How Do We Know How Much Cancer Is Caused By Individual Risk Factors?

When we talk about things that people can do to reduce their chances of getting cancer, a common question is: how much cancer is caused by specific risk factors? For example, someone might want to know how much lung cancer in Alberta is caused by smoking.

To answer this common question, in 2015 the Alberta Cancer Prevention Legacy Fund supported a project led by cancer researchers (Dr. Christine Friedenreich and Dr. Darren Brenner) at Alberta Health Services. Their goal was to estimate the proportion of cancer in Alberta that was caused by modifiable, lifestyle and environmental factors. The research project accomplished this goal by using estimates of ‘Population Attributable Risk’ (described below). The results from this work are used throughout the Alberta Prevents Cancer website and can be found on pages about specific cancer sites and individual risk factors.

What is Population Attributable Risk?

Population Attributable Risk (PAR) is a measure used to estimate how many instances of cancer can be attributed to specific risk factors (which include things such as: smoking, physical activity, healthy body weight, radon, etc.). This corresponds to the theoretical proportion of cancers that could be prevented if a particular risk factor was removed from the population. For example, the PAR for cancer associated with tobacco smoking for Alberta represents the proportion of cancers in Alberta that we can consider to be caused by tobacco.

What information do we need to estimate Population Attributable Risk?

Population Attributable Risk is estimated using two main pieces of information: the relative risk linking a risk factor with cancer from scientific studies and the prevalence of the risk factor in the Alberta population. The relative risk is a measure that allows us to quantify the strength of the relationship between an individual risk factor and cancer. Risk factors that are more likely to cause cancer have larger relative risks. The prevalence of the risk factor quantifies how common the risk factor is in the Alberta population.

Does this consider exposure today or in the past?

For the most part, cancer is a slow developing disease. This means that when we talk about cancers being diagnosed today, they are likely caused by exposures to risk factors that occurred in the past. Consequently, when we are estimating the number of cancers being diagnosed today that are due to an individual risk factor using Population Attributable Risks (PAR), we need to use information about the prevalence of a given risk factor from the past. Although these statistics are based on past exposure, what they also mean is that changing your exposure today will have an impact on your future cancer risk.

How do we know how many cancer cases are caused by specific risk factors?

Population Attributable Risk measures give us an estimate of the proportion of cancers caused by each risk factor. We can figure out how many actual cases of cancer this represents by multiplying the total number of cases of a specific cancer type diagnosed in Alberta by the population attributable risk percentage. For this project, we obtained data describing the number of new cancer cases diagnosed among adults in Alberta in 2015 from the Canadian Cancer Registry. These data were taken from the Canadian Cancer Registry in 2018 and as such, may differ slightly from numbers published by the Surveillance and Reporting Department in CancerControl Alberta. To learn more about the cancer data used in our project, please see our Technical Appendix.

The PAR study was funded by the Alberta Cancer Prevention Legacy Fund (ACPLF).