Does the trauma system protect against the weekend effect?

Abstract

Background: Occurrence on weekends or at night has been associated with poor outcomes for time-sensitive conditions including ST elevation myocardial infarction, stroke, and cardiac arrest. We sought to determine whether the "weekend effect" exists for injured patients at our trauma center.

Methods: We performed a retrospective cohort study at a Level I trauma center (2006-2008). The relative risks of mortality associated with weekend or night arrival were estimated using unadjusted and adjusted analyses.

Results: Four thousand three hundred eighty-two patients were included. One-third of patients (34.0%) arrived on weekends, and 23.3% of patients arrived at night (12:00 midnight to 6:00 am). Average age was 43.2 years (44.2 weekdays vs. 41.4 weekends, p < 0.001 and 45.1 days vs. 37.5 nights, p < 0.001), 72.3% were men (72.6 weekdays vs. 71.8 weekends, p = not significant (NS) and 71.0% days vs. 76.8% nights, p < 0.001), overall Injury Severity Score was 13.7 (13.7 weekdays vs. 13.6 weekends, p = NS and 13.7 days vs. 13.3 nights, p = NS), and overall Glasgow Coma Scale score was 13.6 (13.5 weekdays vs. 13.6 weekends, p = NS and 13.7 days vs. 13.4 nights, p < 0.05). In unadjusted analyses, no survival difference was detected for patients presenting on weekends (5.2% vs. 5.3%; odds ratio [OR], 0.98; and 95% confidence interval [CI], 0.75-1.28) or at night (4.4% vs. 5.5%; OR, 0.81; and 95% CI, 0.58-1.11). In adjusted analyses controlling for age, sex, Injury Severity Score, Glasgow Coma Scale score, and arrival hypotension, no survival difference was detected on weekends (OR, 1.03 and 95% CI, 0.71-1.51) or at night (OR, 0.79 and 95% CI, 0.49-1.25).

Conclusion: Differential mortality on off-hours is not seen at our Level I trauma center. Outcomes that are independent of time of day and day of week may be because of the explicit requirements for trauma centers to be fully staffed and operational at all times. There are implications for staffing and systems solutions for other time-sensitive disease including ST elevation myocardial infarction, stroke, and cardiac arrest. Interventions may include the development of a categorization system based on emergency care capabilities, development of explicit staffing requirements, and requiring an emergency care-specific quality improvement program.