Serological testing is an important tool in the diagnostic work-up of suspe
cted celiac disease. Our aim was to apply a decision analysis model to comp
are the costs of serological testing versus small bowel biopsy in the diagn
ostic work-up of celiac disease. A cost-minimization approach was employed.
A decision analysis model with three diagnostic arms was designed using Da
ta Version 3.5: anti-gliadin antibody versus endomysial antibody versus sma
ll bowel biopsy. Response to gluten-free diet was considered diagnostic of
celiac disease; lack of response prompted a small bowel biopsy to definitiv
ely exclude celiac disease. Baseline probabilities were varied using sensit
ivity analysis. Sensitivity analysis revealed that the endomysial antibody
strategy was least costly, provided the prevalence of celiac disease was le
ss than 42%; above this anti-gliadin antibody became the most economical op
tion. In conclusion, initial screening with endomysial antibody is the leas
t costly strategy for diagnosing celiac disease in a low risk population. A
ntigliadin antibody becomes the cheaper strategy for higher risk population
s.