A Case Control Study Investigating Drinking Water & Dairy Products in the Aetiology of Crohn’s Disease – A Possible Role for Mycobacterium avium paratuberculosis – The CMAW Study
DWI 0841

July 2005


This report presents the final results of a case control study investigating the role of drinking water potentially contaminated with Mycobacterium avium paratuberculosis (MAP) in the aetiology of Crohn’s disease (CD). Patients with CD were recruited from five centres in England and their exposure to proxy measures of MAP contamination was compared with that of controls from a similar geographical area to examine potential risk factors for CD. Exposure to water, milk or dairy products potentially contaminated with MAP will be associated with CD was the primary hypothesis under investigation. Geographical information systems (GIS) software was used to estimate a risk score based on the type of water treatment and source of water supplied to each individual. Odds ratios (OR) for each risk factor were calculated using logistic regression models. Multivariable logistic regression was used to control for the confounding effect of known risk factors. A total of 218 cases and 812 controls were recruited. No significant association was observed between measures of potential contamination of water sources with MAP (surface versus ground water OR 1.08, 95 % CI 0.86 – 1.37), water intake (boiled water OR 0.99, 95 % CI 0.88 – 1.11, or unboiled water OR 1.01, 95 % CI 0.90 – 1.15) or water treatment (sedimentation OR 0.84, 95 % CI 0.31 – 2.28, coagulation OR 0.96, 95 % CI 0.58 – 1.58, flocculation OR 0.73, 95 % CI 0.44 – 1.20 and filtration OR 0.78, 95 % CI 0.51 – 1.18). Neither was there an association with bottled water (OR 0.99, 95 % CI 0.89 – 1.09). Although dairy product intake (OR 0.96, 95 % CI 0.85 – 1.07) was not associated with an effect, consumption of pasteurized milk (OR 0.82, 95 % CI 0.69 – 0.97) was associated with a reduced risk of CD.

In addition to the primary hypotheses tested several other variables remained in the final model. Both smoking (OR 1.31, 95 % CI 1.12 – 1.53) and family history (OR 7.13, 95 % CI 3.37 – 15.08) of CD, two well known risk factors, were significantly associated with an increased risk. In addition, holidays abroad (OR 0.52, 95 % CI 0.32 – 0.84) and fruit consumption (OR 0.78, 95 % CI 0.67 – 0.92) were negatively associated with risk whilst meat intake (OR 1.40, 95 % CI 1.17 – 1.67) was associated with an increased risk of developing CD. Sub group analysis showed non-stricturing non-penetrating CD to be associated with increased meat intake (OR 1.48, 95 % CI 1.19 – 1.85) while stricturing and penetrating disease were not. CD affecting only the ileocolon was significantly associated with increased meat intake (OR 1.58, 95 % CI 1.26 – 1.98). CD of the upper GI was the only disease site significantly associated with farm holidays (OR 3.55, 95 % CI 1.17 – 10.79) and oral contraceptive use (OR 1.17, 95 % CI 1.01 – 1.36).

This study does not provide any evidence of an increased risk of CD in association with exposure to the two primary hypotheses (drinking water and milk) for transmission pathways of MAP. Consequently this study does not support a role for MAP or drinking water in the aetiology of CD. This conclusion is further supported by the lack of any association with farm holidays and contact with farm animals. Although fruit consumption was not one of the primary hypotheses, the negative association with CD has been noted in previous studies, adding weight to this finding. To our knowledge this is the first study to identify a negative association with travel abroad or a positive association with meat consumption. As these observations were not part of the primary hypothesis care must be exercised in their interpretation until confirmed in subsequent research, as such findings could have arisen by chance. If the association with meat consumption is confirmed, we do not consider that this would indicate an infectious aetiology, but possibly a link with total animal protein consumption, a finding that has been noted previously.

Copies of this report may be available as an Acrobat pdf download under the 'Post 2000 Reports' heading on the DWI website.