E. Kwon1, C. Jones1, E. Haut1, J. V. Sakran1, A. J. Kent1 1The Johns Hopkins University School Of Medicine,Department Of Surgery,Baltimore, MD, USA
Introduction:
Laparoscopic Cholecystectomy (LC) is one of the most commonly performed surgeries in the U.S. This study aimed to determine which patient, hospital, and payor characteristics are associated with differences in mortality, hospital utilization, and total billed charges in urgently treated patients with high mortality risk or medical complexity.
Methods:
The Maryland Health Services Cost Review Commission database was queried for adult patients who had urgent or emergent LC between 2008-2017 with medical complexity or mortality risk scores of 3 or greater by Medicare APR DRG classification available in the dataset. Open cholecystectomy was excluded to reduce effects of physiologic outliers and operative practice variation. We examined associations of hospital, patient, and payor characteristics with outcomes of mortality, length of stay (LOS), days in intensive care (IcuLOS), and total charges billed. Multivariate analyses were performed using negative binomial regression modeling, adjusting for individual patient characteristics. Likelihoods reported as adjusted Incidence Rate Ratios (aIRR) were derived from exponentiated negative binomial coefficients.
Results:
We identified a total of 13,885 high-risk patients with urgent LC. As age of the patient increased, LOS, IcuLOS, total charge, and mortality all significantly increased (see table). Mortality was worse in patients who were Asian when compared to White (aIRR=2.23; p<0.05) and in those on Medicare or Medicaid when compared to private insurance (aIRR = 2.26, 2.60; p<0.05). LOS was longer at trauma centers, urban locations, and larger hospitals; African Americans or “other” race and those with Medicare, Medicaid, or “other” insurance also had longer LOS (p<0.001). IcuLOS was increased at urban or trauma centers and at larger hospitals; those on Medicaid had longer IcuLOS (p<0.05). IcuLOS was shorter at academic centers and “other” race. Charges were higher at academic, urban, or trauma centers, larger hospitals; charges were also higher in African Americans or “other” race and in patients with Medicare, Medicaid, or “other” insurance (p<0.001). Self-pay patients had lower charges (aIRR=0.93; p<0.01), and female patients had shorter LOS, shorter IcuLOS, and lower charges.
Conclusion:
Structural hospital, insurance, and demographic factors had substantial and heterogeneous effects on resource utilization, charges, and mortality in high risk patients undergoing urgent LC. These patterns may reflect racial, economic, and geographic disparities. Such effects should be taken into consideration in designing policy and systems to care for complex acute general surgical patients.
