Comparison of time to endoscopy and outcome between weekend/weekday hospital admissions in patients with upper GI hemorrhage.

2012 
OBJECTIVE: Recent findings suggest that time to endoscopy is prolonged in patients admitted on the weekend with upper gastrointestinal hemorrhage (UGIH), which may result in increased adverse outcomes. This study was designed to determine if these findings hold true for a community gastroenterology practice. METHODS: This retrospective study reviewed patients admitted to a community teaching hospital from January 1, 2008, through October 31, 2008 with the primary diagnosis of UGIH. UGIH was further defined as acute variceal hemorrhage (AVH) or non-variceal hemorrhage (NVUGIH). The primary groups were based on weekend vs weekday admission. Time to endoscopy, adverse outcomes, presenting symptom, and length of stay were analyzed. RESULTS: One hundred seventy-four patients were included (50 weekend; 124 weekday). Most patients (94.25%) received upper endoscopy within 24 hours of admission. Mean time to endoscopy was shorter for weekend admission compared to weekday (7.52 hours vs 10.82 hours; P=0.012) for the entire group. No statistically significant difference was detected in AVH patients (6.37 hours vs 4.37 hours; P=0.09), but a difference was observed in the NVUGIH group (7.65 hours vs 11.45 hours, P=0.015). Adverse outcomes were not associated with weekend admission (P=0.583). There was no difference in mean length of stay (3.08 days vs 3.85 days; P=0.131) or mean units of blood transfused (2.44 units vs 2.07 units, P=0.417) between admission groups. CONCLUSIONS: Patients admitted to this community teaching hospital with UGIH on the weekend did not experience delayed endoscopy, increased adverse outcomes, or longer length of stay compared to those admitted on a weekday. The previously reported "weekend effect" was not observed. In fact, patients admitted with NVUGIH on the weekend received upper endoscopy earlier than patients admitted during the week.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    20
    References
    19
    Citations
    NaN
    KQI
    []