A prospective study of the incidence of childhood celiac disease

J Pediatr. 2003 Sep;143(3):308-14. doi: 10.1067/s0022-3476(03)00282-8.


Objectives: To estimate the frequency of celiac disease (CD) in children in the general population of Denver, Colorado.

Study design: From 22,346 newborns characterized as expressing 0, 1, or 2 HLA-DR3(DQB1*0201) alleles, 987 were selected for a prospective stratified cohort study. Participants were followed for as long as 7 years with serial testing for serum IgA anti-transglutaminase antibodies and for evidence of CD (intestinal mucosal changes or persistent seropositivity).

Results: Of 40 children with at least one positive serologic test, 19 had evidence of CD (10 by biopsy, 9 by persistent seropositivity). Those expressing 0, 1, or 2 HLA-DR3 alleles had, respectively, 0.3% (95% CI, 0.0-2.7), 3.4% (3.0-11.7), and 3.2% (1.0-11.0) risk for evidence of CD by age 5 years. The adjusted risk estimate for evidence of CD by age 5 years for the Denver general population was 0.9% (0.4-2.0), or 1 in 104 (1:49-221). After adjusting for number of HLA-DR3 alleles expressed, risk was higher in females: RR=3.34 (1.00-10.9, P=.048). Evidence of CD was not observed before age 2.6 years.

Conclusions: Celiac disease may affect 0.9% of Denver children by 5 years of age. Children positive for the HLA-DR3 allele and females appear to be at increased risk.

Publication types

  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Celiac Disease / diagnosis*
  • Celiac Disease / epidemiology*
  • Celiac Disease / etiology
  • Child
  • Child, Preschool
  • Cohort Studies
  • Colorado / epidemiology
  • Female
  • Follow-Up Studies
  • HLA-DR Antigens / genetics
  • Humans
  • Incidence
  • Infant
  • Infant, Newborn
  • Longitudinal Studies
  • Male
  • Prospective Studies
  • Risk Factors


  • HLA-DR Antigens