Role of diet in prevention versus treatment of Crohn's disease and ulcerative colitis.

Diet is a modifiable risk factor for disease course and data over the past decade have emerged to indicate its role in Crohn's disease (CD) and ulcerative colitis (UC). However, literature is riddled with misinterpretation of data, often leading to unexpected or conflicting results. The key understanding is that causative factors in disease development do not always proceed to an opportunity to change disease course, once established. Here, we discuss the data on dietary influences in three distinct disease states for CD and UC-predisease, active disease and quiescent disease. We appraise the literature for how our dietary recommendations should be shaped to prevent disease development and if or how that differs for CD and UC induction therapy and maintenance therapy. In UC, principles of healthy eating are likely to play a role in all states of disease. Conversely, data linking dietary factors to CD prevention and treatment are paradoxical with the highest quality evidence for CD treatment being exclusive enteral nutrition, a lactose, gluten and fibre-free diet comprising solely of ultraprocessed food-all dietary factors that are not associated or inversely associated with CD prevention. High-quality evidence from dietary trials is much awaited to expand our understanding and ultimately lead our dietary recommendations for targeted patient populations.

Offering Guidance and Learning to Prescribers to Initiate Parenteral Nutrition using a Validated Electronic Decision TREE (OLIVE TREE).

Parenteral nutrition (PN) is recommended in patients nutritionally at risk and unable to receive oral or enteral nutrition. A standardized electronic PN order format could enhance appropriate PN prescribing. We developed the OLIVE TREE (Offering guidance and Learning to prescribers to Initiate PN using a Validated Electronic decision TREE), embedded in our electronic health record. We aimed to evaluate its validity and impact on physicians' prescribing behavior. A non-randomized before-after study was carried out in a tertiary care center. The OLIVE TREE comprises 120 individual items. A process validation was performed to determine interrater agreement between a pharmacist and the treating physician. To estimate the proportion of patients for whom the OLIVE TREE had an effective and potential impact on physicians' prescribing behavior, a proof of concept study was conducted. The proportion of patients for whom PN was averted and the proportion of decisions not in line with the recommendation were also calculated. The process validation in 20 patients resulted in an interrater agreement of 95.0%. The proof of concept in 73 patients resulted in an effective and potential impact on prescribing behavior in 50.7% and 79.5% of these patients, respectively. Initiation of PN was not averted and recommendations of the OLIVE TREE were overruled in 42.5% of the patients. Our newly developed OLIVE TREE has a good process validity. A substantial impact on prescribing behavior was observed, although initiation of PN was not avoided. In the next phase, the decision tree will be implemented hospital-wide.