Kevin G. Buell

Rationale: Early mobilization of mechanically ventilated patients within 72 hours of intubation is underutilized despite its known clinical benefits. The most cited barrier to implementation is physiologic instability. Objectives: To quantify eligibility for early mobilization across established physiologic criteria in a multicenter cohort of mechanically ventilated adults and estimate the relative impact of structural and process implementation barriers. Methods: Retrospective cohort of adults receiving invasive mechanical ventilation for ≥4 hours across nine CLIF Consortium health systems (2018–2024) and MIMIC-IV (2008–2022). Four physiologic eligibility criteria, from clinical trials (Chicago, TEAM) and consensus guidelines (Green, Yellow), were applied between 8am and 5pm within 72 hours of intubation. Random-effects meta-analysis compared eligibility across criteria and quantified the relative effect of two implementation barriers: weekday-only availability (structural) and 4-hour continuous eligibility (process). Measurements and Main Results: Among 223,281 hospitalizations, 86.9–95.6% met eligibility for mobilization within 72 hours of intubation across all 4 criteria. Compared with Chicago trials criteria, eligibility was reduced under TEAM (SHR 0.782, 95% CI 0.746–0.820) and Green (SHR 0.631, 0.611–0.651) but not Yellow (SHR 0.990, 0.976–1.005). Restricting eligibility to weekdays (structural) or requiring continuous eligibility (process) each reduced eligibility by 26-39% across all four criteria (SHR range 0.61–0.74); imposing both barriers together reduced eligibility by 45-56% (SHR range 0.44-0.55). Conclusions: Most mechanically ventilated adults met physiologic eligibility criteria for early mobilization within 72 hours of intubation. Structural and process barriers, such as staff availability on weekends and sustained eligibility for care coordination, reduced opportunity more potently than physiologic instability alone.

Hyperkalemia is a potentially life-threatening condition among patients presenting to the Emergency Department (ED). However, most epidemiological studies have focused on those with end-stage renal disease (ESRD) and there is limited recent, large-scale, and robust data available on the incidence and management of hyperkalemia for non-ESRD patients in the ED. This was a retrospective cohort study of adults (age ≥ 18 years) without ESRD presenting to the ED with hyperkalemia from 1/1/2016 to 12/31/2024. ICD-10 codes were used to identify patients in the Epic Cosmos database. The primary outcome was the incidence of hyperkalemia from all ED visits. Secondary outcomes included rates of admission, cardiac arrest, hemodialysis, and medications administered to treat hyperkalemia. Data were analyzed using summary statistics and odds ratios (OR) with 95% confidence intervals (CI). Among 246,235,769 ED visits, 803,186 (0.33%) had an ICD-10 code for hyperkalemia and 539,033 of those (67.11%) were admitted to the hospital. Cardiac arrest occurred in 18,044 (2.25%) patients with hyperkalemia. Sodium bicarbonate was given in 38.75%, calcium gluconate or chloride in 32.64%, sodium zirconium cyclosilicate in 24.68%, sodium polystyrene sulfonate in 23.12%, and patiromer in 3.04%. Only 2.99% received hemodialysis. Among adult patients without ESRD, hyperkalemia is an uncommon but important condition, with twothirds requiring admission and two percent experiencing cardiac arrest. Sodium bicarbonate and calcium administration is common, while hemodialysis is rare.

Real-world data from vendor-aggregated health information exchanges represent a powerful resource for retrospective clinical research. Epic's Cosmos platform, a large-scale centralized data warehouse, enables querying deidentified electronic health record data. We describe the iterative development of a generalizable clinical informatics workflow for retrospective studies using Cosmos. We applied the Plan-Do-Study-Act cycle to iteratively refine a collaborative research process using Cosmo's SlicerDicer interface, focusing on setting, team, technical preparation, and workflow. We identified multiple areas of improvement to facilitate streamlined collaborative studies. Our institutional review board confirmed Cosmos studies as nonhumans subject research. We expanded collaborations across institutions. We identified key insights for team definition across clinical, research, and skill domains. We expanded our team based on technical skills, domain expertise, and educational aims. We optimized query building and cohort construction, data analysis and validation, and communication processes. We clarified and optimized a collaborative workflow across clinical and informatics expertise. Our collaborative approach to secondary data analysis in Cosmos supports the development of meaningful clinical evidence to support high-quality, well-evidenced patient care. This study demonstrates the application of the Plan-Do-Study-Act cycle to a collaborative workflow for clinically focused secondary data analysis. Our approach allows rapid, reproducible cohort construction and analysis, is adaptable across clinical domains, and scales to multiorganizational collaboration. This approach offers a model for others seeking to develop key clinical insights via retrospective studies within the Cosmos data aggregation tool.