CoviD19
We developed and validated an accurate in-hospital mortality prediction score in a live EHR for automatic and continuous calculation using a novel model that improved upon SOFA.
The COVID-19 pandemic created an emergent need for a novel, accurate, and location and context-sensitive EHR-computable tool to predict mortality in hospitalized patients with and without COVID-19. Because developing a new score can take years, a predictive model must rely on well-validated scores.
In contrast, COVID-19 is a novel disease for which existing scores may be of limited but unknown predictive value.
As such, a predictive framework relying on multiple previously validated scores that can incorporate new information but only keeps the new inputs that explicitly improve performance is required. Stacked generalization provides a solution. A stacked model is built upon one or more baseline model(s) (e.g. SOFA) and incorporates additional models only when they improve prediction.
Materials and Methods
We developed, verified, and deployed a stacked generalization model to predict mortality using data available in the EHR by combining five previously validated scores and additional novel variables reported to be associated with COVID-19-specific mortality.
We verified the model with prospectively collected data from 12 hospitals in Colorado between March 2020 and July 2020. We compared the area under the receiver operator curve (AUROC) for the new model to the SOFA score and the Charlson Comorbidity Index.
Results
The prospective cohort included 27,296 encounters, of which 1,358 (5.0%) were positive for SARS-CoV-2, 4,494 (16.5%) required intensive care unit care, 1,480 (5.4%) required mechanical ventilation, and 717 (2.6%) ended in death. The Charlson Comorbidity Index and SOFA scores predicted mortality with an AUROC of 0.72 and 0.90, respectively. Our novel score predicted mortality with AUROC 0.94. In the subset of patients with COVID-19, the stacked model predicted mortality with AUROC 0.90, whereas SOFA had AUROC of 0.85.
Discussion
Stacked regression allows a flexible, updatable, live-implementable, ethically defensible predictive analytics tool for decision support that begins with validated models and includes only novel information that improves prediction.
read the paper abstract at https://academic.oup.com/jamia/advance-article/doi/10.1093/jamia/ocab100/6273353
read the entire paper at https://academic.oup.com/jamia/advance-article-pdf/doi/10.1093/jamia/ocab100/37905236/ocab100.pdf
Lire l'article complet sur : academic.oup.com
Social media is a popular source of information about health, including COVID-19 and testing information. One of the most popular communication channels used by adolescents and young adults who search for health information is TikTok—an emerging social media platform.
Objective: The purpose of this study was to describe TikTok videos related to COVID-19 testing.
Methods: The hashtag #covidtesting was searched, and the first 100 videos were included in the study sample. At the time the sample was drawn, these 100 videos garnered more than 50% of the views for all videos cataloged under the hashtag #covidtesting.
The content characteristics that were coded included mentions, displays, or suggestions of anxiety, COVID-19 symptoms, quarantine, types of tests, results of test, and disgust/unpleasantness. Additional data that were coded included the number and percentage of views, likes, and comments and the use of music, dance, and humor.
Results: The 100 videos garnered more than 103 million views; 111,000 comments; and over 12.8 million likes. Even though only 44 videos mentioned or suggested disgust/unpleasantness and 44 mentioned or suggested anxiety, those that portrayed tests as disgusting/unpleasant garnered over 70% of the total cumulative number of views (73,479,400/103,071,900, 71.29%) and likes (9,354,691/12,872,505, 72.67%), and those that mentioned or suggested anxiety attracted about 60% of the total cumulative number of views (61,423,500/103,071,900, 59.59%) and more than 8 million likes (8,339,598/12,872,505, 64.79%). Independent one-tailed t tests (α=.05) revealed that videos that mentioned or suggested that COVID-19 testing was disgusting/unpleasant were associated with receiving a higher number of views and likes.
Conclusions: Our finding of an association between TikTok videos that mentioned or suggested that COVID-19 tests were disgusting/unpleasant and these videos’ propensity to garner views and likes is of concern. There is a need for public health agencies to recognize and address connotations of COVID-19 testing on social media.
read the study report at https://publichealth.jmir.org/2021/6/e29528/
Lire l'article complet sur : publichealth.jmir.org