Comparación de las estructuras de difusión de información errónea y verídica en las redes sociales durante una emergencia de salud pública / Contrasting Misinformation and Real-Information Dissemination Network Structures on Social Media During a Health Emergency

RESUMEN Objetivos. Elaborar un esquema operativo integral para detectar la información errónea principal sobre el zika distribuida en Twitter® en el 2016; reconstruir las redes por las que se difunde información mediante retuiteo; contrastar la información verídica frente a la errónea con diversos parámetros; e investigar cómo se difundió en las redes sociales la información errónea sobre el zika durante la epidemia. Métodos. Revisamos sistemáticamente los 5 000 tuits más retuiteados con información sobre el zika en inglés, definimos "información errónea" a partir de la evidencia, buscamos tuits que tuvieran información errónea y conformamos un grupo equiparable de tuits con información verídica. Elaboramos un algoritmo para reconstruir las redes de retuiteo de 266 tuits con información errónea y 458 tuits equiparables con información verídica. Calculamos y comparamos nueve parámetros para caracterizar la estructura de las redes a varios niveles, entre los dos grupos. Resultados. En los nueve parámetros se aprecian diferencias estadísticamente significativas entre el grupo de información verídica y el de información errónea. La información errónea en general se difunde mediante estructuras más sofisticadas que la información verídica. También hay una considerable variabilidad intragrupal. Conclusiones. Las redes de difusión de la información errónea sobre el zika en Twitter fueron sustancialmente diferentes que las de información verídica, lo cual indica que la información errónea se sirve de mecanismos de difusión distintos. Nuestro estudio permitirá formar una comprensión más holística de los desafíos que plantea la información errónea sobre salud en las redes sociales.

ABSTRACT Objectives. To provide a comprehensive workflow to identify top influential health misinformation about Zika on Twitter in 2016, reconstruct information dissemination networks of retweeting, contrast mis- from real information on various metrics, and investigate how Zika misinformation proliferated on social media during the Zika epidemic. Methods. We systematically reviewed the top 5000 English-language Zika tweets, established an evidence-based definition of "misinformation," identified misinformation tweets, and matched a comparable group of real-information tweets. We developed an algorithm to reconstruct retweeting networks for 266 misinformation and 458 comparable real-information tweets. We computed and compared 9 network metrics characterizing network structure across various levels between the 2 groups. Results. There were statistically significant differences in all 9 network metrics between real and misinformation groups. Misinformation network structures were generally more sophisticated than those in the real-information group. There was substantial within-group variability, too. Conclusions. Dissemination networks of Zika misinformation differed substantially from real information on Twitter, indicating that misinformation utilized distinct dissemination mechanisms from real information. Our study will lead to a more holistic understanding of health misinformation challenges on social media.

Diagnostic Value of Coronary Artery Calcium Score for Cardiovascular Disease in African Americans: The Jackson Heart Study.

Background: The role of coronary artery calcium (CAC) as a screening tool for cardiovascular disease (CVD) risk in African Americans (AAs) is unclear. We compared the diagnostic accuracy for CVD prevalence using the CAC score and the Framingham Risk Score (FRS) in an adult population of AAs. Methods: CAC was measured in 2944 participants AAs. Approximately 8% of this cohort had known CVD defined as prior myocardial infarction, stroke, percutaneous coronary intervention, coronary artery bypass grafting and peripheral artery disease. Logistic regression, receiver operating characteristic (ROC) and net reclassification index (NRI) analysis were used adjusting for age, gender, systolic blood pressure (SBP), total and high-density lipoprotein (HDL) cholesterol, smoking status, diabetes mellitus (DM), body mass index (BMI), blood pressure medication and statin use. Participants with prevalent clinical CVD and DM were classified as high FRS risk. Results: The mean age of participants was 60 years, 65% were females, 26% had DM, 50% were obese and 30% were current or former smokers. Prevalent CVD was associated with older age, higher SBP, lower HDL and total cholesterol, and higher CAC. The prevalence of CAC was 83% in participants with prevalent CVD and 45% in those without CVD. CAC was independently associated with prevalent CVD in our multivariable model [OR (95% CI): 1.22 (1.12 -1.32), p< 0.0001]. In ROC analysis, CAC improved the diagnostic accuracy (c statistic) of the FRS from 0.617 to 0.757 (p < 0.0001) for prevalent CVD. Addition of CAC to FRS resulted in net reclassification improvement of 4% for subjects with known CVD and 28.5% in those without CVD. Conclusion: In AAs, CAC is independently associated with prevalent CVD and improves the diagnostic accuracy of FRS for prevalent CVD by 14%. Addition of CAC improves the NRI of those with prevalent CVD by 4% and the NRI of individuals without CVD by 28.5%. Determination of CAC may be useful in CVD risk stratification in AAs.