Development and Validation of a Machine Learning COVID-19 Veteran (COVet) Deterioration Risk Score.

BACKGROUND AND OBJECTIVE: To develop the COVid Veteran (COVet) score for clinical deterioration in Veterans hospitalized with COVID-19 and further validate this model in both Veteran and non-Veteran samples. No such score has been derived and validated while incorporating a Veteran sample. DERIVATION COHORT: Adults (age ≥ 18 yr) hospitalized outside the ICU with a diagnosis of COVID-19 for model development to the Veterans Health Administration (VHA) (n = 80 hospitals). VALIDATION COHORT: External validation occurred in a VHA cohort of 34 hospitals, as well as six non-Veteran health systems for further external validation (n = 21 hospitals) between 2020 and 2023. PREDICTION MODEL: eXtreme Gradient Boosting machine learning methods were used, and performance was assessed using the area under the receiver operating characteristic curve and compared with the National Early Warning Score (NEWS). The primary outcome was transfer to the ICU or death within 24 hours of each new variable observation. Model predictor variables included demographics, vital signs, structured flowsheet data, and laboratory values. RESULTS: A total of 96,908 admissions occurred during the study period, of which 59,897 were in the Veteran sample and 37,011 were in the non-Veteran sample. During external validation in the Veteran sample, the model demonstrated excellent discrimination, with an area under the receiver operating characteristic curve of 0.88. This was significantly higher than NEWS (0.79; p < 0.01). In the non-Veteran sample, the model also demonstrated excellent discrimination (0.86 vs. 0.79 for NEWS; p < 0.01). The top three variables of importance were eosinophil percentage, mean oxygen saturation in the prior 24-hour period, and worst mental status in the prior 24-hour period. CONCLUSIONS: We used machine learning methods to develop and validate a highly accurate early warning score in both Veterans and non-Veterans hospitalized with COVID-19. The model could lead to earlier identification and therapy, which may improve outcomes.

Cardiovascular disease in women.

Cardiovascular disease (CVD) remains the most important health issue facing women and continues to be their number one cause of morbidity and mortality. Women are disproportionately affected by CVD compared with men. It is diagnosed less often and treated less aggressively in the inpatient and outpatient settings; as a result, women have poorer outcomes. It is therefore imperative that physicians take steps to screen women for the risks associated with CVD and actively education them on primary and secondary prevention.

Buying prescription drugs on the internet: promises and pitfalls.

Patients can save time and money by purchasing drugs from Internet pharmacies, but they can also end up with counterfeit or substandard medications. Online pharmacies bypass the safeguards of a doctor-patient relationship, creating a dangerous opportunity for prescription drug abuse and unchecked medication interactions and side effects.

Using computerized physician order entry to decrease insurance denials.

Insurance denials delay payments for tests to medical institutions and can decrease patient satisfaction due to unexpected billing. Our institution utilizes an ambulatory electronic health record (EHR) for routine clinical care that includes computerized physician order entry (CPOE). At our institution as well as others, considerable cost is associated with inappropriate diagnostic coding of needed procedures and tests by the physician. We developed a series of CPOE alerts and order sets targeting specific tests to address this problem. As a result, preliminary data shows that insurance denials fell by up to 37% for the targeted tests.