Mysql

Tyche was set up as a browser-based program running on a Linux server. It employs Joomla!^® (Open Source Matters, Inc., Joomla community, New York, NY, USA) for user management and several publicly licensed JavaScript libraries: dropzone (Matias Meno, MIT License) is used to upload images, and cornerstone (Chris Hafey, MIT License) and cornerstone Tools (Chris Hafey, MIT License) are used to display, analyze, and annotate DICOM images. Its programming languages are PHP (the PHP development team, Zend Technologies) and JavaScript^® (Oracle Corporation, Austin, TX, USA), as well as MySQL (Oracle Corporation, Austin, TX, USA) for database management. All connections are secure sockets layer (SSL)-encrypted.

The prediction outcome of the study is the probability of in-hospital mortality, defined as patient’s condition upon leaving the hospital. Based on previous studies [19 (link)-22 (link)] and experts’ opinion (a total of 6 independent medical professionals and cardiologists in West China Hospital of Sichuan University), demographics, comorbidities, vital signs, and laboratory findings (Multimedia Appendix 2) were extracted from the eICU-CRD, using Structured Query Language (MySQL) queries (version 5.7.33; Oracle Corporation). The following tables from eICU-CRD were used: “diagnosis,” “intakeoutput,” “lab,” “patient,” and “nursecharting.” Except for the demographic characteristics, other variables were collected during the first 24 hours of each ICU admission. Furthermore, to avoid overfitting, Least Absolute Shrinkage and Selection Operator (LASSO) is used to select and filter the variables [23 (link),24 (link)].

All analyses were conducted on weighted data, as recommended by the CDC’s NCHS website. The weighting is calculated using the most recent census data to provide a stratified representation of the national patient population. All participants’ records were stored in a relational database using the open-source database software MySQL (v. 5.7.11, Oracle, Redwood Shores, CA, USA). All analytics were performed using the open-source statistical computing software R (v 3.2.3, R Foundation, Vienna, Austria). The functions of svydesign and svyglm from the R package survey were used to account for stratified, clustered, and weighted variables in the NHAMCS data. Wald tests of association were used to determine significance for bivariate analyses. Stepwise regression via backward elimination was used including all independent variables mentioned above. Separate independent logistic regression models were run holding pain score as the sole independent variable and opioid use as the dependent variable. CDC detailed documentation of the NHAMCS instrument, methodology and data files that were used as the basis for these analyses are available elsewhere [25 ].