Demographic and Medical Features of Inpatients Who Avoid Gluten: A Case-control Study

Introduction: Although an increasing number of people adhere to a gluten-free diet, non-celiac gluten sensitivity remains poorly understood. Many patients not meeting the criteria for celiac disease nevertheless have intestinal and extraintestinal symptoms that they manage by avoiding gluten. Others may avoid gluten due to the belief that this diet confers other medical benefits. The specific medical diagnoses associated with gluten avoidance among patients without celiac disease have not been investigated.Table: Table. Demographic characteristicsTable: Table. Multivariable Analysis of ComorbiditiesMethods: All hospitalized patients, adults and children, at Columbia University Medical Center ordered for a gluten-free diet in 2011-2016 were identified, excluding those with a documented diagnosis of celiac disease based on history, duodenal biopsy, or positive celiac disease serologies. We matched by age and gender each gluten-free patient to 2 inpatients on a regular diet. Prevalences of comorbidities including cardiovascular, gastrointestinal, neurological, and psychiatric disorders were then identified for the gluten-free patients vs. controls. We used conditional logistic regression adjusting for race, ethnicity, and insurance status to determine the odds ratio (OR) for each comorbidity of being on a gluten-free diet. Results: Of the 769 inpatients on a gluten-free diet, the majority (63.6%) did not have celiac disease (Fig. 1). Gluten-free patients were more likely than controls to be non-Hispanic whites (OR 2.92, P<0.001). They were more likely to be admitted to the neurology service (OR 1.74, P=0.015) or pediatric neurology (OR 5.57, P<0.001). On multivariable analysis, they were less likely to have hypertension (OR 0.38, P<0.001) or diabetes (OR 0.58, P=0.015), and more likely to have inflammatory bowel disease (OR 1.56, P=0.041), irritable bowel syndrome (OR 6.16, P<0.001), hyperthyroidism (OR 2.73, P=0.014), hypothyroidism (OR 2.06, P<0.001), lupus (OR 2.87, P=0.027), and autism spectrum disorder (OR 23.42, P<0.001).Figure: The light gray bar represents the number of hospitalized patients without Celiac disease who received a gluten-free diet per year. Dark gray represents the number of hospitalized Celiac disease patients receiving gluten-free diet orders per year.Conclusion: Non-celiac gluten avoiding patients are more likely than controls to be non-Hispanic whites. They have a lower prevalence of hypertension and diabetes, but higher prevalence of inflammatory bowel disease, irritable bowel syndrome, thyroid disease, lupus, and autism spectrum disorder. This suggests that patients with these difficult to manage and often highly symptomatic disorders have turned to a gluten-free diet for perceived benefit. More research is needed to evaluate the efficacy of avoiding gluten in these disorders.