Quantifying Lives Lost Due to Variability in Emergency General Surgery (EGS) Outcomes: Why We Need A National EGS Quality Improvement Program.

2021
Background Nearly four million Americans present to hospitals with conditions requiring emergency general surgery annually, facing significant morbidity and mortality. Unlike elective surgery and trauma, there is no dedicated national quality improvement program to improve Emergency General Surgery (EGS) outcomes. Our objective is to estimate the number of excess deaths that could potentially be averted through EGS quality improvement in the United States. Methods Adults with the American Association for the Surgery of Trauma-defined EGS diagnoses were identified in the Nationwide Emergency Department Sample 2006-2014. Hierarchical logistic regression was performed to benchmark treating hospitals into reliability adjusted mortality quintiles. Weighted generalized linear modeling was used to calculate the relative-risk of mortality at each hospital quintile, relative to best-performing quintile. We then calculated the number of excess deaths at each hospital quintile versus the best-performing quintile using techniques previously used to quantify potentially preventable trauma deaths. Results Twenty six million EGS patients were admitted and 6.5 million (25%) underwent an operation. In-hospital mortality varied from 0.3% to 4.1% across the treating hospitals. Relative to the best performing hospital quintile, an estimated 158,177 (153,509-162-736) excess EGS deaths occurred at lower-performing hospital quintiles. Overall, 47% of excess deaths occurred at the worst-performing hospitals, while 27% of all excess deaths occurred among the operative cohort.Conclusion:Nearly 200,000 excess EGS deaths occur across the US each decade. A national initiative to enable structures and processes-of-care associated with optimal EGS outcomes is urgently needed to achieve "Zero Preventable Deaths after Emergency General Surgery." Level of evidence III, epidemiological.
    • Correction
    • Source
    • Cite
    • Save
    16
    References
    1
    Citations
    NaN
    KQI
    []
    Baidu
    map