The researchers, led by doctoral student Megan Phillips, examined dietary information on more than 22,000 participants in the Physicians' Health Study.
The participants were asked to report their consumption of different categories of fish: tuna, dark meat fish such as salmon, sardines, bluefish, general fish and shellfish, including lobster, shrimp and scallops.
Roughly 10 percent of the participants reported eating fish less than once a week, while 31 percent ate it less than twice a week and 48 percent ate fish less than 5 times a week. About 11 percent of the men surveyed reported eating fish five times or more per week.
The researchers tracked the men's diets for up to 19.4 years, and found that those who reported eating the most fish -- five times or more per week -- experienced a 40 percent lower risk of developing colorectal cancer, compared to men who ate the least fish.
The men who reported eating fish two to five times per week experienced a 20 percent lower risk, while those who ate fish fewer than twice a week still experienced a 13 percent risk reduction.
Phillips' findings were recently presented at the American Association for Cancer Research meeting in Boston. Phillips notes that her study was based on the idea that the men's pattern of fish consumption that was reported in the initial questionnaire held true for many years.
The men who consumed the most fish per week may have also led a generally healthier lifestyle than the men who ate little to no fish, Phillips said, which could have contributed to their lower risk.