I read with interest the debate on population based screening for coeliac disease (Gut 2003:52;168–9 and 170–1). Antagonists of population based screening hold the view that there is no evidence that screening the general population or instituting a gluten free diet in asymptomatic coeliac disease will reduce mortality.

Contrary to early beliefs, coeliac disease is one of the most common disorders affecting up to 1% of the general population, regardless of ethnic or geographical origin. In patients with coeliac disease, mortality rate is higher than the general population by a factor of 1.9–3.8, mainly due to complications (table 1) such as malignancy.^1–3

The important fact is that this increased mortality can be reduced to that of the general population after 1–5 years on a gluten free diet.⁴ Moreover, there has been concern about the increased risk of osteoporosis and concurrent autoimmune disease in patients with untreated coeliac disease.⁵

Recently available sensitive serological assays have led to awareness that the typical form of coeliac disease (diarrhoea, weight loss, abdominal distension, and failure to thrive) represents only a small proportion of coeliac disease patients. Additionally, most diagnosed cases in adult life have an atypical presentation (table 1) or often clinically silent disease.⁶

A combination of high prevalence and atypical presentation (table 1) has led to under diagnosis of this condition, resulting in a ratio of known (previously diagnosed) to undiagnosed coeliac disease cases as high as 1 to 7. Cases detected with screening usually manifest atypical or minimal complaints, such as abdominal pain, fatigue, mood changes, and iron deficiency.

It is important to recognise that some asymptomatic undiagnosed cases may emerge later on due to development of symptoms. This aspect of coeliac disease has been clearly shown in a recent Finnish study in a cohort of 3654 children. At the time of the first blood collection in 1994, none of the subject had received a diagnosis of coeliac disease. However, 56 (1.5%) cases had a positive test for coeliac disease when antitissue transglutaminase (tTG) and antiendomysial antibody tests were performed on sera seven years later. Interestingly, 10 of 56 (17%) anti-tTG positive cases had already received a diagnosis between 1994 and 2001 because of abdominal complaints.⁷

These results raise many questions. Although the natural history of the silent form of coeliac disease remains unclear, there is growing evidence that, if untreated, it may be associated with symptoms and complications such as anaemia, osteoporosis, stunted growth, autoimmune disease, and small bowel lymphoma, leading to increased mortality compared with the general population.

The ethical dilemma that the medical world faces today is whether or not to turn our backs on one of the most common genetic conditions, that is easy to diagnose, and in which complications could be prevented by adherence to a gluten free diet. In view of the usual delay in diagnosis and the high morbidity related to untreated coeliac disease, there should be a low threshold for usage of serological testing in both primary and secondary care settings. There is also a pressing need to have an open debate not only within the medical community but also involving the general public to address the appropriateness of population based screening of this easily treatable common genetic condition.