Rapid Responses to:

EDITORIALS: David Forman and Karen J Goodman The epidemiology of stomach cancer: correlating the past with the present BMJ 2000; 320: 1682-1683 [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

An overlooked co-factor?

Maureen Minchin   (5 July 2000)

Gastric cancer and Crohn's disease-contrasting early environments

Danielle Morris   (21 July 2000)

An overlooked co-factor? 5 July 2000
Maureen Minchin Send response to journal: Re: An overlooked co-factor?	In reading this article I was struck by an number of possibly related facts. Poor SES is associated with more than higher rates of infection: in many situations it also means both artificial feeding and greater dummy use. Both of these inevitably mean greater exposure not only to pathogens but also to carcinogens such as nitrosamines and no doubt other chemicals found in teats and dummies right up until the 1980s and longer, as well as greater exposure to water-borne pathogens and carcinogens (asbestos, lead, PVC) at the time that the infant gut is just developing. Infant feeding is a potentially critical intervening variable (or perhaps even causative factor) that is rarely addressed much less teased out well in these studies. Why not?
Gastric cancer and Crohn's disease-contrasting early environments 21 July 2000
Danielle Morris, Wellcome training fellow in Epidemiology Royal Free Hospital Send response to journal: Re: Gastric cancer and Crohn's disease-contrasting early environments	Editor – We note with interest Leon et al (1) geographical correlation between gastric cancer mortality and infant mortality at birth. We have previously published a similar but opposing relationship between infant mortality at birth and incidence of Crohn’s disease: Crohn’s disease incidence being maximal in those countries where infant mortality is lowest (2). We believe that this reflects improvement in hygienic conditions with time, which once reaching a threshold becomes a risk for Crohn’s disease. This may be due to changing exposure pattern to infectious agents in early life: at an older age and lower dose. This is supported by studies of family composition that suggest that having no older siblings is a risk factor for developing Crohn’s disease. (3) In contrast, infection with H.pylori has been inversely correlated with inflammatory bowel disease in some studies (4), and is a marker for overcrowding, large families, and high birth order, as well as being a gastric carcinogen. When mortality from gastric cancer (obtained from the same sources as Leon’s study) was correlated with Crohn’s disease incidence data (the largest published study with age-standardised data available for that country from the last 15 years), a strong inverse relationship was found. This is apparent for both sexes and across 26 different countries. (For males r2=0.35, p=0.003, for females r2=0.39, p=0.002). This association was not seen when comparing mortality from cervix, prostate or lung cancer with Crohn’s disease, and was also not present for ulcerative colitis. One possible source of confounding is that falling mortality from gastric cancer will indicate improvements in socio-economic and healthcare factors related to survival rather than simply gastric cancer incidence. However, gastric cancer incidence, where comparable figures were available, did correlate well with mortality in this study (r2=.76, p<0.0001). Notwithstanding this and other limitations of ecological studies, (5), we believe this adds support for the role of contrasting early environmental influences in the aetiology of both Crohn’s disease and gastric carcinoma. Patterns of early infection and colonisation (although by different agents) are likely to be important in the aetiology of both diseases. Danielle L Morris, Wellcome Training Fellow in epidemiology Wisemail@Compuserve.com Scott M Montgomery, Senior Lecturer in Epidemiology Inflammatory Bowel Disease Study Group Royal Free and University College Hospital Medical School, London NW3 2PF, UK. 1. Leon DA, Davey Smith G. Infant mortality, stomach cancer, stroke, and coronary heart disease: ecological analysis. BMJ 2000;320:1705-1706. 2. Montgomery SM, Pounder RE, Wakefield AJ. Infant mortality and the incidence of Crohn's disease. Lancet 1997;349:472-473. 3. Montgomery SM, Twamley SI, Morris DL, Pounder RE, Wakefield AJ. Birth order influences IBD risk and phenotype. Gastroenterology 1999 A2063. 4. Halme L, Rautelin H, Leidenius M, et al. Inverse correlation between Helicobacter pylori infection and inflammatory bowel disease. J Clin Pathol 1996;49:65-67. 5. Forman D, Goodman KJ. The epidemiology of stomach cancer: correlating the past with the present. BMJ 2000;320:1682-1683.