS/NMOSD free of fractures at the index date were included. Matched controls were selected based on age, sex, and the presence of hypertension, diabetes mellitus, and dyslipidemia. The mean follow-up durations after the index date were 4.40/4.08 years for patients with MS/NMOSD and 4.73/4.28 for their matched controls. RESULTS: The adjusted hazard ratios (aHRs) with 95% confidence intervals of any, hip, and vertebral fractures were 1.81 (1.43-2.28), 3.36 (1.81-6.24), and 2.01 (1.42-2.99) times higher for patients with MS than for controls, respectively, and they were 1.85 (1.47-2.34), 3.82 (2.05-7.11), and 2.84 (1.92-4.21) times higher for patients with NMOSD than for controls, respectively. No significant differences were observed in the incidence of fractures between the MS and NMOSD groups. Patients with MS/NMOSD had a 1.8-fold higher risk of fracture than matched controls, and the risk of hip fracture was especially high (3- to 4-fold higher). CONCLUSIONS: Clinicians need to regularly assess patients with MS/NMOSD for the risk of fractures and take preventative measures to reduce it.

Impact of Sleep Duration on Decline in Kidney Function in Adult Patients with Hypertension: A Community-Based Prospective Cohort Study.

BACKGROUND: Sleep duration is associated with various health conditions, including chronic kidney disease. However, the association between sleep duration and decline in kidney function in the South Korean population remains unclear. We aimed to investigate the impact of sleep duration on kidney function decline in adult patients with hypertension. METHODS: This cohort study was performed using data obtained from the Korean Genome and Epidemiology Study; 2,837 patients with hypertension who initially had normal kidney function were included. Glomerular filtration rates (GFRs) were estimated at baseline and throughout the 16 years of follow-up. A person was considered to have a decline in kidney function if they had a GFR <60 mL/min/1.73 m2. Sleep duration data were obtained through interviewer-assisted questionnaires. Sleep durations were classified as short (<6 hours), normal (≥6 hours but <9 hours), and long (≥9 hours). The Cox proportional hazards model was applied, with adjustments for covariates. RESULTS: After adjusting for covariates, sleep duration was not associated with a decline in kidney function. However, among men with poorly controlled hypertension at baseline, compared to men with normal sleep durations, men with sleep durations <6 hours had a significantly higher risk of kidney function decline (hazard ratio, 1.56; 95% confidence interval, 1.02-2.36). CONCLUSION: Short sleep duration did not seem to be associated with an increased risk of decline in kidney function; however, it may be a risk factor for the decline in kidney function in men with poorly controlled hypertension.

Examining the adoption and implementation of behavioral electronic health records by healthcare professionals based on the clinical adoption framework.

BACKGROUND: While various quantitative studies based on the Unified Theory of Acceptance and Use of Technology (UTAUT) and Technology Acceptance Models (TAM) exist in the general medical sectors, just a few have been conducted in the behavioral sector; they have all been qualitative interview-based studies. OBJECTIVE: The purpose of this study is to assess the adoption dimensions of a behavioral electronic health record (EHR) system for behavioral clinical professionals using a modified clinical adoption (CA) research model that incorporates a variety of micro, meso, and macro level factors. METHODS: A questionnaire survey with quantitative analysis approach was used via purposive sampling method. We modified the existing CA framework to be suitable for evaluating the adoption of an EHR system by behavioral clinical professionals. We designed and verified questionnaires that fit into the dimensions of the CA framework. The survey was performed in five US behavioral hospitals, and the adoption factors were analyzed using a structural equation analysis. RESULTS: We derived a total of seven dimensions, omitting those determined to be unsuitable for behavioral clinical specialists to respond to. We polled 409 behavioral clinical experts from five hospitals. As a result, the ease of use and organizational support had a substantial impact on the use of the behavioral EHR system. Although the findings were not statistically significant, information and service quality did appear to have an effect on the system's ease of use. The primary reported benefit of behavioral EHR system adoption was the capacity to swiftly locate information, work efficiently, and access patient information via a mobile app, which resulted in more time for better care. The primary downside, on the other hand, was an unhealthy reliance on the EHR system. CONCLUSIONS: We demonstrated in this study that the CA framework can be a useful tool for evaluating organizational and social elements in addition to the EHR system's system features. Not only the EHR system's simplicity of use, but also organizational support, should be considered for the effective implementation of the behavioral EHR system. TRIAL REGISTRATION: The study was approved by the Institutional Review Board of Seoul National University Bundang Hospital (IRB No.: B-1904-534-301).

Development and implementation of patient-level prediction models of end-stage renal disease for type 2 diabetes patients using fast healthcare interoperability resources.

This study aimed to develop a model to predict the 5-year risk of developing end-stage renal disease (ESRD) in patients with type 2 diabetes mellitus (T2DM) using machine learning (ML). It also aimed to implement the developed algorithms into electronic medical records (EMR) system using Health Level Seven (HL7) Fast Healthcare Interoperability Resources (FHIR). The final dataset used for modeling included 19,159 patients. The medical data were engineered to generate various types of features that were input into the various ML classifiers. The classifier with the best performance was XGBoost, with an area under the receiver operator characteristics curve (AUROC) of 0.95 and area under the precision recall curve (AUPRC) of 0.79 using three-fold cross-validation, compared to other models such as logistic regression, random forest, and support vector machine (AUROC range, 0.929-0.943; AUPRC 0.765-0.792). Serum creatinine, serum albumin, the urine albumin-to-creatinine ratio, Charlson comorbidity index, estimated GFR, and medication days of insulin were features that were ranked high for the ESRD risk prediction. The algorithm was implemented in the EMR system using HL7 FHIR through an ML-dedicated server that preprocessed unstructured data and trained updated data.

Changes in Diagnostic Performance of Thyroid Cancer Screening before and after the Korean Thyroid Imaging Reporting and Data System Revision.

BACKGROUND: Since the era of "thyroid cancer epidemic," many Korean academic societies discouraged the use of ultrasonography in healthy individuals and revised the Korean Thyroid Imaging Reporting and Data System to address the overscreening and overdiagnosis issues. This study aimed to evaluate the change in the diagnostic effectiveness of thyroid cancer screening over the last decade. METHODS: This single-center, retrospective observational study analyzed the data of 125,962 thyroid nodules obtained during cancer screening at the health promotion center of Seoul National University Bundang Hospital from 2010 to 2019. Only 327 thyroid cancer cases pathologically confirmed by fine-needle aspiration (FNA) were included in the study. The strength of the association between the number of FNA and (1) the number of thyroid cancer diagnoses, (2) the positive predictive values (PPVs), and (3) the difference in PPV from the previous year were evaluated using Pearson's correlation analysis. RESULTS: The number of thyroid FNA biopsies as well as the thyroid cancer diagnoses decreased from 2010 to 2019 (166 to 48 [-71.1%] vs. 43 to 22 [-48.8%]). The PPV of FNA biopsies increased from 25.9% to 45.8% (+76.8%) and was negatively correlated with the number of FNA biopsies performed (R=-0.87, P<0.001). The difference in PPV from the previous year increased similarly but without statistical significance (R=-0.59, P=0.09). CONCLUSION: The diagnostic efficiency of thyroid cancer screening has increased over the last decade, as evidenced by the increasing PPV of FNA biopsies.

Risk of Parkinson's disease in multiple sclerosis and neuromyelitis optica spectrum disorder: a nationwide cohort study in South Korea.

BACKGROUND: Neurodegeneration is associated with pathogenesis of both multiple sclerosis (MS) and neuromyelitis optica (NMOSD). Parkinson's disease (PD) is a representative neurodegenerative disease, however, whether MS or NMOSD is associated with risk of PD is not known. METHODS: MS and NMOSD cohorts were collected from the Korean National Health Insurance Service between 1 January 2010 and 31 December 2017, using International Classification of Diseases 10th revision diagnosis codes and information in the Rare Intractable Disease management programme. The PD incidence rate that occurred after a 1-year lag period was calculated and compared with that of a control cohort matched for age, sex, hypertension, diabetes and dyslipidaemia in a 1:5 ratio. RESULTS: The incidence rates of PD in patients with MS and NMOSD were 3.38 and 1.27 per 1000 person-years, respectively, and were higher than that of their matched control groups. The adjusted HR of PD was 7.73 (95% CI, 3.87 to 15.47) in patients with MS and 2.61 (95% CI, 1.13 to 6.02) in patients with NMOSD compared with matched controls. In both patients with MS and NMOSD, there were no significant differences in relative risk when stratified by sex, age, diabetes, hypertension and dyslipidaemia. CONCLUSION: The PD risk was higher in patients with MS and NMOSD compared with healthy controls and was particularly high in patients with MS. Further investigations should be performed to determine the pathophysiology and occurrence of PD in patients with MS and NMOSD.

Machine Learning-Based COVID-19 Patients Triage Algorithm Using Patient-Generated Health Data from Nationwide Multicenter Database.

INTRODUCTION: A prompt severity assessment model of patients with confirmed infectious diseases could enable efficient diagnosis while alleviating burden on the medical system. This study aims to develop a SARS-CoV-2 severity assessment model and establish a medical system that allows patients to check the severity of their cases and informs them to visit the appropriate clinic center on the basis of past treatment data of other patients with similar severity levels. METHODS: This paper provides the development processes of a severity assessment model using machine learning techniques and its application on SARS-CoV-2-infected patients. The proposed model is trained on a nationwide data set provided by a Korean government agency and only requires patients' basic personal data, allowing them to judge the severity of their own cases. After modeling, the boosting-based decision tree model was selected as the classifier while mortality rate was interpreted as the probability score. The data set was collected from all Korean citizens with confirmed COVID-19 between February 2020 and July 2021 (N = 149,471). RESULTS: The experiments achieved high model performance with an approximate precision of 0.923 and area under the curve of receiver operating characteristic (AUROC) score of 0.950 [95% tolerance interval (TI) 0.940-0.958, 95% confidence interval (CI) 0.949-0.950]. Moreover, our experiments identified the most important variables affecting the severity in the model via sensitivity analysis. CONCLUSION: A prompt severity assessment model for managing infectious people has been attained through using a nationwide data set. It has demonstrated its superior performance by surpassing that of conventional risk assessments. With the model's high performance and easily accessible features, the triage algorithm is expected to be particularly useful when patients monitor their health status by themselves through smartphone applications.

Recent trends of healthcare information and communication technologies in pediatrics: a systematic review.

As information communication technology (ICT) has advanced, the healthcare industry has embraced it to reduce medical costs, improve health outcomes, and increase patient satisfaction. Healthcare ICT revolutionizes pediatric healthcare. This study aimed to categorize and synthesize findings from the literature regarding the application of ICT in pediatric patients. This systematic review is based on a comprehensive search of Embase, MEDLINE, and Google Scholar. Study selection and coding were performed independently by 2 researchers, followed by narrative categorization. To reflect current trends in ICT for pediatrics, we adopted the Hype cycle technology classification developed by the advisory and information technology firm, Gartner, and the classification of digital health interventions by the World Health Organization. This study included a total of 135 studies. The analysis revealed 7 main types of ICT for pediatrics: (1) telehealth (39 papers), (2) precision medicine (2 papers), (3) automated decision support systems (17 papers), (4) electronic health records (7 papers), (5) patient portals (7 papers), (6) artificial intelligence (AI) (39 papers), and (7) mobile and wearable technologies (20 papers). In particular, we consistently found references to ICT for pediatrics as well as changing and improving healthcare for children. Further studies are required to determine how we can improve ICT productivity for pediatrics, particularly through AI. This study's results will help healthcare delivery organizations and technology companies consider the future direction of pediatric healthcare.