N. Goel1, L. Gutnik1, G. Reznor1, D. Rivera Morales1, A. Salim1, M. J. Zinner1, Z. R. Cooper1 1Brigham And Women’s Hospital,Trauma, Burns And Surgical Critical Care,Boston, MA, USA
Introduction:
The Two-Midnight Rule established by the Centers for Medicare and Medicaid Services in 2013, states that inpatient tests, treatments, and services are eligible for payment only if the hospital stay crosses two midnights (MN). Therefore, to avoid non-payment from Medicare, it is critically important for surgeons to correctly identify patients who will stay less than 2 MN (observation status) at the time of admission. Small bowel obstruction (SBO) is the most common cause of surgical admission with estimated costs of over $1 billion per year. The purpose of this study was to identify patient factors predictive of a hospital stay less than two midnights (MN) among patients > 65 years old who present to the emergency room with small bowel obstruction.
Methods:
A retrospective review of patients > 65 years old admitted to a tertiary academic medical center with small bowel obstruction from 2006-2013. We used data from the Research Depository of Patient Records (RPDR) to identify patients with small bowel obstruction using ICD-9 codes (560.0 intussusception, 560.1 paralytic ileus, 560.81 adhesions (intestinal/peritoneal), 560.8 other specified obstruction, 560.9 other unspecified obstruction). We used chart review to collect patient demographics (age, race, sex, zip code) and clinical characteristics (neoplasm, inflammatory bowel disease (IBD), recent surgery, laboratory values and radiographic findings including prior anastomosis, adhesions, mass or tumor, or other) at the time of presentation to the emergency room. The unit of analysis was the hospital admission for SBO. Chi-square tests were used to compare differences between patients admitted for < 2 MN and > 2 MN and multivariate logistic regression was used to identify predictors of admission for < 2 MN.
Results:
Among 855 older patients admitted with SBO, 39 (4.8%) stayed < 2 MN. As compared to longer hospital stay, stay < 2MN was associated with age 65-74 years (71.8% vs. 53.1%, p=0.03), IBD (10.3% vs. 1.6%, p=0.006) and prior anastomosis on CT(15.4% vs. 3.2%, (p=0.001). However, only a minority of patients staying < 2MN had any of these characteristics. In multivariate logistic regression, age 65-74 (OR 2.10, 95% CI 1.02-4.32), IBD (OR 6.07, 95% CI 1.76-20.91), and prior anastomosis (OR 4.67, 95% CI 1.72-12.60) were predictive.
Conclusion:
Approximately 5% of older patients previously admitted for SBO would not meet current CMS criteria for admission. Although age 65-74 years, presence of IBD, and obstruction due to prior anastomosis, predict stay < 2MN, most patients who stay ≥ 2MN also have these characteristics. Useful identifers of older patients with SBO who will stay < 2 MN are elusive. Under the CMS rules, hospitals are at high risk of reduced payment for some older patients admitted with SBO.