Search

Background Many studies have evaluated stroke using claims data; most of these studies have defined ischemic stroke by using an operational definition following the rule-based method. Rule-based methods tend to overestimate the number of patients with ischemic stroke. Objective We aimed to identify an appropriate algorithm for phenotyping stroke by applying machine learning (ML) techniques to analyze the claims data. Methods We obtained the data from the Korean National Health Insurance Service database, which is linked to the Ilsan Hospital database (n = 30,897). The performance of prediction models (extreme gradient boosting [XGBoost] or long short-term memory [LSTM]) was evaluated using the area under the receiver operating characteristic curve (AUROC), the area under precision-recall curve (AUPRC), and calibration curve. Results In total, 30,897 patients were enrolled in this study, 3,145 of whom (10.18%) had ischemic stroke. XGBoost, a tree-based ML technique, had the AUROC was 93.63% and AUPRC was 64.05%. LSTM showed results similar to those of the rule-based method. The F1 score was 70.01%, while the AUROC was 97.10% and AUPRC was 85.70%, which was the highest. Conclusions We proposed recurrent neural network based deep learning techniques to improve stroke phenotyping. We anticipate the ability to produce rapid and accurate results.

BACKGROUND In South Korea, many studies have evaluated stroke using claims data; most of these studies have defined ischemic stroke by using an operational definition following the rule-based method, assigning the I63 diagnosis code for admission, and incorporating the imaging or drug claims data. Rule-based methods tend to overestimate the number of patients with ischemic stroke. OBJECTIVE We aimed to identify an appropriate algorithm for phenotyping stroke by applying machine learning (ML) techniques to analyze the claims data. METHODS We obtained the data from the Korean National Health Insurance Service database, which is linked to the National Health Insurance Ilsan Hospital database (n=30,897). The ML techniques were constructed based on the results of a previous chart review (gold standard), and the demographic variables such as qualifications, examination/screening, medical utilization, and total medical costs were used as features. The performance of prediction models (extreme gradient boosting [XGBoost] or long short-term memory [LSTM]) was evaluated using the area under the receiver operating characteristic curve (AUROC), the area under precision-recall curve (AUPRC), and calibration curve. RESULTS In total, 30,897 patients were enrolled in this study, 3,145 of whom (10.18%) had ischemic stroke. XGBoost, a tree-based ML technique, had the AUROC was 93.63% and AUPRC was 64.05%. LSTM, such as neural network based deep learning methods, showed results similar to those of the rule-based method. The F1 score was 70.01%, while the AUROC was 97.10% and AUPRC was 85.70%, which was the highest. CONCLUSIONS We proposed recurrent neural network based deep learning techniques to improve stroke phenotyping. We anticipate the ability to produce rapid and accurate results by utilizing the National Health Information Database.

Background This study aimed to examine the association between familial aggregation of chronic kidney disease (CKD) and risk of CKD development and its progression. Methods This nationwide family study comprised 881,453 cases with newly diagnosed CKD between 2004 and 2017 and 881,453 controls without CKD matched by age and sex, using data from the Korean National Health Insurance Service with linkage to the family tree database. The risks of CKD development and disease progression, defined as an incident end-stage renal disease (ESRD), were evaluated. Results The presence of any affected family member with CKD was associated with a significantly higher risk of CKD with adjusted ORs (95% CI) of 1.42 (1.38–1.45), 1.50 (1.46–1.55), 1.70 (1.64–1.77), and 1.30 (1.27–1.33) for individuals with affected parents, offspring, siblings, and spouses, respectively. In Cox models conducted on patients with predialysis CKD, risk of incident ESRD was significantly higher in those with affected family members with ESRD. The corresponding HRs (95% CI) were 1.10 (1.05–1.15), 1.38 (1.32–1.46), 1.57 (1.49–1.65), and 1.14 (1.08–1.19) for individuals listed above, respectively. Conclusions Familial aggregation of CKD was strongly associated with a higher risk of CKD development and disease progression to ESRD.

Longitudinal studies of the association between sex and adverse clinical outcomes in patients with chronic kidney disease (CKD) are scarce. We assessed whether major outcomes may differ by sex among CKD patients. We analyzed a total of 1780 participants with non-dialysis CKD G1-5 from the KoreaN cohort study for Outcome in patients with Chronic Kidney Disease (KNOW-CKD). The primary outcome was a composite of non-fatal cardiovascular events or all-cause mortality. Secondary outcomes included fatal and non-fatal cardiovascular events, all-cause mortality, and a composite kidney outcome of ≥ 50% decline in estimated glomerular filtration rate from baseline or the onset of end-stage kidney disease. There were 1088 (61%) men and 692 (39%) women in the study cohort. The proportion of smokers was significantly higher in men (24% vs. 3%). During 8430 person-years of follow-up, 201 primary outcome events occurred: 144 (13%) in men and 57 (8%) in women, with corresponding incidence rates of 2.9 and 1.7 per 100 person-years, respectively. In multivariable Cox models, men were associated with a 1.58-fold (95% CI 1.06–2.35) higher risk of composite outcome. Propensity score matching analysis revealed similar findings (HR 1.81; 95% CI 1.14–2.91). Risk of all-cause mortality was significantly higher in men of the matched cohort. However, there was no difference in the risk of CKD progression. In the subgroup with coronary artery calcium (CAC) measurements, men had a higher likelihood of CAC progression. In Korean CKD patients, men were more likely to experience adverse cardiovascular events and death than women.

Background Most women with systemic lupus erythematosus (SLE) are diagnosed with the disease in their reproductive years, but the incidence and prevalence of SLE among women of childbearing age have not been studied. The objective of this study was to estimate the incidence and prevalence of SLE among the Korean women of childbearing age. Methods Women aged 20 to 44 years with SLE were identified from National Health Insurance Service – National Health Information Database (2009-2016), which contain health information of approximately 97% of the Korean population. SLE was defined by International Classification of Diseases, 10th revision code, M32. Incidence and prevalence were calculated per 100,000 person-years and stratified by year and age. Results A total of 12,756 women with SLE were identified. The incidence of SLE from 2011 to 2016 among women in childbearing years was 8.18/100,000 person-years (95% CI 7.94–8.43), with the highest incidence in 2016 (8.56/100,000 person-years, 95% CI 7.95–9.17) and the lowest incidence in 2012 (7.85/100,000 person-years, 95% CI 7.28–8.42). The prevalence of SLE from 2009 to 2016 among women in childbearing years was 77.07/100,000 person-years (95% CI 75.76–78.39), with the highest prevalence in 2014 (79.47/100,000 person-years, 95% CI 77.64–81.30) and the lowest in 2010 (74.19/100,000 person-years, 95% CI 72.45–75.93). The peak age for SLE incidence was between 25–39 years, and lower incidence was seen in the early (20–24 years) and late (40–44 years) childbearing age periods. There was an increasing trend in prevalence according to age in women of childbearing age, with the highest prevalence occurring in the 40–44 age group. Conclusions The risk and burden of SLE are high among women during their childbearing years. This calls for special attention to this particular population group when allocating health resources.

A chronic disease management program was implemented in April 2012 to lower out-of-pocket costs for repeat visits to the same clinic. The aim of this study was to investigate the association between participating in this program and the onset of complications among patients with hypertension using whole-nation claims data.We used National Health Insurance Service data (2011-2018) and patients with newly detected hypertension from 2012 to 2014 were selected. Chronic disease management program reduces the out-of-pocket expenses of consultation fee from 30% to 20% when patients enroll in this program by agreeing to visit the same clinic for the treatment of hypertension or diabetes. As the dependent variable, acute myocardial infarction (MI), stroke, chronic kidney disease (CKD), and heart failure (HF) were selected. For analysis, cox proportional hazards model was used.Total participants were 827,577, among which 102,831(12.6%) subjects participated in the chronic disease management. Participants of the chronic disease management program were more likely to show lower hazard ratios (HRs) than those of non-participants in terms of all complications (MI: HR, 0.75; 95% confidence interval [CI], 0.68-0.82; stroke: HR, 0.75; 95% CI, 0.72-0.78; CKD: HR, 0.90; 95% CI, 0.85-0.96; HF: HR, 0.56; 95% CI, 0.52-0.61).The results showed that participants of the chronic disease management program were less likely to have hypertension complications compared to non-participants. Enhancing the participation rate may be related to better outcomes and reducing medical expenses among patients with chronic diseases.

We investigated the distribution of medical service uses for Bell's palsy and Ramsay Hunt syndrome between medicine and traditional Korean medicine using the National Health Insurance Service National Sample Cohort data of Korea from 2006 to 2015. Patients were identified with diagnostic codes and medication or treatment claim codes. For Bell's palsy, there were 5,970 (68.8%) patients who used traditional Korean medical service only, whereas for Ramsay Hunt syndrome, there were 749 (93.6%) patients who used medical service only. The proportion of traditional Korean medical service use was higher than that of medical service use in patients with Bell's palsy, while the opposite was found in patients with Ramsay Hunt syndrome.

Clinical practice guidelines recommend a target blood pressure (BP)130/80 mm Hg to reduce cardiovascular risk. However, the optimal BP to prevent chronic kidney disease (CKD) is unknown.Population-based retrospective cohort study.10.5 million adults who participated in the National Health Insurance Service National Health Checkup Program in South Korea between 2009 and 2015 and had an estimated glomerular filtration rate (GFR) ≥ 60 mL/min/1.73 mBaseline and time-updated systolic BP (SBP) as a continuous variable and categorized as110, 110 to 119, 120 to 129, 130 to 139, or≥140 mm Hg.Incident CKD GFR categories 3 to 5 (CKD G3-G5), defined as de novo development of estimated GFR60 mL/min/1.73 mCox proportional hazards regression for baseline BP and marginal structural analysis for time-updated BP.During 49,169,311 person-years of follow-up, incident CKD G3-G5 developed in 172,423 (1.64%) individuals with a crude event rate of 3.51 (95% CI, 3.49-3.52) per 1,000 person-years. Compared to a baseline SBP of 120 to 129 mm Hg, HRs for incident CKD G3-G5 for the110, 110 to 119, 130 to 139, and≥140 mm Hg categories were 0.84 (95% CI, 0.82-0.85), 0.92 (95% CI, 0.91-0.94), 1.11 (95% CI, 1.09-1.12), and 1.30 (95% CI, 1.28-1.31), respectively. For time-updated SBPs, corresponding HRs were 0.57 (95% CI, 0.56-0.59), 0.79 (95% CI, 0.78-0.80), 1.58 (95% CI, 1.55-1.60), and 2.49 (95% CI, 2.45-2.53), respectively. Treated as a continuous exposure, each 10-mm Hg higher SBP was associated with 35% higher risk for incident CKD G3-G5 (95% CI, 1.35-1.36).Use of International Classification of Diseases codes to assess comorbid condition burden; residual confounding, and potential selection bias cannot be excluded.In this large national cohort study, higher SBPs were associated with higher risk for incident CKD G3-G5. These findings support evaluation of SBP-lowering strategies to reduce the development of CKD.

Background Few data on the association between height and the risk of sudden cardiac death (SCD) in Asian populations are available. Objective The purpose of this study was to assess the risk of SCD as related to height both in a representative Korean population and in specific subgroups. Methods This is a retrospective cohort study of 410,119 Koreans age ≥20 years from the Korean National Health Insurance Service–National Sample Cohort, who underwent a national health examination. SCD cases were adjudicated based on information within the claims database. Cox proportional hazard models were applied to estimate hazard ratios and 95% confidence intervals for the association between height and SCD. Potential mediators included demographic factors, health-related habits, and specific cardiovascular comorbidities. Results During an 8.45-year follow-up period, a total of 1341 SCDs occurred. Various factors, including short stature, were identified as risk factors for SCD. Multivariable regression analysis revealed that a 10-cm increase in height was associated with a 14% decreased risk for SCD. This relationship remained significant among the elderly, postmenopausal women, and individuals without cardiovascular disease. Conclusion Our results indicate that short stature is a significant risk factor for SCD in a Korean population, thus supporting previously published findings correlating height to SCD risk in non-Asian populations.

Objective To examine the association between income level and incident chronic kidney disease (CKD) in adults with normal baseline kidney function. Patient and Methods We studied the association between income level categorized into deciles and incident CKD in a national cohort comprised of 7,405,715 adults who underwent National Health Insurance Service health examinations during January 1, 2009, to December 31, 2015, with baseline estimated glomerular filtration rates (eGFRs) ≥60 mL/min/1.73 m2. Incident CKD was defined as de novo development of eGFR Results During a median follow-up of 4.8 years, there were 122,032 of 7,405,715 (1.65%) and 55,779 of 7,405,715 (0.75%) incident CKD events based on model 1 and 2 definitions, respectively. Compared with income levels in the sixth decile, there was an inverse association between lower income level and higher risk for CKD up to the fourth decile, above which no additional reduction (model 1) or slightly higher risk for CKD (model 2) was observed at higher income levels. The multivariable-adjusted hazard ratios from the lowest to fourth deciles were 1.30 (95% CI, 1.26-1.33), 1.16 (95% CI, 1.13-1.19), 1.07 (95% CI, 1.05-1.10), and 1.06 (95% CI, 1.03-1.09) in model 1 and 1.32 (95% CI, 1.27-1.37), 1.18 (95% CI, 1.14-1.22), 1.08 (95% CI, 1.04-1.13), and 1.05 (95% CI, 1.01-1.09) in model 2, respectively. These associations persisted across various subgroups of age, sex, and comorbidity status. Conclusion In this large nationwide cohort, lower income levels were associated with higher risk for incident CKD.

The associations between hypertension, diabetes, and dyslipidemia with Bell’s palsy have been controversial and only a few studies have assessed risk factors for Bell’s palsy based on population-based data. The aim of the present study was to evaluate whether sociodemographic factors such as sex, age, residence, household income, and metabolic diseases such as hypertension, diabetes, and dyslipidemia were risk factors for Bell’s palsy using the National Health Insurance Service National Sample Cohort data of Korea. Patients who visited an outpatient clinic twice or more or had one or more admission and received steroid medication under the International Classification of Diseases diagnostic codes for Bell’s palsy from 2006 to 2015 were defined as patients with Bell’s palsy in this study. The associations between sociodemographic factors and metabolic diseases to Bell’s palsy were analyzed with univariate and multivariate Cox proportional hazard regression models. There were 2708 patients with Bell’s palsy recorded from 2006 to 2015. Male sex, advanced age, residence in a location other than the capital and metropolitan cities, hypertension, and diabetes were significant risk factors for Bell’s palsy. This study is significant for patients and providers because we analyzed the relationships using a population-based database over a long-term follow-up period.

Background To determine the risk of pregnancy complications and adverse offspring outcomes in Korean women with rheumatic diseases (RDs). Methods Women aged 20–44 years with pregnancies ending in delivery were identified from the National Health Insurance Service-National Health Information Database (2009–2016). Women with RD including systemic lupus erythematosus (SLE), seropositive rheumatoid arthritis (SPRA), and ankylosing spondylitis (AS) (n = 4,284) were age-matched with controls (n = 26,023). Outcome variables included threatened abortion (TA), preterm birth (PB), preeclampsia/eclampsia (PE/E), intrauterine growth retardation (IGR), urinary tract infection, low birth weight (LBW) offsprings, and offspring death within 1 year of birth. Results Women with RDs had increased risks for cesarean section delivery (odds ratio [OR], 1.5; 95% confidence interval [CI], 1.4–1.6), TA (OR, 1.4; 95% CI, 1.2–1.5), PB (OR, 2.4; 95% CI, 1.9–3.2), PE/E (OR, 4.4; 95% CI, 3.3–5.9), and IGR (OR, 2.4; 95% CI, 2.0–3.1) than the controls. The risk of pregnancy complications was increased in SLE and SPRA pregnancies but not in AS pregnancies. Offsprings of women with RDs had an increased risk of LBW (OR, 4.0; 95% CI, 3.2–4.9). The offspring mortality rate within 1 year of birth was higher in women with RDs (6.2/10,000 persons) than in the controls (4.9/10,000 persons). Conclusion Women with RDs are at a risk of developing pregnancy complications, and the risk of LBW offsprings and offspring death within 1 year of birth is increased in these women. Therefore, this population requires special attention during their childbearing years., Graphical Abstract

Objective Many thyroid cancer patients are exposed to long-term thyroid-stimulating hormone (TSH) suppression, often as lifetime treatment, and are consequently at risk for cardiovascular disease. We investigated the incidence of coronary heart disease (CHD) and ischemic stroke among thyroid cancer patients compared with matched control subjects. Design Retrospective cohort study. Methods A total of 182 419 subjects who received thyroidectomy for thyroid cancer during 2004–2012 were selected from the Korean National Health Insurance data, which cover approximately 97% of the entire Korean population. Propensity score matching was used to select non-cancer controls. Cox proportional hazards regression analysis was used to determine relative risk of coronary heart disease and ischemic stroke. Mean follow-up was 4.32 years. Results Thyroid cancer patients had elevated risk for CHD and ischemic stroke with hazard ratio (HR) of 1.15 (95% confidence interval (CI): 1.10–1.22) and 1.15 (1.09–1.22), respectively. This risk was increased in those who took a higher dosage of levothyroxine (HR: 1.47, 95% CI: 1.34–1.60 for CHD and HR: 1.56, 95% CI: 1.42–1.72 for ischemic stroke among those who took ≥170 μg/day levothyroxine). Although risk of atrial fibrillation was dose-dependently associated with levothyroxine dosage, it represents only a small proportion of ischemic stroke incidence (4.4%, 128/2914). Conclusions The risk for CHD and ischemic stroke was higher in thyroid cancer patients who received thyroidectomy, and the dosage of levothyroxine administered appears to play a major role. Greater caution is suggested for the screening and treatment of thyroid cancer and subsequent TSH suppression therapy, as well as proper management for cardiovascular disease prevention.

Background Almost all Koreans are covered by mandatory national health insurance and are required to undergo health screening at least once every 2 years. We aimed to develop a machine learning model to predict the risk of developing hepatocellular carcinoma (HCC) based on the screening results and insurance claim data. Methods The National Health Insurance Service-National Health Screening database was used for this study (NHIS-2020-2-146). Our study cohort consisted of 417,346 health screening examinees between 2004 and 2007 without cancer history, which was split into training and test cohorts by the examination date, before or after 2005. Robust predictors were selected using Cox proportional hazard regression with 1000 different bootstrapped datasets. Random forest and extreme gradient boosting algorithms were used to develop a prediction model for the 9-year risk of HCC development after screening. After optimizing a prediction model via cross validation in the training cohort, the model was validated in the test cohort. Results Of the total examinees, 0.5% (1799/331,694) and 0.4% (390/85,652) in the training cohort and the test cohort were diagnosed with HCC, respectively. Of the selected predictors, older age, male sex, obesity, abnormal liver function tests, the family history of chronic liver disease, and underlying chronic liver disease, chronic hepatitis virus or human immunodeficiency virus infection, and diabetes mellitus were associated with increased risk, whereas higher income, elevated total cholesterol, and underlying dyslipidemia or schizophrenic/delusional disorders were associated with decreased risk of HCC development (p C-index, AUC, and Brier skill score were 0.857, 0.873, and 0.078, respectively. Conclusions Machine learning-based model could be used to predict the risk of HCC development based on the health screening examination results and claim data.

The rapid spread of COVID-19 has resulted in the shortage of medical resources, which necessitates accurate prognosis prediction to triage patients effectively. This study used the nationwide cohort of South Korea to develop a machine learning model to predict prognosis based on sociodemographic and medical information. Of 10,237 COVID-19 patients, 228 (2.2%) died, 7,772 (75.9%) recovered, and 2,237 (21.9%) were still in isolation or being treated at the last follow-up (April 16, 2020). The Cox proportional hazards regression analysis revealed that age > 70, male sex, moderate or severe disability, the presence of symptoms, nursing home residence, and comorbidities of diabetes mellitus (DM), chronic lung disease, or asthma were significantly associated with increased risk of mortality (p ≤ 0.047). For machine learning, the least absolute shrinkage and selection operator (LASSO), linear support vector machine (SVM), SVM with radial basis function kernel, random forest (RF), and k-nearest neighbors were tested. In prediction of mortality, LASSO and linear SVM demonstrated high sensitivities (90.7% [95% confidence interval: 83.3, 97.3] and 92.0% [85.9, 98.1], respectively) and specificities (91.4% [90.3, 92.5] and 91.8%, [90.7, 92.9], respectively) while maintaining high specificities >90%, as well as high area under the receiver operating characteristics curves (0.963 [0.946, 0.979] and 0.962 [0.945, 0.979], respectively). The most significant predictors for LASSO included old age and preexisting DM or cancer; for RF they were old age, infection route (cluster infection or infection from personal contact), and underlying hypertension. The proposed prediction model may be helpful for the quick triage of patients without having to wait for the results of additional tests such as laboratory or radiologic studies, during a pandemic when limited medical resources must be wisely allocated without hesitation.

Background Increased body mass index (BMI) has been associated with a higher risk of severe coronavirus disease 2019 (COVID-19) infections. However, whether obesity is a risk factor for contracting COVID-19 has hardly been investigated so far. Methods We examined the association between BMI level and the risk of COVID-19 infection in a nationwide case-control study comprised of 3788 case patients confirmed to have COVID-19 between 24 January and 9 April 2020 and 15 152 controls matched by age and sex, who were aged 20 years or more and underwent National Health Insurance Service (NHIS) health examinations between 2015−2017, using data from the Korean NHIS with linkage to the Korea Centers for Disease Control and Prevention data. Our primary exposure of interest was BMI level, categorized into 4 groups: Results Of the entire 18 940 study participants, 11 755 (62.1%) were women, and the mean age of the study participants was 53.7 years (standard deviation, 13.8). In multivariable logistic regression models adjusted for sociodemographic, comorbidity, laboratory, and medication data, there was a graded association between higher BMI levels and higher risk of COVID-19 infection. Compared to normal-weight individuals, the adjusted odds ratios in the overweight and obese individuals were 1.13 (95% confidence interval [CI], 1.03–1.25) and 1.26 (95% CI, 1.15–1.39), respectively. This association was robust across age and sex subgroups. Conclusions Higher BMI levels were associated with a higher risk of contracting COVID-19.

Background: Diabetes mellitus (DM), one of the most common metabolic diseases worldwide, is associated with malignancies at many sites. The relationship between DM and bladder or kidney cancer is still controversial. We investigated the associations between DM and bladder or kidney cancer in a population-based exposure-control matched study.Method: Using the Korean National Health Insurance System sample cohort database, we identified patients who were diagnosed as having DM between January 2006 and December 2010. A control group was also selected from the database using propensity score matching to adjust for age and sex. An exposure-control matched study was conducted until December 2015. The two groups were compared for baseline characteristics and bladder or kidney cancer incidence. Cox’s proportional hazard regression analyses were performed to investigate associations between bladder or kidney cancer and each clinical variable.Results: In total, 132,570 patients with DM were identified. In addition, 132,570 patients without DM matched by age and sex (control group) were selected. The mean age was 54 years, and the male-to-female ratio was 1:1. During the follow-up period, 619 patients in the DM group and 511 in the control group were diagnosed as having bladder cancer. Of those diagnosed as having kidney cancer, 419 patients were in the DM group and 285 were in the control group. Adjusted Cox’s proportional hazard regression analysis revealed that DM is associated with bladder and kidney cancer.Conclusions: DM is associated with bladder and kidney cancer. Older men are at an increased risk for both cancers.

Background Although increased height is associated with a risk of atrial fibrillation (AF), the mechanism is not well understood. We aimed to explore whether this association varies with metabolic conditions. Methods and results We used the database from the 14-year Korea National Health Insurance Service–National Sample Cohort. The data of 368,206 adults older than 20 years who received a health check-up were analyzed to explore the association of height and AF risk. Cox proportional hazards regression models were used to compute hazard ratios (HRs) and 95% confidence intervals (CIs) for associations of height with the risk of AF. During the median follow up duration of 8.46 years, 2641 (0.72%) patients were diagnosed with AF at 3,070,724 person-years. Overall, greater height was significantly associated with AF risk (HR per 5 cm, 1.22; 95% CI, 1.03–1.05). The association did not vary with age, sex, obesity, hypertension, and diabetes. Conclusion Metabolic conditions do not affect the higher risk of AF in tall people.

Gastrectomy can lead to bone loss. Previous studies have suggested that there is an increased risk of fracture in gastric cancer survivors. However, these studies were performed without proper control groups. Therefore, we used Korean national health insurance data to compare the fracture risk in gastric cancer survivors who received gastrectomy to that of the general population.A total of 133,179 gastric cancer survivors were included and matched to noncancer controls using 1:1 propensity score matching. Cox proportional hazards regression analysis was used to determine the relative risk of fracture between the gastric cancer survivors and matched controls. We also examined the factors associated with fracture in gastric cancer survivors.Compared with the matched controls, gastric cancer survivors had an elevated risk of osteoporotic fracture (hazard ratio [HR] 1.61, 95% confidence interval [CI] 1.53-1.70), which was most prominent in patients who underwent total gastrectomy (HR 2.18, 95% CI 1.96-2.44) and adjuvant chemotherapy (HR 2.01, 95% CI 1.81-2.23). In multivariate analysis, anemia was significantly associated with increased fracture risk (aHR 1.34, 95% CI 1.13-1.59), while decrease in weight5% was not (aHR 1.06, 95% CI 0.89-1.25).Gastric cancer survivors who underwent gastrectomy had an increased osteoporotic fracture risk than did matched controls. Total gastrectomy, adjuvant chemotherapy, and anemia were associated with an even higher risk in these patients. Additional studies are needed to establish optimal strategies, such as screening for osteoporosis and preventive interventions, that will reduce fracture risk in this population.

The aim of this study was to investigate the level of satisfaction for orthodontic treatment among adult patients. In addition, the influencing host factors were monitored for their associations with satisfaction.A questionnaire was designed to measure the level of satisfaction in 10 items; overall satisfaction, tooth alignment, facial appearance, eating and chewing, confident smile and self-image, retention state, treatment duration, treatment costs, intention to recommend, and relief of previous concerns, using a 5-point Likert scale. Total satisfaction was calculated by averaging the Likert scores from the 10 items. The survey was conducted, and the results from 298 adults were evaluated.For the overall satisfaction item, 45.0% were very satisfied, and 39.9% were satisfied, resulting in a satisfaction ratio of 84.9%. Total satisfaction score was 3.9. The level of satisfaction for tooth alignment and confident smile and self-image were significantly higher than facial appearance and eating and chewing (P 0.001). Patients aged 50 and above were more satisfied than the younger ones, and men were more satisfied than women (P 0.05).Overall, adult patients were highly satisfied with orthodontic treatment. Age, sex, motivation, expected concern, and discomfort influenced the level of satisfaction.

Background and objective The purpose of this study was to build a 10-year stroke prediction model and categorize a probability of stroke using the Korean national health examination data. Then it intended to develop the algorithm to provide a personalized warning on the basis of each user's level of stroke risk and a lifestyle correction message about the stroke risk factors. Methods Subject to national health examinees in 2002–2003, the stroke prediction model identified when stroke was first diagnosed by following-up the cohort until 2013 and estimated a 10-year probability of stroke. It sorted the user's individual probability of stroke into five categories – normal, slightly high, high, risky, very risky, according to the five ranges of average probability of stroke in comparison to total population – less than 50 percentile, 50–70, 70–90, 90–99.9, more than 99.9 percentile, and constructed the personalized warning and lifestyle correction messages by each category. Results Risk factors in stroke risk model include the age, BMI, cholesterol, hypertension, diabetes, smoking status and intensity, physical activity, alcohol drinking, past history (hypertension, coronary heart disease) and family history (stroke, coronary heart disease). The AUC values of stroke risk prediction model from the external validation data set were 0.83 in men and 0.82 in women, which showed a high predictive power. The probability of stroke within 10 years for men in normal group (less than 50 percentile) was less than 3.92% and those in very risky group (top 0.01 percentile) was 66.2% and over. The women's probability of stroke within 10 years was less than 3.77% in normal group (less than 50 percentile) and 55.24% and over in very risky group. Conclusions This study developed the stroke risk prediction model and the personalized warning and the lifestyle correction message based on the national health examination data and uploaded them to the personal health record service called My Health Bank in the health information website - Health iN . By doing so, it urged medical users to strengthen the motivation of health management and induced changes in their health behaviors.

Background: Hepatic steatosis measured with controlled attenuation parameter (CAP) using transient elastography predicts metabolic syndrome in the general population. We investigated whether CAP predicted metabolic syndrome in chronic kidney disease patients. Methods: CAP was measured with transient elastography in 465 predialysis chronic kidney disease patients (mean age, 57.5 years). Results: The median CAP value was 239 (202-274) dB/m. In 195 (41.9%) patients with metabolic syndrome, diabetes mellitus was more prevalent (105 [53.8%] vs. 71 [26.3%], P < 0.001), with significantly increased urine albumin-to-creatinine ratio (184 [38-706] vs. 56 [16-408] mg/g Cr, P = 0.003), high sensitivity C-reactive protein levels (5.4 [1.4-28.2] vs. 1.7 [0.6-9.9] mg/L, P < 0.001), and CAP (248 [210-302] vs. 226 [196-259] dB/m, P < 0.001). In multiple linear regression analysis, CAP was independently related to body mass index (β = 0.742, P < 0.001), triglyceride levels (β = 2.034, P < 0.001), estimated glomerular filtration rate (β = 0.316, P = 0.001), serum albumin (β = 1.386, P < 0.001), alanine aminotransferase (β = 0.064, P = 0.029), and total bilirubin (β = -0.881, P = 0.009). In multiple logistic regression analysis, increased CAP was independently associated with increased metabolic syndrome risk (per 10 dB/m increase; odds ratio, 1.093; 95% confidence interval, 1.009-1.183; P = 0.029) even after adjusting for multiple confounding factors. Conclusion: Increased CAP measured with transient elastography significantly correlated with and could predict increased metabolic syndrome risk in chronic kidney disease patients.

Objective: We investigated the risk of sudden sensorineural hearing loss (SSNHL) in patients with autoimmune diseases compared with a control group in a population-based study using a National Health Insurance Service National Sample Cohort data from Korea. Methods:We enrolled autoimmune-disease patients who were ≥20 years of age in 2006 into the autoimmune-disease group, and selected a control group with similar demographic characteristics. We tracked the 2 groups from 2006 to 2015 and compared the proportions of patients who developed SSNHL. Results: Of 13,250 patients in the autoimmune-disease group, 145 experienced an SSNHL event (1.09%). Of the 66,250 in the control group, 484 experienced an SSNHL event (0.73%). The SSNHL risk was significantly higher in the autoimmune-disease group than in the control group. SSNHL incidence was significantly higher among patients with antiphospholipid syndrome (APS), multiple sclerosis (MS), rheumatoid arthritis (RA), and connective-tissue diseases including Sjögren syndrome and Behçet disease. In detailed stratified analyses using reinforced additional diagnostic codes, only RA patients had a significantly higher SSNHL incidence than the control group. Conclusion: The association of several autoimmune diseases with SSNHL was evaluated in this large-scale, population-based, big-data study. The risk of SSNHL was significantly higher in patients with APS, MS, RA, and connective-tissue diseases including Sjögren syndrome and Behçet disease than in patients without autoimmune diseases. SSNHL, in particular, was significantly associated with RA according to detailed analyses using reinforced additional diagnostic codes.

This study was performed to evaluate the long-term changes in mandibular width, lower facial width, and ramus angulation after intraoral vertical ramus osteotomy (IVRO) and to identify the factors influencing these changes. This retrospective study included 53 consecutive patients with mandibular prognathism who underwent IVRO with (n=33) or without (n=20) Le Fort I osteotomy. Postero-anterior cephalograms and frontal facial photographs obtained before, 1 month after, and at least 24 months after IVRO were used for measurements. A linear mixed model and paired t-tests were used to analyze temporal changes and the associated influencing factors. The mandibular width increased immediately after surgery (P0.05), but decreased continuously thereafter. The ramus angulation showed negligible change within the first month (P0.05) and decreased thereafter up to approximately 36 months. The amounts of mandibular setback and posterior impaction and the length of time postoperative influenced these changes. The lower facial width changed, although inconsistently, within 3mm over time (P0.05). In conclusion, the mandibular width increased after IVRO but seemed to normalize within approximately 3 years. The lower facial width did not reflect underlying skeletal changes. Therefore, long-term transverse changes after IVRO can be considered clinically irrelevant.

Objectives: To identify risk factors for apical root resorption (ARR) of maxillary and mandibular incisors using mathematical quantification of apical root displacement (ARD) and multiple linear mixed-effects modeling. Materials and Methods: Periapical radiographs of maxillary and mandibular incisors and lateral cephalograms of 135 adults were taken before and after orthodontic treatment. ARR was measured on the periapical radiographs, and movement of central incisors was evaluated on the superimposed pre- and posttreatment lateral cephalograms. ARD was mathematically calculated from pretreatment tooth length, inclination change, and movement of the incisal edge. Linear mixed-effects model analysis was performed to identify risk factors for ARR, and standardized coefficients (SCs) were calculated to investigate the relative contribution of the risk factors to ARR. Results: Vertical ARD showed the highest SCs for both maxillary and mandibular incisors. Horizontal ARD showed the second highest SC for mandibular incisors but was not significantly correlated with the ARR of maxillary incisors. When horizontal and vertical ARDs were included in the mixed-effects model, the use of self-ligating brackets was significantly correlated with increased ARR of mandibular incisors. Conclusions: ARD is a critical factor for ARR after orthodontic treatment. Careful monitoring of ARR is recommended for patients requiring significant ARD of incisors.

Small studies suggest improvement of cardiovascular risk profiles and reduced cardiovascular mortality after gastrectomy for gastric cancer. In our study using national health insurance data, we investigated coronary heart disease (CHD) and ischemic stroke incidence among gastric cancer patients who received gastrectomy compared with general population controls. A total of 98,936 gastric cancer patients were included in the study and matched to non-cancer controls via 1:1 propensity score matching. Competing risk regression analysis was used to determine the relative risk of CHD and ischemic stroke, with cancer mortality as the competing risk. Changes in cardiovascular risk profile before and after gastrectomy were tested, and factors associated with CHD or ischemic stroke incidence among gastric cancer patients were analyzed. Compared with the matched controls, gastric cancer patients who received gastrectomy were shown to have a decreased risk for both CHD (subdistribution hazard ratio [SHR] 0.60, 95% confidence interval [CI] 0.57–0.63) and ischemic stroke (SHR 0.72, 95% CI 0.69–0.75). Decreases in body mass index (BMI), blood pressure, blood sugar, and lipid and hemoglobin levels were marked after gastrectomy. Among gastric cancer patients, those who lost > 5% of their body weight were at lower risk of developing CHD (SHR 0.82, 95% CI 0.71–0.96) and ischemic stroke (SHR 0.85, 95% CI 0.75–0.98). CHD and ischemic stroke risk decreased after gastrectomy. The amount of weight loss and accompanying metabolic changes seemed to mediate the reduction of such risk. Reassessment of cardiovascular risk factors after gastrectomy and consideration of cardiovascular risk in the selection of treatment modality are suggested.

Background There is a general consensus that elevated serum beta-2 microglobulin (B2M) levels measured at a single time-point are significantly associated with mortality in patients on maintenance dialysis. To date, the majority of prior studies that have examined B2M-associated mortality have been conducted in prevalent hemodialysis patients with little residual renal function (RRF). However, studies in incident peritoneal dialysis (PD) patients are lacking. Moreover, changes in serum B2M levels over time have not been considered in this population. Methods We examined the association of time-updated and baseline serum B2M levels with mortality in a 10-year cohort of 725 incident PD patients who were maintained on dialysis between January 2006 and December 2011 using Cox proportional hazards regression analyses. Patients were categorized into tertiles according to B2M levels. Results During a median follow-up of 38 (interquartile range 23-64) months, 258 (35.4%) deaths occurred, including 106 (14.6%) and 86 (11.9%) deaths from cardiovascular and infectious causes, respectively. The lowest B2M tertile was associated with a higher risk of all-cause and infectious mortality compared with the middle tertile: the hazard ratios (95% confidence interval) for all-cause deaths were 2.12 (1.38-3.26) and 2.20 (0.96-5.05) in time-varying analyses and 1.52 (1.07-2.17) and 2.41 (1.19-4.88) in baseline analyses. Subgroup analyses showed that this association was particularly observed in females, older patients, those with comorbidities such as diabetes, a lower body mass index, lower albumin levels or those with higher RRF (all P for interactions Conclusions In incident PD patients, lower B2M levels were independently associated with overall and infectious mortality. These associations can be potentially modified by malnutrition, inflammation and RRF.

Background: Subcortical vascular dementia (SVaD) is one of the most common dementias, after Alzheimer's disease (AD) dementia. Few survival analyses in SVaD patients have been reported. Methods: The dates and causes of death of 146 SVaD and 725 AD patients were included. We used the Cox proportional hazards model to compare survival between SVaD and AD patients and to explore possible factors related to survival of SVaD patients. Results: The median survival time after the onset of SVaD (109 months) was shorter than that recorded for AD (152 months). The most common cause of death in SVaD was stroke (47.1%). Factors associated with shorter survival in SVaD were late onset, male sex, worse baseline cognition, absence of hypertension and a family history of stroke. Conclusions: Stroke prevention may be important in SVaD treatment because 47.1% of SVaD patients died of stroke. A family history of stroke and absence of hypertension were associated with a shorter survival in SVaD, suggesting the existence of genetic or unknown risk factors.

Long-term administration of supraphysiologic dosages of levothyroxine can have detrimental effect on the bone. We aimed to investigate fracture incidence among post-thyroidectomy thyroid cancer patients compared with a matched comparison group, and explore the association between levothyroxine dosage and fracture risk. From the Korean National Health Insurance database, virtually all thyroid cancer patients who received thyroidectomy in Korea from January 1, 2004 to December 31, 2012 were included. Matched subjects were selected by 1:1 propensity score matching. Cox proportional hazards regression analysis was used to determine relative risk of osteoporotic fracture. Of 185,956 thyroid cancer patients identified, fracture events were observed in 1096 subjects (0.56%) over a mean 4.35 years of follow-up. Compared to the matched comparison group, thyroid cancer patients had no elevated risk of osteoporotic fracture (hazard ratio [HR] 1.03; 95% confidence interval [CI], 0.94 to 1.12); however, the highest dosage group (≥170 μg/day) showed significantly higher risk (HR 1.25; 95% CI, 1.07 to 1.45), while the second quartile dosage group (115-144 μg/day) showed lower risk (HR 0.71; 95% CI, 0.59 to 0.84) compared to a matched comparison group. When the second quartile dosage group was considered as reference, increased fracture risk was observed in those who took either lower (first quartile: adjusted HR 1.31; 95% CI, 1.08 to 1.59) or higher dosage of levothyroxine (third quartile: adjusted HR 1.50; 95% CI, 1.26 to 1.79; fourth quartile: adjusted HR 1.79; 95% CI, 1.51 to 2.13). Thyroid cancer patients were more likely to be treated with osteoporosis medication (HR 1.22; 95% CI, 1.18 to 1.26) than the matched comparison group. Both high and low dosage of levothyroxine treatment was associated with a higher risk for fractures in a J-shaped dose-dependent manner in post-thyroidectomy patients. Future studies are needed to determine how to optimize thyroid-stimulating hormone (TSH) suppression and how to screen and manage fracture risk. © 2018 American Society for Bone and Mineral Research.