Seroprevalence, correlates, and characteristics of undetected coeliac disease in England

Abstract

Objective: To examine the seroprevalence, correlates, and characteristics of undetected coeliac disease in a large adult population sample in Cambridge, UK.

Methods: The Cambridge General Practice Health Study invited individuals from 12 general practices, aged 45-76 years, to attend for a health survey that included a bone density measurement, between 1990 and 1995. A total of 7550 participants' serum samples were tested for antiendomysial antibody (EMA). Seroprevalence of undetected coeliac disease was based on EMA positivity. Differences between EMA positive and negative participants of various physiological correlates and reported characteristics were estimated by multivariate logistic and linear regression and adjusted for age, sex, social class, and smoking behaviour.

Results: The seroprevalence of undetected coeliac disease in this general population sample aged 45-76 was 1.2% (95% confidence interval (CI) 0.9-1.4). EMA positive participants (n=87) were on average slightly lighter by 2.2 kg (p=0.08), were more likely to have reported their general health as being "good or excellent" (odds ratio (OR) 1.76 (95% CI 0.90-3.46)), and were less likely to report being a current smoker (OR for current versus never 0.36 (95% CI 0.14-0.90)) than EMA negative participants. EMA positivity was associated with an 8% reduction in mean serum cholesterol (0.5 mmol/l; p<0.01) and reductions in mean haemoglobin (0.3 g/dl; p<0.01), total protein (1.0 g/l; p<0.05), and corrected serum calcium (0.02 mmol/l; p<0.05). There was an increased risk of osteoporosis in EMA positive participants (OR 3.1 (95% CI 1.3-7.2)) and of mild anaemia (OR 4.6 (95% CI 2.5-8.2)) compared with EMA negative participants.

Conclusions: Undetected coeliac disease is likely to affect approximately 1% of the population of England aged 45-76 years, a value similar to several other countries. Those affected report "better health" but they do have an increased risk of osteoporosis and mild anaemia. In contrast, they have a favourable cardiovascular risk profile that may afford protection from ischaemic heart disease and stroke.

Figure 1

Overview of study participants. *Participants with probable coeliac disease (treated)—those who reported taking a gluten free diet and having a medical condition coded as malabsorption (including coeliac disease) and who were antiendomysial antibody (EMA) negative. †Participants with probable coeliac disease (untreated)—those who reported having a medical condition coded as malabsorption (including coeliac disease) but did not report taking a gluten free diet, and who were EMA positive. ‡Participants with possible coeliac disease—those who reported taking a gluten free diet but did not report having a medical condition coded as malabsorption (including coeliac disease) and who were EMA negative. §Participants not previously diagnosed with coeliac disease—those that did not report being on a gluten free diet, or having any other medical condition coded as malabsorption (including coeliac disease). This group was further subdivided into those who were EMA negative (no evidence of coeliac disease) and those who were EMA positive (undetected coeliac disease). The latter group was further subdivided on the basis of human antitissue transglutaminase antibody (tTGA) results