Critique 064: Association of lifestyle and environmental factors with the risk of cancer – 13 December 2011
Parkin DM, Boyd L, Walker LC. The fraction of cancer attributable to lifestyle and environmental factors in the UK in 2010. Summary and conclusions. British Journal of Cancer 2011;105:S77 – S81.
Authors’ Abstract
This analysis estimates the fraction of cancers occurring in the UK in 2010 that can be attributed to sub-optimal, past exposures of 14 lifestyle and environmental risk factors. For each of 18 cancer types, we present the percentage of cases attributable to one or all of the risk factors considered [tobacco, alcohol, four elements of diet (consumption of meat, fruit and vegetables, fibre, and salt), overweight, lack of physical exercise, occupation, infections, radiation (ionising and solar), use of hormones, and reproductive history (breast feeding)].
Exposure to less than optimum levels of the 14 factors was responsible for 42.7% of cancers in the UK in 2010 (45.3% in men, 40.1% in women) – a total of about 134 000 cases. Tobacco smoking is by far the most important risk factor for cancer in the UK, responsible for 60 000 cases (19.4% of all new cancer cases) in 2010. The relative importance of other exposures differs by sex. In men, deficient intake of fruits and vegetables (6.1%), occupational exposures (4.9%) and alcohol consumption (4.6%) are next in importance, while in women, it is overweight and obesity (because of the effect on breast cancer) – responsible for 6.9% of cancers, followed by infectious agents (3.7%).
Population-attributable fractions provide a valuable quantitative appraisal of the impact of different factors in cancer causation, and are thus helpful in prioritising cancer control strategies. However, quantifying the likely impact of preventive interventions requires rather complex scenario modelling, including specification of realistically achievable population distributions of risk factors, and the timescale of change, as well as the latent periods between exposure and outcome, and the rate of change following modification in exposure level.
Forum Comments
It has long been recognized that certain lifestyle habits relate to the risk of certain cancers (e.g., smoking and lung cancer). In this paper from the UK, the authors estimate the proportion of cancer in the population associated with a variety of lifestyle and environmental factors. They find that smoking has, by far, the largest effect on the risk of cancer, with 19.4% of cancer cases in the UK attributable to tobacco use. A poor diet (less intake of fruits and vegetables and fiber and greater intake of meat and salt), obesity, and alcohol were the next most important factors that relate to cancer, with alcohol being calculated to relate to 4.0% of cancer cases in the UK.
Specific Comments on this paper: Forum reviewers considered this report to be based on a well-done analysis to estimate the effects of various lifestyle and environmental exposures on the risk of cancer. It used methodology that is preferable to that used in some similar recent analyses, such as a 2009 paper from the World Cancer Research Fund/American Institute for Cancer Research.1 The authors appropriately used exposures several years earlier than the occurrence of cancer, allowing a lag-time for the development of cancer and lowering the risk that early signs of cancer may have resulted in changes in lifestyle habits. They also used dose-response effects for exposures based, when possible, on country-specific data (rather than using the same value for all countries). Relating risk associated with alcohol consumption, it is unfortunate that they were unable to calculate estimates separately by type of alcoholic beverage consumed.
As the authors pointed out, it is not possible to adjust adequately for possible interactions of various exposures in their relation to cancer risk. As stated by one Forum reviewer: “I agree with some of the authors’ caution on the interpretation of the population-attributable risk (PAR); it is a theoretical number that can seldom be achieved. In contrast, a generalized impact factor is closer to achievable impact as it assumes a specified proportion of prevalence of the exposure that can be eliminated in a specific time frame. It should also be pointed out that the individual PARs presented in the paper are very crude as they do not account for joint effects of the individual risk factors listed. In reality, people may have a cluster of several risk factors acting together or one after another to influence cancer risk.”
A Forum reviewer was also unsure about the estimated effects in the paper on cancer risk associated with the intake of fruits and vegetables. “A deficit in intake of fruit and vegetables [less than 5 servings (400 g) per day] was reported in this paper to be an important risk factor in men, responsible for 6.1% of cancer cases. However, in the very large EPIC study2 only a very small inverse association between intake of total fruits and vegetables and cancer risk was observed; given the small magnitude of the observed associations, caution should be applied in the interpretations by the authors of the results of the present paper. Moreover, in the EPIC study, the reduced risk of cancer associated with high vegetable intake was restricted to women and was small (HR = 0.98, 95% CI = 0.97 to 0.99).2”
Another reviewer pointed out how these analyses provide somewhat lower estimates of the risk of cancer associated with alcohol consumption than often reported. He was pleased that the authors appeared to use a little more restraint (perhaps more common sense) in providing reasonable implications of their research. The authors also commented on current trends in these risk factors: the third most important factor, obesity, is increasing in prevalence in the UK while there is evidence that excessive alcohol consumption is decreasing. As they stated in their conclusions: “The proportion of men and women drinking more than the recommended maximum has been falling. It is quite possible, therefore, that the burden of alcohol-related cancers is around its maximum at present, and will fall in the future.”
Choosing no alcohol consumption as the “Optimum Exposure Level: Most Forum reviewers agreed that using no alcohol intake as the “theoretically optimum exposure level” is not appropriate. There is undoubtedly a threshold for the adverse effect of alcohol on many cancers. The strongest effects of alcohol shown in the paper were for upper aero-digestive tract cancers, which are primarily (if not exclusively) diseases associated with heavy drinking. Obviously, in terms of cardiovascular and many other diseases that show a J-shaped relation (as well as and total mortality), using no intake as “optimum” would be a serious error.
As one Forum reviewer stated: “Reasonable doubt exists as the evidence base for the theoretical optimum exposure level. For example, with no alcohol as the optimum exposure level you would certainly calculate a large number of alcohol-related cancers; however, the evidence that nil alcohol is the optimum exposure level is weak. Stratification by alcohol intake in the EPIC study2 suggested a stronger reduction in risk in heavy drinkers and was confined to cancers caused by smoking and alcohol.”
Another Forum reviewer was not as concerned about using no alcohol as an optimum intake for cancer: “With the exception of non-Hodgkin lymphoma, renal, thyroid, and a few other cancers, the use of nil ethanol as the optimum level for cancer endpoints is not too terrible.” However, taking a dose-response curve into account would provide better estimates of effect.
A reviewer reminded the Forum that underlying genetic mechanisms surely modify the effects of any environmental exposure on the risk of cancer. Thus, the estimates provided in this paper may help quantify the overall risks associated with certain lifestyle and environmental factors, but cannot be easily applied to the risk for single individuals.
References from Forum review:
1. World Cancer Research Fund (WCRF)/American Institute for Cancer Research (AICR) Policy and Action for Cancer Prevention. Food, Nutrition and Physical Activity: A Global Perspective. AICR:Washington, DC, 2009.
2. Boffetta P, Couto, E, Wachmann J, Ferrari P, Trichopoulos D, et al. Fruit and vegetable intake and overall cancer risk in the European Prospective Investigation Into Cancer and Nutrition (EPIC). J Natl Cancer Inst 2010;102:529–537.
Forum Summary
It has been well established that certain lifestyle habits relate to the risk of certain cancers (e.g., smoking and lung cancer). In a well-done analysis, the authors estimate the proportion of cancer in the population associated with a variety of lifestyle and environmental factors. They find that smoking has, by far, the largest effect on the risk of cancer, with 19.4% of cancer cases in the UK attributable to tobacco use. A poor diet (less intake of fruits and vegetables and fiber and greater intake of meat and salt), obesity, and alcohol are the next most important factors that relate to cancer, with alcohol being calculated to relate to 4.0% of cancer cases in the UK.
Forum reviewers considered this to be a well-done paper that used epidemiologic methods that are preferable to those used in some previous such analyses. Generally, they disagreed with the authors that no alcohol consumption was the theoretical “optimum exposure level,” as the risk of certain cancers seems to increase primarily from heavy drinking. Further, they found reason to believe that the purported effects related to diet may have been over-estimated.
Nevertheless, this paper provides considerable new information on lifestyle and environmental factors that may relate to the risk of cancer. It puts into perspective the importance of targeting certain behaviors for the potential reduction in the risk of cancer.
* * *
Comments on this critique by the International Scientific Forum on Alcohol Research were provided by the following members:
Erik Skovenborg, MD, Scandinavian Medical Alcohol Board, Practitioner, Aarhus, Denmark
Creina Stockley, clinical pharmacology, Health and Regulatory Information Manager, Australian Wine Research Institute, Glen Osmond, South Australia, Australia
Harvey Finkel, MD, Hematology/Oncology, Boston University Medical Center, Boston, MA, USA
Arne Svilaas, MD, PhD, general practice and lipidology, Oslo University Hospital, Oslo, Norway
David Van Velden, MD, Dept. of Pathology, Stellenbosch University, Stellenbosch, South Africa
Luc Djoussé, MD, DSc, Dept. of Medicine, Division of Aging, Brigham & Women’s Hospital and Harvard Medical School, Boston, MA, USA
Gordon Troup, MSc, DSc, School of Physics, Monash University, Victoria, Australia;
R. Curtis Ellison, MD, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA