Patient-Level Prediction of Cardio-Cerebrovascular Events in Hypertension Using Nationwide Claims Data.

BACKGROUND: Prevention and management of chronic diseases are the main goals of national health maintenance programs. Previously widely used screening tools, such as Health Risk Appraisal, are restricted in their achievement this goal due to their limitations, such as static characteristics, accessibility, and generalizability. Hypertension is one of the most important chronic diseases requiring management via the nationwide health maintenance program, and health care providers should inform patients about their risks of a complication caused by hypertension. OBJECTIVE: Our goal was to develop and compare machine learning models predicting high-risk vascular diseases for hypertensive patients so that they can manage their blood pressure based on their risk level. METHODS: We used a 12-year longitudinal dataset of the nationwide sample cohort, which contains the data of 514,866 patients and allows tracking of patients' medical history across all health care providers in Korea (N=51,920). To ensure the generalizability of our models, we conducted an external validation using another national sample cohort dataset, comprising one million different patients, published by the National Health Insurance Service. From each dataset, we obtained the data of 74,535 and 59,738 patients with essential hypertension and developed machine learning models for predicting cardiovascular and cerebrovascular events. Six machine learning models were developed and compared for evaluating performances based on validation metrics. RESULTS: Machine learning algorithms enabled us to detect high-risk patients based on their medical history. The long short-term memory-based algorithm outperformed in the within test (F1-score=.772, external test F1-score=.613), and the random forest-based algorithm of risk prediction showed better performance over other machine learning algorithms concerning generalization (within test F1-score=.757, external test F1-score=.705). Concerning the number of features, in the within test, the long short-term memory-based algorithms outperformed regardless of the number of features. However, in the external test, the random forest-based algorithm was the best, irrespective of the number of features it encountered. CONCLUSIONS: We developed and compared machine learning models predicting high-risk vascular diseases in hypertensive patients so that they may manage their blood pressure based on their risk level. By relying on the prediction model, a government can predict high-risk patients at the nationwide level and establish health care policies in advance.

Cultural values and cross-cultural video consumption on YouTube.

Video-sharing social media like YouTube provide access to diverse cultural products from all over the world, making it possible to test theories that the Web facilitates global cultural convergence. Drawing on a daily listing of YouTube's most popular videos across 58 countries, we investigate the consumption of popular videos in countries that differ in cultural values, language, gross domestic product, and Internet penetration rate. Although online social media facilitate global access to cultural products, we find this technological capability does not result in universal cultural convergence. Instead, consumption of popular videos in culturally different countries appears to be constrained by cultural values. Cross-cultural convergence is more advanced in cosmopolitan countries with cultural values that favor individualism and power inequality.

The Effect of Bad News and CEO Apology of Corporate on User Responses in Social Media.

While social media has become an important platform for social reputation, the emotional responses of users toward bad news have not been investigated thoroughly. We analyzed a total of 20,773 Twitter messages by 15,513 users to assess the influence of bad news and public apology in social media. Based on both computerized, quantitative sentiment analysis and in-depth qualitative analysis, we found that rapid public apology effectively and immediately reduced the level of negative sentiment, where the degree of change in sentiments differed by the type of interactions users engaged in. The majority of users who directly conversed with corporate representatives on the new media were not typical consumers, but experts and practitioners. We extend the existing cognitive model and suggest the audiences' psychological reaction model to describe the information processing process during and after an organizational crisis and response. We also discuss various measures through which companies can respond to a crisis properly in social media in a fashion that is different from conventional mass media.