Eczema and Celiac Disease: The All of Us Association Does Not Justify a Gluten-Free Trial Before Testing
Dr. Arsy Editorial education, checked November 9, 2026. Sources do not endorse #003. This does not replace individualized medical care.
Does an eczema diagnosis mean gluten should be removed from the diet? A 2026 US study found an association between eczema and celiac disease, but it did not test a gluten-free diet or diagnose the cause of anyone’s rash. Before changing food intake, it helps to understand both the numbers and how a diet change can interfere with celiac testing.
Recorded diagnoses in a volunteer research population
The All of Us analysis, published online June 8, 2026, used electronic health records from 393,601 adults. Participants were enrolled between May 2017 and October 2023. The researchers identified eczema and celiac disease through diagnostic coding and compared adults with and without recorded eczema. This was cross-sectional research, not follow-up of newly developed celiac disease.
| Finding | Numbers | Interpretation |
|---|---|---|
| Adults with eczema | 175 of 9,488 had recorded celiac disease, approximately 1.8% | Most did not have a recorded celiac diagnosis. |
| Adults without eczema | 2,584 of 384,113, approximately 0.7% | A comparator prevalence, not an annual incidence rate. |
| Unadjusted association | OR 2.77; 95% CI 2.37–3.23 | Other population differences can influence this estimate. |
| Most extensively adjusted model | OR 1.70; 95% CI 1.44–2.00 | Association persisted after measured demographic, autoimmune and atopic conditions were included. |
Why the model changed the estimate
Demographic adjustment produced an OR of 2.67. Adding selected autoimmune conditions reduced it to 2.28; including allergic rhinitis, asthma and eosinophilic esophagitis reduced it further to 1.70. The stepwise pattern shows that the headline estimate depends on what the model accounts for. The final OR is not a 70% probability of celiac disease and cannot establish that eczema caused it.
The database lacked eczema severity information, and diagnoses were not necessarily made by dermatologists. Volunteer recruitment and differences in health-care contact can affect which conditions appear in records. The cross-sectional design cannot establish which condition came first. Proposed shared immune or barrier mechanisms are hypotheses rather than directly tested explanations in this study. NIH database participation is also not NIH endorsement of a product or a screening policy.
Testing comes before an unsupervised gluten experiment
The NIDDK diagnostic guide explains that clinicians use medical and family history, examination and selected tests, commonly blood tests and small-intestine biopsies. It advises against starting a gluten-free diet before diagnostic testing because that can affect the results. If gluten has already been restricted, discuss that fact with the clinician rather than designing your own reintroduction.
Celiac disease is not diagnosed from eczema, digestive symptoms or a risk ratio alone. NIDDK also distinguishes the celiac-associated rash dermatitis herpetiformis, for which skin biopsy may be considered. The study did not demonstrate that all people with eczema need routine screening, or that broad food elimination improves their skin. Symptoms, family history and other risk factors should guide a clinical conversation.
Our thyroid-testing analysis illustrates another association-versus-testing question. Read about vitiligo associations, positive food IgE tests and emotional wellbeing for different care concerns. None substitutes for a celiac diagnosis or a nutritionally appropriate treatment plan.
Dr. Arsy Moisturizer #003 supports cosmetic hydration, moisture-barrier support and ordinary dry-skin comfort. It does not treat celiac disease, gluten-related disorders or eczema. These adult records establish no pediatric or infant suitability for the moisturizer.