Prevalence of inflammatory bowel disease in British 26 year olds: national longitudinal birth cohortBMJ 1998; 316 doi: https://doi.org/10.1136/bmj.316.7137.1058 (Published 04 April 1998) Cite this as: BMJ 1998;316:1058
- S M Montgomery, research fellow (, )
- D L Morris, Wellcome training fellow in clinical epidemiology,
- N P Thompson, British Digestive Foundation research fellow,
- J Subhani, clinical fellow in gastroenterology,
- R E Pounder, professor of medicine,
- A J Wakefield, reader in experimental gastroenterology
- Correspondence to: Dr Montgomery
- Accepted 21 November 1997
Inflammatory bowel disease has become more common in developed countries this century. Mayberry et al reported incidences of Crohn's disease in Wales of 0.18 cases/105/year in the 1930s and 5.95 cases/105/year in the 1970s.1 We investigated the prevalence of inflammatory bowel disease at age 26 years in a nationally representative birth cohort. Associations of sex and social class with risk of the disease have previously been shown,1-3 and these were also investigated.
Subjects, methods, and results
A postal survey of the 1970 British cohort study was conducted in 1995-6 among individuals aged 25 or 26 years, asking if respondents had a diagnosis of Crohn's disease or ulcerative colitis. The cohort study is a longitudinal study of those living in England, Scotland, and Wales born 5 to 11 April 1970.4 The target population was estimated as 16 000, and we sent questionnaires to the 13 099 cohort members whom we traced. In all, 9803 completed questionnaires were returned; 309 addresses were identified as no longer current; and 12 people refused to participate. Excluding invalid and untraced addresses, the response rate was 77%. The social statistics research unit at City University, London, provided most (7430) of the addresses. To minimise bias, we traced the remaining 2373 cohort members through a letter forwarding service provided by the Driver and Vehicle Licensing Agency. The cohort remained largely representative, with some loss from the most disadvantaged groups: the proportion in social class V at birth dropped from 6.4% to 4.7% in the respondents.
Cohort members who reported inflammatory bowel disease were contacted again for details of their diagnosis and permission to contact their physicians. If permission was not granted, diagnosis was not confirmed. The registrar general's social class was based on father's occupation, collected prospectively in 1970.
The table shows the prevalence of Crohn's disease and ulcerative colitis in the cohort, by social class and age. Thirty two and 27 cohort members reported Crohn's disease and ulcerative colitis respectively. For two reports of Crohn's disease and five of ulcerative colitis, the diagnosis was subsequently refuted by cohort members themselves or their physicians. The diagnosis was confirmed for 21 cohort members with Crohn's disease and 12 with ulcerative colitis. On the basis of physician confirmed cases only, the prevalence per 10 000 was 21.4 (95% confidence interval 12.3 to 30.6) for Crohn's disease, 12.24 (5.3 to 19.2) for ulcerative colitis, and 33.7 (22.2 to 45.1) for inflammatory bowel disease. If it is assumed that the unconfirmed cases had the same disease specific, false positive rates as the entire sample, the estimated prevalences per 10 000 were 29.8 (19.0 to 40.6), 19.4 (10.5 to 28.1), and 49.2 (35.3 to 63.0) respectively.
Social class was modelled by using logistic regression, both as a six category ordinal variable and as a binary (manual v non-manual) dummy. Neither social class nor sex was significantly associated with Crohn's disease, ulcerative colitis, or both diseases combined (P>0.1).
We found a higher prevalence for Crohn's disease and for ulcerative colitis than other studies in Britain have found for comparable age groups (Keighley et al found a prevalence of 6.49/10 000 for Crohn's disease among 25-29 year olds in 19732 and Evans et al 7.59/10 000 for ulcerative colitis among 25-34 year olds in 19603). In part, this may be because a general population based sample was used, but it is also likely to reflect a genuine rise in the prevalence of inflammatory bowel disease, particularly for Crohn's disease. The lack of significant association of both social class and sex with inflammatory bowel disease may be a function of the small number of cases. Alternatively, there may be a homogenisation of the pattern of exposure to risk factors for inflammatory bowel disease that reflects improved material conditions in infancy in comparison with those born earlier this century: improved conditions in early life have been identified as a risk for later inflammatory bowel disease.5 This is a relatively young cohort, and we expect the prevalence of inflammatory bowel disease to continue rising both in the 1970 British cohort study and in the general population.
We are grateful for help from the staff of the social statistics research unit, City University, London, and the staff of the Driver and Vehicle Licensing Agency, Swansea.
Contributors: SMM wrote the original draft of the paper, planned the data collection, was responsible for the data analysis, participated in all components of the study, and will act as guarantor for the paper. All authors contributed to the design, data interpretation, and writing of the paper. DLM was responsible for ensuring confirmation of the diagnosis of Crohn's disease or ulcerative colitis in cohort members who reported having inflammatory bowel disease. NPT and JS assisted in data collection and data preparation. REP and AJW were responsible for originating and overseeing the study.
Funding: This work was supported by the Hayward Foundation and the Enid Linden Trust.
Conflict of interests: None.