The results of this study are interesting, but probably not large enough to be statistically significant. The total number of people with celiac disease in each group was astounding:
2 out of 50 with IBS (4%)
2 out of 24 with dyspepsia (6%)
2 of the 78 controls (2.6%)
These findings do not necessarily contradict previous IBS/celiac disease studies that looked at hospital outpatients who are more likely to have more severe and prolonged symptoms than a group that selects itself from the general public by responding to a questionnaire. Additionally most of the earlier studies that concluded that there was a connection between celiac disease and IBS were conducted before more recent epidemiological studies that have shown just how high the incidence of celiac disease in the general population is--now estimated between 0.8% and 1.3%--this study suggests 2 -3%. These recent epidemiological studies have also shown that a large percentage of celiacs have little or no symptoms, perhaps due to the length of time or the severity of the disease.
A 1 in 20 diagnosis of celiac disease in patients with IBS/dyspepsia is consistent with other studies, and is still high and suggests that testing for celiac disease should be done routinely on these patients. No studies have ever suggested that all or even most IBS patients have celiac disease, just that the incidence is higher than that of the normal population.
I propose that if this study had been done on exactly the same people several years from now, the 2 people in the control group who were found to have celiac disease may well develop symptoms that would put them in either the IBS or dyspepsia group, which would create a statistically significant result that would contradict this studys result. Last, perhaps the results of this study really support a more broad conclusion: Everyone ought to be screened for celiac disease, not just those with symptoms.