Gastrointestinal bleeding (GIB) remains one of the leading causes of emergency hospital admission for gastrointestinal disorders and continues to represent a major source of morbidity and mortality. Yet, there are limited data on the effect of the specialist care setting on the outcomes of the GIB hospitalized patients. In the present study, we aimed to assess the main care process and clinical indicators of hospitalized GIB patients in the Veneto Region from 2000 to 2024, according to the admission ward.  

A retrospective analysis based on the anonymous regional Hospital Discharge Records database was carried out including all discharges from hospitals operating under the National Health Service in the Veneto Region in the period 2000-2024 with a Diagnosis-Related Group (DRG) code equal to 174 (“Gastrointestinal Haemorrhage with Complications”) or 175 (“Gastrointestinal Haemorrhage without Complications”), or any surgical DRG with a principal discharge diagnosis indicative of GIB. Selected cases were stratified into the site of bleeding (upper, lower, unspecified), the admission area (medical, surgical or other) and, for those hospitalized in the medical area, into gastroenterology ward or different ward (e.g., internal medicine). The mean length of stay (LOS) among patients discharged alive and the in-hospital case-fatality risk (CFR) were calculated. Multivariate logistic regression models were used to assess the association between the admission ward (gastroenterology vs other medical ward) and the in-hospital CFR.

During the 25-year observation period, 99,731 patients were hospitalized for GIB in the Veneto Region: 57.4% in medical area, 39.6% in surgical area and 3.0% in other wards. Males represented 55.2% of them and accounted for the most in each area. The mean age of GIB hospitalized patients was 70.6±18.1 years and they were older in medical area (73.7±15.2 years) than in surgical area (69.2±16.7 years). Upper GIB was the most represented site of bleeding among patients admitted to the medical area (53.0%), whereas lower GIB prevailed among patients admitted to the surgical area (52.5%). 

The overall in-hospital CFR was 5.0% and the mean LOS was 9.3±8.3 days. Both indicators were higher among patients admitted to the medical area compared with those hospitalized in a surgical ward: the CFR was 5.4% and 4.0%, respectively, whereas the mean LOS was 9.9±8.2 days and 8.5±7.8 days, respectively. 

Among GIB patients cared for in the medical area (n=57,204), the mean LOS was 7.1±5.9 days in the gastroenterology wards compared to 10.7±8.6 days in other wards. Inside the medical area, significant difference emerged also for the in-hospital CFR, which was 2.3% in gastroenterology wards and 6.4% elsewhere. After adjustment for all the other examined variables (sex, age, calendar year and need for surgery), the adjusted odds of in-hospital death decreased by more than half among patients with upper GIB admitted to a gastroenterology ward compared to those admitted to a different medical ward (Odds Ratio 0.48, 95% CI 0.40 to 0.58, p<0.0001), and about two-thirds among patients with lower (Odds Ratio 0.33, 95% CI 0.25 to 0.43, p<0.0001) or unspecified GIB (Odds Ratio: 0.34, 95% CI 0.27 to 0.43, p<0.0001). 

The management in a gastroenterology ward appears to be associated with improved care process and clinical outcomes, irrespective of the site of bleeding.  These findings may provide guidance to both health policymakers and clinicians involved in the management of patients presenting with GIB.