Osteoporosis is treatable but often overlooked in clinical practice. We aimed to construct prediction models with machine learning algorithms to serve as screening tools for osteoporosis in adults over fifty years old. Additionally, we also compared the performance of newly developed models with traditional prediction models. Data were acquired from community-dwelling participants enrolled in health checkup programs at a medical center in Taiwan. A total of 3053 men and 2929 women were included. Models were constructed for men and women separately with artificial neural network (ANN), support vector machine (SVM), random forest (RF), k-nearest neighbor (KNN), and logistic regression (LoR) to predict the presence of osteoporosis. Area under receiver operating characteristic curve (AUROC) was used to compare the performance of the models. We achieved AUROC of 0.837, 0.840, 0.843, 0.821, 0.827 in men, and 0.781, 0.807, 0.811, 0.767, 0.772 in women, for ANN, SVM, RF, KNN, and LoR models, respectively. The ANN, SVM, RF, and LoR models in men, and the ANN, SVM, and RF models in women performed significantly better than the traditional Osteoporosis Self-Assessment Tool for Asians (OSTA) model. We have demonstrated that machine learning algorithms improve the performance of screening for osteoporosis. By incorporating the models in clinical practice, patients could potentially benefit from earlier diagnosis and treatment of osteoporosis.

Objectives The cross‐sectional study aimed to analyze the association between burnout, work‐related factors, and metabolic syndrome (Mets) in nurses from several departments of a tertiary hospital in Taiwan. Exploring biomarkers could provide for prevention. Methods Demographic data were obtained through a written questionnaire and include the following information: gender, age, education level, psychosocial and work situations, such as departments, working hours, work shift, depression, and sleep time. Burnout was evaluated according to the Chinese Burnout inventory, Mets was evaluated according to the criteria of the National Cholesterol Education Program of Taiwan—Treatment Panel for Adults III (NCEP‐ATP III). Results A total of 1758 nurses participated with a median age of 35.2 years. The prevalence of burnout and Mets was 6.4% and 13.84%, respectively. The results showed that burnout induced higher risk of Mets, odds ratio (OR) 1.70 (95% confidence interval, 1.04‐3.05). Other factors, such as out‐patient nurses, seniority (4‐10 and >10 years), working hours (51‐59 h/wk), nigh shift, Brief Symptom Rating Scale‐5 (score 10‐14 and ≧15), poor self‐rated health status, and inadequate sleep time, led to higher risk of Mets. Biomarkers research showed that Glycated hemoglobin (Hba1c) was significantly associated with burnout nurses (OR = 24.72, P < .001), but thyroid‐stimulating hormone and free thyroxin were not. Conclusions Results suggested positive associations between burnout and Mets in nurses. For nurses with higher seniority, long hours of work, night shifts, poor physical and mental conditions, and poor lifestyle habits in different departments, strategies are needed to prevent burnout and Mets.

Urinary tract infection (UTI) is a common infection worldwide. Numerous studies have demonstrated risk factors for UTI in facilities and communities. In this study, we aimed to identify risk factors for UTI in home care patients. Patients who received home care for at least three months in 2017 were included. A UTI episode was defined by admission for UTI treatment, and/or a visit to an outpatient department for UTI and receiving antibiotic prescription. A total of 598 patients aged 81.9 years were included. Overall 47% (281) had at least one UTI episode. After analysis, urinary catheter indwelling was the most important risk factor (odds ratio (OR), 3.21). Underlying diseases (coronary artery disease (CAD), chronic kidney disease (CKD), diabetes mellitus (DM)) were related to UTI with OR ranging from 1.46 to 1.79. Higher Alb (albumin) (OR 0.68) and Hb (hemoglobin) (OR 0.91) were inversely related to UTI. Polypharmacy (OR 1.84) and lower Barthel index scores (OR 1.62) were also related to UTI by different degrees. In conclusion, apart from the unmodifiable factors, avoiding prolonged urinary catheter placement, unnecessary prescriptions, and keeping good nutritional status may help to prevent UTI in home care patients.

Background: This cross-sectional study aimed to analyze the association between burnout, work-related factors and metabolic syndrome (MetS) among various health-care workers (HCWs) at a tertiary hospital in Taiwan.Methods: Relevant demographic data were obtained through written questionnaires. Information about psychosocial and work conditions, including assigned department, working hours, shifts, and sleep condition, was obtained. Burnout was evaluated according to the Chinese version of Maslach Burnout Inventory–Health Services Survey. MetS was analyzed according to the criteria of the National Cholesterol Education Program of Taiwan–Treatment Panel for Adults III.Results: A total of 1,055 non-doctor/nurse and 2,078 doctor/nurse staff with a median age of 45.2 and 36.1 years participated in this study. The incidence of burnout was nearly 6.42 and 6.68% and that of MetS was 31.4 and 13.5% in non-doctor/nurse and doctor/nurse staff, respectively. The results showed that burnout induced a higher Odds ratio (OR) of MetS in the doctor/nurse group (OR = 1.96, p = 0.01). Other factors, such as night shift and seniority (>10 years), led to a higher OR of MetS, but a decreased risk was observed based on seniority (2–4 years). Further, young female participants and young doctors/nurses with burnout had a higher OR of MetS compare to other groups (OR = 2.43 and 2.32, p < 0.05).Conclusion: The study results suggested positive relationship between burnout and MetS in young female staffs and young doctors/nurses. For doctor/nurse staff with higher seniority or more night shifts, strategies are needed to prevent burnout and MetS.

Background. The aim of this study was to investigate depression and other determinants (sleep-deprived behaviors such as hours spent sleeping, watching television, and on the computer) and their association with excessive daytime sleepiness (EDS) among college freshmen. Methods. Self-administered questionnaires were collected from two colleges in northern Taiwan from July to September 2014. A total of 2643 students (38.7% male; ages ranged 18–23 years; mean age of 18.8 ± 1.2 years) completed an anonymous questionnaire on lifestyle behaviors (including personal habits, sleep duration and quality, and hours spent watching television and on the computer); perception of one’s health, a validated depression scale (Brief Symptom Rating Scale, BSRS-5); insomnia symptoms (the Chinese version of the Athens Insomnia Scale, CAIS); and EDS rated with the Chinese Epworth sleepiness scale (CESS). The data were analyzed using the chi-squared test, t-test, multivariate logistic regression, and multiple linear regression. Results. The prevalence of EDS among college students was approximately 27.1% (717/2643). The risk of EDS was elevated with increasing severity of depression: odds ratio (OR) = 2.8/3.71/5.01 for female, and OR = 3.29/5.07/5.07 for mild/moderate/severe depression for male, respectively (p < 0.05; marginally higher in male severe depression, p = 0.08). If depression score increased by 1 point, CESS score increased by 0.35 point; if time spent on the computer during non-holidays increased by 1 h, CESS score increased by 0.1 point; and for those whose sleep duration increased by 1 h during non-holidays, CESS score decreased by 0.1 point. Conclusions. EDS significantly predicted depression among college freshmen. Using a computer for a long time and less sleep duration during non-holidays contributed to EDS of college freshmen. Youths who experience EDS are recommended to seek assessment for depression symptoms and sleep-deprived behaviors, thus allowing physicians to offer appropriate screening and treatment.

Background: The prevalence of gallstone disease (GSD) increases with age, and the elderly have a much higher mortality risk and incidence of surgical comorbidities. The aim of this study was to explore the relationship between GSD and cardiometabolic risk factors in elderly people with non-alcoholic fatty liver disease (NAFLD). Methods: In this cross-sectional study, we analyzed the data of elderly people who underwent annual health check-ups at a Northern Taiwan health examination center. These data were collected from physical examination, blood tests, abdominal ultrasonography, and medical histories. We excluded those with hepatitis B or C infections, heavy alcohol consumption, or cholecystectomy. Results: The analysis included 3,037 participants with a mean age of 73.6± 6.0 years. Over 70% were overweight or obese, and the overall prevalence of GSD was 17.7%. In our univariate analysis, GSD was positively correlated with age, body mass index, metabolic syndrome, diabetes mellitus (DM), hypertension (HTN), and various metabolic factors (fasting plasma glucose [FPG], triglyceride, uric acid, and high-density lipoprotein cholesterol [HDL-C] levels). After adjustment for age, gender, and body mass index, metabolic syndrome showed a positive association with GSD (odds ratio [OR] 1.31 [95% confidence interval [CI], 1.05– 1.64]; P=0.020). Specific components of metabolic syndrome that increased the risk for GSD in NAFLD elderly include lower levels of HDL-C (OR 1.35 [95% CI, 1.10– 1.66]; P< 0.001) and elevated FPG (OR 1.36 [95% CI, 1.10– 1.69]; P< 0.001). Conclusion: Our study concluded that GSD is significantly associated with metabolic syndrome in elderly people with NAFLD. Reduced HDL-C and elevated FPG both heighten the risk of developing GSD. Therefore, to lower the risk of GSD in NAFLD patients, their FPG levels and HDL-C levels must be regularly followed-up, and these patients should be educated about the symptoms of GSD if they meet the criteria for metabolic syndrome.