Frequency and Variables Associated with Fasting Orders in Inpatients with Ulcerative Colitis: The Audit of Diet Orders - Ulcerative Colitis (ADORE-UC) Study

Introduction: Current clinical practice guidelines suggest that patients with ulcerative colitis (UC) hospitalized because of a disease flare should be offered a normal diet, unless such a diet is not tolerated. Studies of hospitalized patients have demonstrated iatrogenic malnutrition from unjustified or inappropriate nil per os (NPO) or clear liquid diet (CLD) orders. In this study, we aim to characterize the burden of this problem in hospitalized patients with UC. Methods: We conducted a retrospective cohort study of all patients with UC admitted to the gastroenterology service or the general internal medicine service at a tertiary, academic hospital between January 2009 and December 2014, with a length of stay between 2 and 30 days. The frequency and duration of bowel rest and CLD orders was recorded, and the number of meals missed because of these orders was assessed. NPO or CLD diet orders were considered justified if the patient had intractable nausea or vomiting, pancreatitis, bowel obstruction, toxic megacolon or were awaiting endoscopy, or if alternative enteral nutrition was provided. Clinical and demographic factors associated with unjustified underfeeding were identified. Results: A total of 187 admissions among 158 patients with UC were identified during the study period and included in the final analysis. Most admissions were to the gastroenterology service (148/187, 79.1%). The mean age at admission was 35.0 years (SD = 15), and 83/158 (52.5%) were female. The median length of stay was 8 days (interquartile range = 4-12). Registered dietician consultation was obtained in only 32 admissions (17.1%), and admission weight was recorded in only 68 (36.4%) admissions. A total of 252 NPO or CLD dietary orders were encountered in 142 admissions (75.9%). Of those, 112 orders were unjustified (44%). On average, patients with unjustified NPO or CLD orders spent 3 days on an NPO or CLD diet, which corresponded to a mean of 10 missed meals. Characteristics associated with unnecessary fasting included female gender, less frequent endoscopic disease staging, less frequent escalation of therapy to prednisone and/or biologics, and admission to a non-gastroenterology service. Conclusions: There is a high burden of unjustified underfeeding among hospitalized patients with UC, particularly in patients admitted without evidence of objective disease flare. This may lead to nutritional compromise in an at-risk population, and further studies are needed to assess the nutritional impact of unjustified bowel rest on patients with UC. Our findings also suggest that targeted quality improvement interventions are needed to decrease the frequency of inappropriate bowel rest among hospitalized patients with UC.