Children's Hospital Colorado
Advances and Answers in Pediatric Health
U.S. News & World Report ranked in all 10 specialties badge

Advancing child health and pediatric specialty care through clinical discovery, multidisciplinary research and innovation

Study of Celiac Disease in European, U.S. Children Shows Incidence Varies by Region



Key takeaways

  • Celiac disease was common in all studied regions.

  • There was regional variance despite the HLA genotype pointing to the importance of environmental risk factors.

  • Findings of high cumulative incidence of celiac disease warrant a low threshold to screen for it when clinically appropriate.

Background: incidence of celiac disease among U.S. and European children

Many cases of celiac disease can occur without a family history or classic disease symptoms and diagnostic approaches vary by region, so it is difficult to determine the true incidence.

A previous study conducted by Children’s Hospital Colorado researchers reported the 15-year cumulative incidence of biopsy-proven celiac disease in the general pediatric population at 1.9% –– and higher at 3.1% –– when including those who did not undergo a biopsy but had high and persistent autoantibody levels.

The Environmental Determinants of Diabetes in the Young (TEDDY) study is an ongoing multinational prospective birth cohort study to understand the environmental factors that contribute to celiac disease and Type 1 diabetes (T1D).

  • Longitudinal monitoring for celiac disease autoimmunity with autoantibodies to tissue transglutaminase (tTGA)
    • Caused by several factors and can be present without celiac disease
      • Transient or fluctuating antibody levels
      • Low antibody levels without intestinal injury by biopsy
      • Dietary changes influencing additional evaluation
    • 6 clinical research sites, including Children’s Colorado
    • 4 countries
      • United States
        • Colorado
        • Georgia
        • Washington
      • Finland
      • Germany
      • Sweden
    • Represents highest-risk HLA-DQ haplogenotypes for celiac disease

Authors of this study, Marian Rewers, MD, PhD, and Jill Norris, PhD, MPH, including researchers Marisa Stahl, MD, and Edwin Liu, MD, from the Colorado Center for Celiac Disease at Children’s Colorado, utilized this unique cohort of children at high genetic risk for celiac disease to understand differences in population-wide cumulative incidence of celiac disease autoimmunity and celiac disease in the U.S. and Europe.

Methods: celiac disease evaluation and celiac disease autoimmunity population-specific estimates

Study design and population

  • 8,676 children enrolled in TEDDY between September 2004 and February 2010 with eligible haplotypes
    • DQ2.5/DQ2.5
    • DQ2.5/DQ8.1
    • DQ8.1/DQ8.1
    • DQ8.1/DQ4.2
  • Beginning at age 2, participants screened annually for tTGA
    • At first tTGA positive result, all prior collected samples tested for tTGA to determine earliest timepoint of autoimmunity
  • Celiac disease autoimmunity
    • Primary study outcome
    • Defined as positivity (>99th percentile of healthy control sera) in 2 consecutive tTGA tests at least three months apart.
  • Celiac disease
    • Defined a positive blood serum and duodenal biopsy Marsh score >2, or
    • Average tTGA of ≥100 units from two consecutive positive sera without biopsy
  • Stem Cell Donor Registries determined major celiac risk-associated HLA-DQ haplogenotypes at each TEDDY site

Results: risks for celiac disease autoimmunity and celiac disease varied by region

The final study cohort comprised of participant data as of July 31, 2020:

  • 6,628 HLA-typed eligible children
    • DQ2.5, DQ8.1, or both haplotypes
    • Underwent 1 or more tTGA tests
  • 11.5 years median follow-up
  • 580 (9%) children first-degree relative (FDR) with T1D
  • 317 (5%) reported FDR with celiac disease
  • 1,299 (20%) met celiac disease autoimmunity
  • Outcome
    • 41 months median age
  • 529 children (8%) met study diagnostic criteria for celiac disease
  • Previously reported data from TEDDY found the majority of children with celiac disease autoimmunity were asymptomatic

Cumulative risk for celiac disease autoimmunity and celiac disease by site and haplogenotype

  • Compared with the three highest risk HLA-DQ haplogenotypes
  • Risk substantially varied by haplogenotype, clinical center
  • Relative risk by region same regardless of haplogenotype


  • Highest risk in each region, all sites, haplogenotypes
  • Mirrors previous TEDDY reports
  • Among DQ2.5 subjects:
    • 63.1% developed celiac disease autoimmunity by age 10
    • 28.3% developed celiac disease by age 10


  • Lowest disease burden for each region for all haplogenotypes

Finland compared to Colorado

  • Finland has consistently higher incidence of celiac disease autoimmunity
    • 60.4% vs. 50.9% among DQ2.5 subjects
  • Finland has lower incidence of celiac disease
    • 20.3% vs. 22.6% among DQ2.5 subjects

Stem cell donor registries for haplogenotypes frequencies

  • Most common HLA-DQ haplogenotypes in general population
    • DQ2.5/X
    • DQ8.1/X
  • Sweden highest frequency (41.6%)
  • Georgia lowest frequency (34.4%)

Site-specific population cumulative incidence adjusted by regional frequencies of haplogenotypes

  • DQ2.5 homozygosity conferred highest risk of celiac disease autoimmunity and celiac disease
  • DQ2.5 heterozygotes contributed most to cumulative incidence
  • 10-year cumulative incidence highest in Sweden (celiac disease autoimmunity 8.4%, celiac disease 3.0%).
  • Within U.S., Colorado had highest cumulative incidence for both endpoints (celiac disease 6.5%; celiac disease 2.4%)

Sensitivity analyses with lower non-biopsy tTGA threshold for celiac disease

  • When tTGA cutoff to define celiac disease lowered to average two-visit tTGA ≥ 67.4, more children met study serologic criteria for celiac disease
    • Statistically significant differences in risk of celiac disease between clinical sites, countries still observed
    • Regional differences in celiac disease incidence not explained by differential biopsy rates

Discussion and conclusion: Swedish, U.S. children had highest incidence of celiac disease

Several differences remained after adjusting for key factors including HLA haplogenotypes, sex and family history:

  • Highly variable cumulative incidence of celiac disease by age 10 in the general population
    • 0.9% Washington
    • 2.4% Colorado
    • ≤ 3% Sweden
  • Celiac disease autoimmunity rates even higher, demonstrating celiac disease progression has not occurred in all who develop autoimmunity

Study findings were consistent with recent meta-analysis of world-wide incidence data suggesting celiac disease incidence continues to rise, is highest amongst children but depends on region.

Specific and notable findings

  • Disparities in cumulative incidence –– even within U.S. sites
  • Prospective, uniform screening confirmed differences between U.S. and Europe
  • Specific HLA haplogenotype risk does not convey same risk for disease from region to region
  • Finland may have higher rates of low-level or transient autoimmunity or differing environmental factors increasing progression to celiac disease
    • Increased rates of probiotic supplementation in first year of life

Factors contributing to development of celiac disease autoimmunity and celiac disease

There is not yet full understanding of the role the environment plays in the development of celiac disease autoimmunity and celiac disease.

  • Multiple environmental factors likely –– or even more likely an interaction of them –– explain striking differences in autoimmunity amongst regions
  • Non-HLA, epigenetic influences also likely to moderate development

Study authors noted several conclusions

  • The high cumulative incidence of celiac disease, which varies by region, informs future screening practices.
  • It is important for clinicians to have a low threshold for celiac disease screening in the appropriate clinical setting.
  • Ongoing analyses of environmental, genetic, epigenetic exposures in TEDDY cohort will help improve understanding of regional differences in celiac disease and celiac disease autoimmunity.