Neglecting familial colorectal cancer leads to bias in the study results- Dietary patterns and colorectal cancer: results from a Canadian population-based study by Chen et al., 2015.
Sivapalan Selvachsivam, Memorial University of Newfoundland
13 April 2015
Chen et al.'s study on "Dietary patterns and colorectal cancer: results from a Canadian population-based study" in Newfoundland leads to bias in the study because of the founder effects of the study population. Newfoundland, which has a higher rate of colorectal cancer patients than any other province in Canada, is the ideal place to study colorectal cancer and its association with dietary patterns. However, the effects of dietary patterns on colorectal cancer are affected by familial colorectal cancer and dietary habits in Newfoundland. Since Newfoundland has a higher rate of familial colorectal cancer or colorectal cancer associated with genetic factors it affects Chen et al.'s study results. Green et al.'s study in 2007 revealed that the high rate of colorectal cancer in Newfoundland is due to familial or genetic factors and founder mutations in APC and MSH2 are involved in high-risk families. Colorectal cancer due to genetic or familial diseases usually occurs in those younger than 74 years of age. Since the study also included the age group between 20-74 years of age, there is a possibility to include familial colorectal cancer in the study. Therefore, the association between dietary patterns and colorectal cancer is higher in the study than in reality. Thus, this study leads to bias in the results.
One of the study limitations is that the control has more healthy diet patterns and physical activity than the real picture. This it increases the dietary pattern difference between the cases and control. Further, a meat-based diet is a common dietary pattern in Newfoundland, like any other province in Canada. A healthy diet including fresh meat, fish, and vegetables are expensive, as in any other province. Therefore, a meat-based diet is a common dietary pattern in Newfoundland and overestimates the association between colorectal cancer and the meat diet pattern.
The exclusion of genetically associated or familial colorectal cancer in the study can give a better association between colorectal cancer and dietary patterns. Otherwise, a prospective cohort study with three different dietary patterns after excluding genetic or familial causes for colorectal cancer may find an actual relationship between dietary patterns and colorectal cancer. Gao et al.'s study in 2007 in the US, is a prospective cohort study done to determine the relationship between dietary patterns and Parkinson’s disease, and is an example of a prospective study.
References
Chen, Z., Wang, P. P., Woodrow, J., Zhu, Y., Roebothan, B., Mclaughlin, J. R., & Parfrey, P. S. (2015). Dietary patterns and colorectal cancer: results from a Canadian population-based study. Nutrition Journal, 14(1), 8.
Gao, X., Chen, H., Fung, T. T., Logroscino, G., Schwarzschild, M. A., Hu, F. B., & Ascherio, A. (2007). Prospective study of dietary pattern and risk of Parkinson’s disease. The American Journal of Clinical Nutrition, 86(5), 1486-1494.
Green, R. C., Green, J. S., Buehler, S. K., Robb, J. D., Daftary, D., Gallinger, S., ... & Younghusband, H. B. (2007). Very high incidence of familial colorectal cancer in Newfoundland: a comparison with Ontario and 13 other population-based studies. Familial Cancer, 6(1), 53-62.
Neglecting familial colorectal cancer leads to bias in the study results- Dietary patterns and colorectal cancer: results from a Canadian population-based study by Chen et al., 2015.
13 April 2015
Chen et al.'s study on "Dietary patterns and colorectal cancer: results from a Canadian population-based study" in Newfoundland leads to bias in the study because of the founder effects of the study population. Newfoundland, which has a higher rate of colorectal cancer patients than any other province in Canada, is the ideal place to study colorectal cancer and its association with dietary patterns. However, the effects of dietary patterns on colorectal cancer are affected by familial colorectal cancer and dietary habits in Newfoundland. Since Newfoundland has a higher rate of familial colorectal cancer or colorectal cancer associated with genetic factors it affects Chen et al.'s study results. Green et al.'s study in 2007 revealed that the high rate of colorectal cancer in Newfoundland is due to familial or genetic factors and founder mutations in APC and MSH2 are involved in high-risk families. Colorectal cancer due to genetic or familial diseases usually occurs in those younger than 74 years of age. Since the study also included the age group between 20-74 years of age, there is a possibility to include familial colorectal cancer in the study. Therefore, the association between dietary patterns and colorectal cancer is higher in the study than in reality. Thus, this study leads to bias in the results.
One of the study limitations is that the control has more healthy diet patterns and physical activity than the real picture. This it increases the dietary pattern difference between the cases and control. Further, a meat-based diet is a common dietary pattern in Newfoundland, like any other province in Canada. A healthy diet including fresh meat, fish, and vegetables are expensive, as in any other province. Therefore, a meat-based diet is a common dietary pattern in Newfoundland and overestimates the association between colorectal cancer and the meat diet pattern.
The exclusion of genetically associated or familial colorectal cancer in the study can give a better association between colorectal cancer and dietary patterns. Otherwise, a prospective cohort study with three different dietary patterns after excluding genetic or familial causes for colorectal cancer may find an actual relationship between dietary patterns and colorectal cancer. Gao et al.'s study in 2007 in the US, is a prospective cohort study done to determine the relationship between dietary patterns and Parkinson’s disease, and is an example of a prospective study.
References
Chen, Z., Wang, P. P., Woodrow, J., Zhu, Y., Roebothan, B., Mclaughlin, J. R., & Parfrey, P. S. (2015). Dietary patterns and colorectal cancer: results from a Canadian population-based study. Nutrition Journal, 14(1), 8.
Gao, X., Chen, H., Fung, T. T., Logroscino, G., Schwarzschild, M. A., Hu, F. B., & Ascherio, A. (2007). Prospective study of dietary pattern and risk of Parkinson’s disease. The American Journal of Clinical Nutrition, 86(5), 1486-1494.
Green, R. C., Green, J. S., Buehler, S. K., Robb, J. D., Daftary, D., Gallinger, S., ... & Younghusband, H. B. (2007). Very high incidence of familial colorectal cancer in Newfoundland: a comparison with Ontario and 13 other population-based studies. Familial Cancer, 6(1), 53-62.
Competing interests
None declared