Early detection of nasopharyngeal carcinoma through machine-learning-driven prediction model in a population-based healthcare record database.

OBJECTIVE: Early diagnosis and treatment of nasopharyngeal carcinoma (NPC) are vital for a better prognosis. Still, because of obscure anatomical sites and insidious symptoms, nearly 80% of patients with NPC are diagnosed at a late stage. This study aimed to validate a machine learning (ML) model utilizing symptom-related diagnoses and procedures in medical records to predict nasopharyngeal carcinoma (NPC) occurrence and reduce the prediagnostic period. MATERIALS AND METHODS: Data from a population-based health insurance database (2001-2008) were analyzed, comparing adults with and without newly diagnosed NPC. Medical records from 90 to 360 days before diagnosis were examined. Five ML algorithms (Light Gradient Boosting Machine [LGB], eXtreme Gradient Boosting [XGB], Multivariate Adaptive Regression Splines [MARS], Random Forest [RF], and Logistics Regression [LG]) were evaluated for optimal early NPC detection. We further use a real-world data of 1 million individuals randomly selected for testing the final model. Model performance was assessed using AUROC. Shapley values identified significant contributing variables. RESULTS: LGB showed maximum predictive power using 14 features and 90 days before diagnosis. The LGB models achieved AUROC, specificity, and sensitivity were 0.83, 0.81, and 0.64 for the test dataset, respectively. The LGB-driven NPC predictive tool effectively differentiated patients into high-risk and low-risk groups (hazard ratio: 5.85; 95% CI: 4.75-7.21). The model-layering effect is valid. CONCLUSIONS: ML approaches using electronic medical records accurately predicted NPC occurrence. The risk prediction model serves as a low-cost digital screening tool, offering rapid medical decision support to shorten prediagnostic periods. Timely referral is crucial for high-risk patients identified by the model.

E-cigarettes and smoking cessation among adolescent smokers.

Smokers of any age can reap substantial health benefits from quitting or reducing their smoking. E-cigarettes have been promoted as a potentially promising product for tobacco harm reduction because e-cigarettes deliver nicotine vapor without many of the hazardous chemical combustion byproducts produced by combustible cigarette smoking. However, there remains an ongoing debate on whether the use of e-cigarettes is effective in combustible cigarette smoking cessation or reduction in both adolescents and adults. Our study uses data from the 2015 (baseline) and from the 2017 (follow-up) waves of the Taiwan Adolescent to Adult Longitudinal Study (TAALS), which is a large nationwide representative cohort study of health behaviors among adolescents in Taiwan. We analyzed the data using logistic regression and multivariate regression with a post-stratification weighting procedure. Among the 474 adolescent combustible cigarette users at baseline, the use of e-cigarettes had no association with smoking cessation (aRR = 0.99, 95% CI = 0.66, 1.50). Furthermore, the use of e-cigarettes was also not associated with change in combustible cigarette consumption among all adolescent combustible cigarette users at follow-up (Coef. = 0.62, 95% CI = - 36.85, 38.09). In summary, our findings suggest that e-cigarettes may not aid tobacco control among adolescent smokers. Policy makers should be cautious of the potential harms that e-cigarette may bring to young people when they are developing e-cigarette regulations.

Bullying Experiences, Depression, and the Moderating Role of Resilience Among Adolescents.

Introduction: Resilience refers to the ability to adapt to difficult situation or adversity. Resilience is what gives people the psychological strength to cope with stress and hardship. Previous studies have investigated the relationship between resilience and bullying victimization and mental health problems. But whether the moderating effect of resilience against depression varies among victims of different types of bullying victimization remains unknown. Methods: The study used data from the Taiwan Adolescent to Adult Longitudinal Study (TAALS), which was a school based, nationwide, longitudinal study conducted among adolescents in Taiwan. Between 2015 and 2019, the survey was repeated three times to capture changes in health behaviors. Meanwhile, our study is a cross-sectional study focusing on the 2nd follow-up survey of the TAALS, where we recruited 4,771 Grade 7 (12-13 years) and Grade 10 (15-16 years) students who had experienced bullying at school. Results: This study confirms the protective effect of resilience on depression among adolescents who have experienced bullying. The mode resilience score was used as a reference group. Compared to the reference group, victims of verbal bullying from the lowest resilience group were at the greatest risk of depression (OR = 5.91, CI = 4.38-7.99). Compared to the reference group, victims of cyber bullying from the highest resilience group had the lowest risk of depression (OR = 0.72, CI = 0.57-0.90). Conclusion: Regardless of the type of bullying victimization, resilience has been shown to offer protection against depression. Specifically, higher resilience levels offer the greatest protection against depression for victims of cyber bullying compared to other three types of bullying victimization. Early interventions to reduce negative effects of bullying victimization may start with increasing an individual's resilience during adolescence.

Thoracic surgery improved overall survival in patients with stage IIIB-IV epidermal growth factor receptor-mutant lung adenocarcinoma who received and responded to tyrosine kinase inhibitor treatment.

PURPOSE: No large-scale, prospective, randomized study has evaluated the effect of thoracic surgery on patients with unresectable stage IIIB-IV epidermal growth factor receptor (EGFR)-mutant lung adenocarcinoma who received and responded to EGFR tyrosine kinase inhibitor (TKI) treatment. Therefore, we designed a propensity-score-matched, nationwide, population-based, cohort study to investigate the effects of thoracic surgery on patients with EGFR-mutant lung adenocarcinoma. PATIENTS AND METHODS: We included patients with unresectable stage IIIB-IV EGFR-mutant lung adenocarcinoma and categorized them into two groups according to their treatment modalities and compared their outcomes: the case group consisted of patients who underwent thoracic surgery for lung tumors after receiving and responding to EGFR-TKI treatment and the comparison group consisted of patients who received EGFR-TKI treatment alone until tumor progression. Patients in both groups were matched at a ratio of 1:4. RESULTS: The matching process yielded a final cohort of 1395 patients (279 and 1,116 in the case and comparison groups, respectively) who were eligible for further analysis. According to multivariable Cox regression analyses, the adjusted hazard ratio (aHR; 95% confidence interval [CI]) for thoracic surgery for lung tumors after EGFR-TKI use and tumor response (group 2) compared with EGFR-TKI treatment alone (group 1) was 0.445 (0.351-0.564). CONCLUSIONS: Thoracic surgery prolonged overall survival in patients with unresectable stage IIIB-IV EGFR-mutant lung adenocarcinoma who received and responded to EGFR-TKI treatment.

Machine-Learning Techniques for Feature Selection and Prediction of Mortality in Elderly CABG Patients.

Coronary artery bypass surgery grafting (CABG) is a commonly efficient treatment for coronary artery disease patients. Even if we know the underlying disease, and advancing age is related to survival, there is no research using the one year before surgery and operation-associated factors as predicting elements. This research used different machine-learning methods to select the features and predict older adults' survival (more than 65 years old). This nationwide population-based cohort study used the National Health Insurance Research Database (NHIRD), the largest and most complete dataset in Taiwan. We extracted the data of older patients who had received their first CABG surgery criteria between January 2008 and December 2009 (n = 3728), and we used five different machine-learning methods to select the features and predict survival rates. The results show that, without variable selection, XGBoost had the best predictive ability. Upon selecting XGBoost and adding the CHA2DS score, acute pancreatitis, and acute kidney failure for further predictive analysis, MARS had the best prediction performance, and it only needed 10 variables. This study's advantages are that it is innovative and useful for clinical decision making, and machine learning could achieve better prediction with fewer variables. If we could predict patients' survival risk before a CABG operation, early prevention and disease management would be possible.

Treatment-Emergent Co-Morbidities and Survival in Patients With Metastatic Castration-Resistant Prostate Cancer Receiving Abiraterone or Enzalutamide.

Secondary hormone therapy, abiraterone and enzalutamide, has improved outcomes for metastatic castration-resistant prostate cancer (mCRPC) and prolonged patients' lives significantly. Various studies have compared the cancer-related outcomes, adverse effects, and drug-induced comorbidities in patients with mCRPC who are treated with abiraterone or enzalutamide. However, few studies have explored associations between survival and comorbidities or comprehensive analyzed newly developed comorbidities during and after secondary hormone therapy. We attempted to clarify whether the Charlson comorbidity index (CCI) overall or itemized is predictive for overall survival, and we compared newly developed comorbidities between abiraterone and enzalutamide groups. We extracted data about expenses and comorbidities for patients who have mCRPC, received abiraterone and enzalutamide and met pre-examination operation criteria between September 2016 and December 2017 from the Taiwan National Health Insurance database. A total of 1153 patients with mCRPC who received abiraterone (n = 782) or enzalutamide (n = 371) with or without previous chemotherapy were included. We used the propensity score to match confounding factors, including age, pre-existing comorbidities, and precipitating factors for comorbidity (e.g., hypertension, hyperlipidemia), to eliminate selection bias in the comparison of newly developed comorbidities. Cox regression analysis was used for overall survival. We found that enzalutamide is superior to abiraterone with regard to overall survival. Our study revealed that there is no statistically significant difference in development of new comorbidities between abiraterone and enzalutamide group. Moreover, the CCI score, rather than any single item of the CCI, was a statistically significant predictor for overall survival.

Electronic Cigarette Use and Smoking Initiation in Taiwan: Evidence from the First Prospective Study in Asia.

A growing literature indicates that electronic cigarette use increases the risk of subsequent initiation of conventional smoking among cigarette-naïve adolescents in several Western countries. This research assesses the same relationship in an Asian country, Taiwan. The Taiwan Adolescent to Adult Longitudinal Study is a school-based survey that was carried out in two waves in 2014 (baseline) and in 2016 (follow-up). It employs probability sampling to create nationally representative samples of students in junior high school (mean age 13, 7th grade at baseline) and in senior high school (mean age 16, 10th grade at baseline). Data from this survey were analyzed via logistic regression to estimate the association between ever use of e-cigarettes at baseline and smoking initiation at follow-up, accounting for susceptibility to smoking, socio-demographic profile, depression status, and peer support. Among the 12,954 cigarette-naïve students surveyed, those with e-cigarette experience at baseline exhibited higher odds of smoking initiation at follow-up (Odds Ratio = 2.14, 95% CI (1.66, 2.75), p < 0.001). For the first time, we confirmed, through a longitudinal survey, a prospective association between ever use of e-cigarettes and smoking initiation in an Asian adolescent population. The restrictive policy on e-cigarettes currently in force in Taiwan is justified to prevent both e-cigarette and cigarette use among adolescents.

The Taiwan Adolescent to Adult Longitudinal Study (TAALS): Methodology and Cohort Description.

The objective of the study was to introduce the methodology and report on cohort description of Taiwan Adolescent to Adult Longitudinal Study (TAALS). TAALS is the first nationwide longitudinal survey among Taiwan adolescents, linked with the National Health Insurance Research Database (NHIRD) to obtain complete medical records of respondents in the future. The TAALS project employed the principle of probability proportional to size (PPS) sampling method. Data were collected by questionnaire from 18 064 school students participating in 2015 formal survey, with good sample representation via a goodness-of-fit test after weighting adjustments. Through expert evaluation and statistics tests, TAALS shows a well nationally representation, validity, and reliability. Results indicate that the vocational school students had poor healthy behavior than other education systems, supporting the hypothesis that different learning environment will develop different health behaviors. TAALS can serve as a foundation for analyzing health trajectories of Taiwan adolescents.