The third paradigm provides an opportunity to look beyond any dichotomy between "standardized" versus "real-life" characteristics of the health care system and study designs. Namely, future research will determine whether the identification of these contextual factors can help to best design randomized controlled trials that provide better estimates of drugs' effectiveness.

Patients with severe COVID-19 have overwhelmed healthcare systems worldwide. We hypothesized that machine learning (ML) models could be used to predict risks at different stages of management and thereby provide insights into drivers and prognostic markers of disease progression and death. From a cohort of approx. 2.6 million citizens in Denmark, SARS-CoV-2 PCR tests were performed on subjects suspected for COVID-19 disease; 3944 cases had at least one positive test and were subjected to further analysis. SARS-CoV-2 positive cases from the United Kingdom Biobank was used for external validation. The ML models predicted the risk of death (Receiver Operation Characteristics—Area Under the Curve, ROC-AUC) of 0.906 at diagnosis, 0.818, at hospital admission and 0.721 at Intensive Care Unit (ICU) admission. Similar metrics were achieved for predicted risks of hospital and ICU admission and use of mechanical ventilation. Common risk factors, included age, body mass index and hypertension, although the top risk features shifted towards markers of shock and organ dysfunction in ICU patients. The external validation indicated fair predictive performance for mortality prediction, but suboptimal performance for predicting ICU admission. ML may be used to identify drivers of progression to more severe disease and for prognostication patients in patients with COVID-19. We provide access to an online risk calculator based on these findings.

Medication review for older patients with polypharmacy in the emergency department (ED) is crucial to prevent inappropriate prescribing. Our objective was to assess the feasibility of a collaborative medication review in older medical patients (≥65 years) using polypharmacy (≥5 long-term medications). A pharmacist performed the medication review using the tools: Screening Tool of Older Persons’ potentially inappropriate Prescriptions (STOPP) criteria, a drug–drug interaction database (SFINX), and Renbase® (renal dosing database). A geriatrician received the medication review and decided which recommendations should be implemented. The outcomes were: differences in Medication Appropriateness Index (MAI) and Assessment of Underutilization Index (AOU) scores between admission and 30 days after discharge and the percentage of patients for which the intervention was completed before discharge. Sixty patients were included from the ED, the intervention was completed before discharge for 50 patients (83%), and 39 (61.5% male; median age 80 years) completed the follow-up 30 days after discharge. The median MAI score decreased from 14 (IQR 8-20) at admission to 8 (IQR 2-13) 30 days after discharge (p < 0.001). The number of patients with an AOU score ≥1 was reduced from 36% to 10% (p < 0.001). Thirty days after discharge, 83% of the changes were sustained and for 28 patients (72%), 1≥ medication had been deprescribed. In conclusion, a collaborative medication review and deprescribing intervention is feasible to perform in the ED.