Allergic reactions to medication range from mild to severe or even life-threatening. Proper documentation of patient allergy information is critical for safe prescription, avoiding drug interactions, and reducing healthcare costs. Allergy information is regularly obtained during the medical interview, but is often poorly documented in electronic health records (EHRs). While many EHRs allow for structured adverse drug reaction (ADR) reporting, a free-text entry is still common. The resulting information is neither interoperable nor easily reusable for other applications, such as clinical decision support systems and prescription alerts. Current approaches require pharmacists to review and code ADRs documented by healthcare professionals. Recently, the effectiveness of machine algorithms in natural language processing (NLP) has been widely demonstrated. Our study aims to develop and evaluate different NLP algorithms that can encode unstructured ADRs stored in EHRs into institutional symptom terms. Our dataset consists of 79,712 pharmacist-reviewed drug allergy records. We evaluated three NLP techniques: Naive Bayes—Support Vector Machine (NB-SVM), Universal Language Model Fine-tuning (ULMFiT), and Bidirectional Encoder Representations from Transformers (BERT). We tested different general-domain pre-trained BERT models, including mBERT, XLM-RoBERTa, and WanchanBERTa, as well as our domain-specific AllergyRoBERTa, which was pre-trained from scratch on our corpus. Overall, BERT models had the highest performance. NB-SVM outperformed ULMFiT and BERT for several symptom terms that are not frequently coded. The ensemble model achieved an exact match ratio of 95.33%, a F₁score of 98.88%, and a mean average precision of 97.07% for the 36 most frequently coded symptom terms. The model was then further developed into a symptom term suggestion system and achieved a Krippendorff’s alpha agreement coefficient of 0.7081 in prospective testing with pharmacists. Some degree of automation could both accelerate the availability of allergy information and reduce the efforts for human coding.

Chinese economic development has led to a significant rise in internal migration over the last 20 years, including large numbers of elderly. When elderly Chinese people migrate, they still register their residency to their place of origin and often do not register with the new administrative office at the destination due to the household registration (hukou) system in China. Thus, most of these migrant elderly do not receive full social services, possibly leading to poor subjective well-being. This study aims to qualitatively examine the level of subjective well-being and its determinants among migrants and local elderly in Dongguan City of Guangdong province, one of the most rapid economically developing areas in China. We also present the results of in-depth interviews among 27 elderly, 15 elderly migrants and 12 local elderly living in Dongguan. The results reveal that the overall subjective well-being of the two groups were good. Most migrants believed their well-being had remained stable or even improved over time due to family reunion and a better physical environment. Elderly’s most valuable needs and the main reason of migration is family reunion; however, inadequacy of social support, including community support and government support (e.g., gift during holiday season, free health examination, healthcare expenditure reimbursement), cannot be neglected for maintaining a good level of well-being. The well-being of migrant elderly can be further enhanced by promoting social services and welfare, recreational activities, and enhancing healthcare reimbursement in their new home.

Several studies have shown the health effects of air pollutants, especially in China, North American and Western European countries. But longitudinal cohort studies focused on health effects of long-term air pollution exposure are still limited in Southeast Asian countries where sources of air pollution, weather conditions, and demographic characteristics are different. The present study examined the association between long-term exposure to air pollution and self-reported morbidities in participants of the Thai cohort study (TCS) in Bangkok metropolitan region (BMR), Thailand. This longitudinal cohort study was conducted for 9 years from 2005 to 2013. Self-reported morbidities in this study included high blood pressure, high blood cholesterol, and diabetes. Air pollution data were obtained from the Thai government Pollution Control Department (PCD). Particles with diameters ≤10 μm (PM₁₀), sulfur dioxide (SO₂), nitrogen dioxide (NO₂), ozone (O₃), and carbon monoxide (CO) exposures were estimated with ordinary kriging method using 22 background and 7 traffic monitoring stations in BMR during 2005-2013. Long-term exposure periods to air pollution for each subject was averaged as the same period of person-time. Cox proportional hazards models were used to examine the association between long-term air pollution exposure with self-reported high blood pressure, high blood cholesterol, diabetes. Results of self-reported morbidity were presented as hazard ratios (HRs) per interquartile range (IQR) increase in PM₁₀, O₃, NO₂, SO₂, and CO. After controlling for potential confounders, we found that an IQR increase in PM₁₀ was significantly associated with self-reported high blood pressure (HR = 1.13, 95% CI: 1.04, 1.23) and high blood cholesterol (HR = 1.07, 95%CI: 1.02, 1.12), but not with diabetes (HR = 1.05, 95%CI: 0.91, 1.21). SO₂ was also positively associated with self-reported high blood pressure (HR = 1.22, 95%CI: 1.08, 1.38), high blood cholesterol (HR = 1.20, 95%CI: 1.11, 1.30), and diabetes (HR = 1.21, 95%CI: 0.92, 1.60). Moreover, we observed a positive association between CO and self-reported high blood pressure (HR = 1.07, 95%CI: 1.00, 1.15), but not for other diseases. However, self-reported morbidities were not associated with O₃ and NO₂. Long-term exposure to air pollution, especially for PM₁₀ and SO₂ was associated with self-reported high blood pressure, high blood cholesterol, and diabetes in subjects of TCS. Our study supports that exposure to air pollution increases cardiovascular disease risk factors for younger population.

Over time, large amounts of clinical data have accumulated in electronic health records (EHRs), making it difficult for healthcare professionals to navigate and make patient-centered decisions. This underscores the need for healthcare recommendation systems that help medical professionals make faster and more accurate decisions. This study addresses drug recommendation systems that generate an appropriate list of drugs that match patients’ diagnoses. Currently, recommendations are manually prepared by physicians, but this is difficult for patients with multiple comorbidities. We explored approaches to drug recommendations based on elderly patients with diabetes, hypertension, and cardiovascular disease who visited primary-care clinics and often had multiple conditions. We examined both collaborative filtering approaches and traditional machine-learning classifiers. The hybrid model between the two yielded a recall at 5 of 76.61%, a precision at 5 of 46.20%, a macro-averaged area under the curve of 74.52%, and an average physician agreement of 47.50%. Although collaborative filtering is widely used in recommendation systems, our results showed that it consistently underperformed traditional classification. Collaborative filtering was sensitive to class imbalances and favored the more popular classes. This study highlighted challenges that need to be addressed when developing recommendation systems in EHRs.

Critical care staff are presented with a large amount of data, which made it difficult to systematically evaluate. Early detection of patients whose condition is deteriorating could reduce mortality, improve treatment outcomes, and allow a better use of healthcare resources. In this study, we propose a data-driven framework for predicting the risk of mortality that combines high-accuracy recurrent neural networks with interpretable explanations. Our model processes time-series of vital signs and laboratory observations to predict the probability of a patient’s mortality in the intensive care unit (ICU). We investigated our approach on three public critical care databases: Multiparameter Intelligent Monitoring in Intensive Care III (MIMIC-III), MIMIC-IV, and eICU. Our models achieved an area under the receiver operating characteristic curve (AUC) of 0.87–0.91. Our approach was not only able to provide the predicted mortality risk but also to recognize and explain the historical contributions of the associated factors to the prediction. The explanations provided by our model were consistent with the literature. Patients may benefit from early intervention if their clinical observations in the ICU are continuously monitored in real time.