Fish Intakes Associated with Less Colorectal Cancer
Geelen et al. American Journal of Epidemiology. 166(10):1116-1125 (2007)
The authors conducted an extensive review (meta-analysis) of 14 studies combined where the frequency of fish consumption in population studies was correlated with the incidence of colorectal cancer. The compilation of these 14 cohort studies indicated that the pooled relative risks for colorectal cancer was 12% lower (relative risk of 0.88) for the highest as compared to the lowest fish consumption category. The pooled relative risks were 22% lower for women and 6% lower for men. In those studies where the difference between the highest intake categories was 7 or more times per month, the pooled relative risk was 22% lower overall. The authors calculated that a 4% decrease in colorectal cancer risk was found per extra time per week where fish was consumed. Thus, fish consumption at 2 times per week will equate to an estimated 8% reduction in the risk of colorectal cancer. Statistical analysis of omega-3 fatty acid consumption (as DHA/EPA) was restricted since such measures of omega-3 intake were only performed in 3 studies. Although there was an overall trend for a lower colorectal cancer incidence with higher intakes of omega-3 fatty acid (using a pooled relative risk of 0.91 suggesting a potential 9% risk reduction), this trend was not statistically significant.
Dr. Holub's Comments:
The present study is of interest in view of the very low intakes of fish in various countries including the United States. For example, an extensive survey of the frequency of fish consumption amounts amongst women in the United States indicated the average 30-day intake of all fish to average at approximately 0.13 g/kg body weight per day. For a 75kg adult, this corresponds to a daily average intake of fish of 9.75g per day which translates into 1 fish serving every 11-12 days (assuming a regular serving of fish to represent a 4oz serving size (114g). Current average intakes are well below the intakes of fish that may be associated with the reduced risk of colorectal cancer based on the present study from The Netherlands.