Does Weekend Hospital Admission Affect Upper Gastrointestinal Hemorrhage Outcomes?: A Systematic Review and Network Meta-Analysis.
J Clin Gastroenterol. 2018 Aug 16;:
Authors: Liu L, Hao D, Liu W, Wang L, Wang X
BACKGROUND: Compared with weekday admissions, weekend admissions are consistently associated with worse patient outcomes, known as the "weekend effect." The weekend effect may have adverse health consequences, including death. To determine the potential impact of the weekend effect on primary (ie, mortality) and secondary outcomes of patients with upper gastrointestinal hemorrhage (UGIH).
MATERIALS AND METHODS: This was a network meta-analysis based on cohort studies. Databases were searched for studies published up to April 2018. The predefined primary outcome was mortality (30-d mortality and in-hospital mortality). The secondary efficacy outcomes were rebleeding rates, use of endoscopic therapy, need for surgery or angiography, mean length of hospital stay, and time to endoscopy. The study protocol was registered with PROSPERO (No. CRD42018094660).
RESULTS: In total, 25 studies, including 28 analyses (N=1,203,202 patients), were eligible. The results revealed a tendency toward increased 30-day mortality and increased in-hospital mortality among weekend admissions. In a subgroup analysis, there were significance differences in mortality according to the study location (ie, Europe) and UGIH type (ie, variceal UGIH), with these subgroups having elevated mortality rates. Moreover, weekday admissions were associated with a significant decrease in rebleeding rates. In the network meta-analysis, the study location (in Europe or Asia) and type of UGIH (ie, variceal UGIH) were associated with an increased likelihood of high in-hospital mortality among weekend admissions.
CONCLUSIONS: The evidence derived from this network meta-analysis supports the idea that weekend admissions are associated with an increased risk of death, especially among variceal UGIH patients in European hospitals.
PMID: 30119093 [PubMed - as supplied by publisher]