Shift-Level Team Familiarity Is Associated with Improved Outcomes in Mechanically Ventilated Adults.

Due to a typesetter error, this is a previous version of this article; It will be replaced shortly by the final accepted version. Rationale: Organizing ICU interprofessional teams-nurses, physicians, and respiratory therapists-is high priority because of workforce crises, but how often clinicians work together (i.e., interprofessional familiarity) remains unexplored. Objectives: Determine if mechanically ventilated patients cared for by teams with greater familiarity have lower mortality, shorter duration of mechanical ventilation, and greater spontaneous breathing trial (SBT) implementation. Methods: Using electronic health records from five ICUs (2018-2019), we identified the interprofessional team that cared for each mechanically ventilated patient each shift, calculated familiarity, and modeled familiarity exposures separately on ICU mortality, duration of mechanical ventilation, and SBT implementation using encounter-level generalized linear regression models with a log-link, unit-level fixed effects adjusting for cofounders, including severity of illness. Measurements and Main Results: Familiarity was defined as how often clinicians worked together for all patients in an ICU (i.e., coreness) and for each patient (i.e., mean team value). Among 4,292 patients (4,485 encounters, 72,210 shifts), unadjusted mortality was 12.9%, average duration of mechanical ventilation was 2.32 days, and SBT implementation was 89%. An increase in coreness and mean team value, by the SD of each, was associated with lower probability of dying (coreness: adjusted marginal effect, -0.038; 95% confidence interval [-0.07 to -0.004]; mean team value: adjusted marginal effect, -0.0034 [-0.054 to -0.014]); greater probability of receiving SBT when eligible (coreness: 0.45 [0.007 to 0.083]; mean team value: 0.012 [-0.017 to 0.042]), and shorter duration of mechanical ventilation (coreness: -0.23 [-0.321 to -0.139]). Conclusions: Interprofessional familiarity was associated with improved outcomes; assignment models that prioritize familiarity might be a novel solution.