This article is more than 5 years old. Information may no longer be current.
Any coffee intake may decrease risk for colorectal cancer
Coffee consumption may reduce risk for colorectal cancer in a dose–response manner, where even intake of one serving of coffee daily reduces risk, according to an analysis of data from the population-based, case–control Molecular Epidemiology of Colorectal Cancer study conducted in northern Israel.
This association persisted for decaffeinated and boiled coffee, results showed.
“We found that drinking coffee is associated with lower risk for colorectal cancer, and the more coffee consumed, the lower the risk,” Stephen Gruber, MD, PhD, MPH, director of University of Southern California Norris Comprehensive Cancer Center of Keck Medicine, said in a press release. “We were somewhat surprised to see that caffeine did not seem to matter. This indicates that caffeine alone is not responsible for coffee’s protective properties.”
Stephen Gruber
Because coffee contains bioactive components — including polyphenols, melanoidins, diterpenes and caffeine — related to colon physiology that may limit growth of polyps, coffee has been suggested as a protective agent for colorectal cancer. However, the evidence of this association has been limited.
Gruber and colleagues sought to further examine this association by coffee type, cancer site and ethnic subgroup. They evaluated data from 5,145 patients (mean age, 68.7 years) with colorectal cancer and 4,097 controls (mean age, 70.7 years) enrolled in the Molecular Epidemiology of Colorectal Cancer study. The major ethnic subgroups in the cohort included Ashkenazi Jewish (61.3%), Sephardi Jewish (21.4%) and Arab (13.5%).
Study participants completed a validated food frequency questionnaire to provide information on consumption of decaffeinated coffee, boiled black coffee, espresso, instant coffee or filter coffee in the year prior to their diagnosis (patients) or interview (controls).
Mean total coffee intake among controls was two servings daily.
Results of an unconditional logistic regression analysis adjusted for study matching factors and known risk factors showed any amount of coffee consumption reduced the odds for developing colorectal cancer 26% compared with not drinking coffee (OR = 0.74; 95% CI, 0.64-0.86).
When stratifying by coffee type, the association persisted for decaffeinated coffee (OR = 0.82; 95% CI, 0.68-0.99) and boiled coffee (OR = 0.82; 95% CI, 0.71-0.94).
Reduction in risk for colorectal cancer with coffee consumption only reached statistical significance in the Ashkenazi Jew and Sephardi Jew populations. However, the association persisted regardless of colon or rectum tumor location.
Compared with drinking less than one serving a day, data showed a reduction in colorectal cancer risk with daily consumption of one to less than two servings (OR = 0.78; 95% CI, 0.68-0.9), two to 2.5 servings (OR = 0.59; 95% CI, 0.51-0.68) and more than 2.5 servings (OR = 0.46; 95% CI, 0.39-0.54).
The researchers noted they could not assess duration of coffee exposure, and there were no standardized measurements for serving size. They also acknowledged that case–control studies may be limited by residual confounding.
“While the evidence certainly suggests this to be the case, we need additional research before advocating for coffee consumption as a preventive measure,” Gruber said. “That being said, there are few health risks to coffee consumption. I would encourage coffee lovers to revel in the strong possibility that their daily mug may lower their risk for colorectal cancer.” – by Alexandra Todak
Disclosure: The researchers report no relevant disclosures.
Perspective
Back to Top
Alfred I. Neugut
The case–control study is the workhorse of cancer epidemiology, providing most of the information we know regarding various putative etiologic risk factors and cancers. The Molecular Epidemiology of Colorectal Cancer study is a beautiful example of this type of study and has been very productive in describing genetic and other etiologic factors associated with colorectal cancer.
In this instance, however, we run into one of the shortcomings of the case–control methodology. Case–control studies can tell us very well about the association between a putative exposure and an outcome, but the reason for the association — ie, whether it is causal — is usually difficult to assess. This is particularly true when one is dealing with a lifestyle factor, such as coffee consumption. It is what distinguishes an observational study from a randomized trial — the fact that self-selection or volunteer bias took place.
Any lifestyle factor — such as smoking, dietary factors, coffee consumption, physical activity, occupational exposures, socioeconomic status and the like — are part of the overall makeup and persona of an individual, and they are all inextricably intertwined and interlinked. Whether you smoke, whether you are an oncologist or a waitress, whether you spend your time watching “Law & Order” reruns (as I do) or jogging 5 miles a day, or whether you eat arugula or hot corned beef sandwiches, are all choices that stem from the various factors that determine your overall sociocultural and socioeconomic features and that interrelate to innumerable other such features and choices in your life. To isolate coffee consumption in a vacuum and look at the nature of how you drink it, how much you drink, whether it is espresso or decaffeinated or boiled, or whether you drink tea or have Coca-Cola for breakfast are all life choices that also stem from your persona and relate to whether you eat arugula, whether you smoke and the other lifestyle choices you make. This is referred to as confounding, and although some efforts can be made to account for some major confounders — like tobacco — ultimately, one cannot fully account for all the confounders of life.
So, how does one get around confounding and learn anything from a case–control study? The answer is that it is damn hard. A set of rules developed by a British epidemiologist, Austin Bradford-Hill, are usually applied. With the accumulation of numerous studies, one can usually separate some of the fog when true important causal relationships exist.
Coffee, as it were, has a checkered past, with a famous paper in 1981 that showed an association between coffee and pancreatic cancer that made a huge splash but was later unable to be replicated. There have been multiple papers regarding coffee and colorectal neoplasia that have as a group, for the most part, been inconclusive — including my own paper on coffee and colorectal adenomas more than 20 years ago. Further, in two meta-analyses cited by the investigators, case–control studies were associated with a protective effect by coffee but cohort studies were not, again raising a red flag with regard to causality.
So, as I take my daily aspirin, reflect on my last screening colonoscopy a few years ago and avoid excessive red meat, I guess I will just enjoy my second cup of coffee of the day for the heck of it and reflect for now on what it is doing to my genitourinary tract rather than my gastrointestinal tract.
References:
Blackburn H and Labarthe D. Am J Epidemiol. 2012;doi:10.1093/aje/kws374.
Je Y, et al. Int J Cancer. 2009;doi:10.1002/ijc.24124.
Lee WC, et al. Ann Epidemiol. 1993;3:239-244.
Li G, et al. Public Health Nutr. 2013;doi:10.1017/S1368980012002601.
MacMahon B, et al. N Engl J Med. 1981;doi:10.1056/NEJM198103123041102.