Using a practice guideline for safely shortening hospital stay for upper gastrointestinal tract hemorrhage.

Making optimal medical decisions within the constraints of a given health care system requires integrating knowledge, experience, compassion, and patient preferences with the best available clinical research. Evidence-based practice guidelines are an increasingly popular means of helping clinicians assimilate research results into practice. In this issue of JAMA , Hay and colleagues 1 describe a practice guideline developed to determine the time of discharge for patients admitted with acute, nonvariceal upper gastrointestinal tract hemorrhage. In preparing this length-of-stay guideline, these investigators used observational studies from the literature and their own database to create and retrospectively validate a risk-stratification model incorporating 4 variables to predict the likelihood of further bleeding: findings at endoscopy, time elapsed from onset of bleeding, hemodynamics, and comorbidities. 2 See also p 2151. The goal of this and other practice guidelines is to influence some aspect of patient care through modification of clinician behavior. However, it has long