1Data on cancer incidence was obtained from the Alberta Cancer Registry for the year 2012.
2 Tobacco Smoking was defined as the proportion of current and former smokers in Alberta in 2001 measured using data from the Canadian Community Health Survey, Cycle 1.1 for lung, oral cavity and pharynx, esophagus, liver, pancreas, colorectum, ovary, kidney, myeloid leukemia and bladder cancers. Comparable data from the Canadian Community Health Survey, Cycle 3.1 (2005) was used for larynx and stomach cancers and Cycle 4.1 (2007) for cervical cancer.
3 Overweight and obesity were defined using data on body mass index (BMI), where BMIs from 25 – 29.9 kg/m2 were classified as overweight and BMIs ≥ 30 kg/m2 were classified as obese. The proportion of overweight and obese individuals was obtained using self-reported height and weight data from the Canadian Community Health Survey, Cycle 1.1 (2000/2001) for esophagus, kidney, gallbladder and endometrial cancers and the Canadian Community Health Survey, Cycle 2.1 (2003) for pancreatic, colorectal and breast cancers.
4 Physical inactivity was characterized using self-reported data on leisure time physical activity from the Canadian Community Health Survey. Respondents were classified as moderately inactive if their reported leisure time activities translated to a daily energy expenditure of 1.5 – 2.9 kcal/kg/day. Inactive individuals were those whose self-reported leisure time activities translated to an energy expenditure of <1.5kcal/kg/day. Data on the proportion of moderately inactive and inactive individuals in Alberta for analyses related to all cancer sites were obtained from the Canadian Community Health Survey, Cycle 2.1 (2003).
5 Alcohol consumption levels in Alberta were estimated using data from the Canadian Community Health Survey Cycle 1.1 (2000/2001) for analyses related to oral cavity/pharynx, esophageal, colorectal and breast cancers. Data from the Canadian Community Health Survey, Cycle 2.1 (2003) were used for analyses related to larynx and liver cancers.
6 Fruit and vegetable consumption levels were estimated using data from the Canadian Community Health Survey. Data from Cycle 2.1 (2003) were used to characterize fruit consumption for analyses related to esophageal, stomach and lung cancers. Data from Cycle 2.2 (2004) were used to characterize vegetable consumption for analysis related to stomach cancer. Data from Cycle 3.1 (2005) were used to characterize vegetable consumption for analysis related to esophageal cancer. Data from 2007/2008 were used to characterize both fruit and vegetable consumption for analyses related to oral cavity, pharynx and larynx cancers.
7 Red and processed meat consumption levels were estimated using data from Alberta’s Tomorrow Project cohort (http://myatp.ca/). Using data from the diet history questionnaire that was completed as part of baseline data collection, the number grams consumed per day of red (beef, pork, lamb, veal, venison, liver etc.) and processed (cold cuts, sausage, ham, hot dogs) meat were estimated among members of the cohort.
8 Fibre consumption levels were estimated using data from Alberta’s Tomorrow Project cohort (http://myatp.ca/). Data from the diet history questionnaire completed as part of baseline data collection were used to estimate total dietary fibre intake in grams per day. A cut point of 23 g/day of dietary fibre intake was used to define sufficient fibre intake, where any intake below 23 g/day was considered insufficient.
9 Salt consumption levels were estimated using data from Alberta’s Tomorrow Project cohort (http://myatp.ca/). Data from the diet history questionnaire completed as part of baseline data collection were used to estimate total dietary salt intake in grams per day. The guideline from Health Canada (http://www.hc-sc.gc.ca/fn-an/nutrition/sodium/index-eng.php) for daily salt intake is 5.75 g/day (2200 – 2300 mg/day sodium) and this was used as the cut point to define excess salt intake.
10 Vitamin D intake from diet and supplements was estimated using data from Alberta’s Tomorrow Project cohort (http://myatp.ca/). Data from the diet history questionnaire completed as part of baseline data collection were used to estimate total dietary vitamin D intake and intake from supplements. The Health Canada guideline of 600 IU of vitamin D per day (http://www.hc-sc.gc.ca/fn-an/nutrition/vitamin/vita-d-eng.php) was used as the cut point to define optimal vitamin D intake. Vitamin D produced by cutaneous sun exposure was not considered in our analysis.
11 Dietary calcium consumption was estimated using data from Alberta’s Tomorrow Project cohort (http://myatp.ca/). Data from the diet history questionnaire completed as part of baseline data collection was used to estimate total dietary calcium intake. The Health Canada guideline for dietary calcium consumption is 1000-1200 mg/day (http://www.hc-sc.gc.ca/fn-an/nutrition/vitamin/vita-d-eng.php), so a cut point of 1100 mg/day was used for this analysis.
12 Data on use of birth control pills and hormone replacement therapy (HRT) were obtained from Alberta’s Tomorrow Project cohort (http://myatp.ca). Questions concerning ever use of birth control pills and ever use of HRT were included as part of baseline data collection for all cohort participants. Women were classified as ‘ever’ users of birth control pills and as ‘ever’ HRT users.
13 Human Papillomavirus (HPV) infection is required for the development of cervical cancer, such that all cervical cancers are considered to be infected. Thus, the prevalence of HPV infection in cervical cancer cases in Alberta was the number of cervical cancers diagnosed in the province in 2012. For all other cancer sites (oral cavity/oropharynx, penis, vagina, vulva, anus) the prevalence of HPV infection in cancer tumours was obtained from the scientific literature. Preference was given, in order, to evidence from Alberta, Canada, North America and Europe as these populations were considered to be most comparable to Alberta.
14 Infection with Epstein-Barr virus (EBV) is common in childhood and adolescence and persists throughout life in certain cells involved in the immune system (B-lymphocytes). Therefore, cancers caused by EBV infection are considered to be those where EBV can be detected within tumour cells. The proportion of Hodgkin’s lymphoma and nasopharyngeal cancer cases where EBV in tumour cells could be detected was estimated from the scientific literature, using estimates for North America as no Alberta-specific estimates were available.
15 The prevalence of Hepatitis B and Hepatitis C infections was estimated at both the population level and in cases of liver cancer. At the population level the prevalence of Hepatitis B and C infections was estimated using data from the Canadian Health Measures Survey (CHMS) for Canada, as Alberta-specific data from this survey were not available. Given that these population-level estimates likely represent an underestimate of the true prevalence of Hepatitis B and C infection as participation in the CHMS is voluntary and individuals had to specifically consent to testing for these viruses, population attributable risk estimates were also conducted using estimates of the frequency of Hepatitis B and Hepatitis C infection in liver cancer cases from the published scientific literature for North America, as Alberta-specific estimates were not available.
16 The prevalence of Helicobacter Pylori (H. pylori) infections was estimated at both the population level and in cases of stomach cancer. As no Alberta-specific survey data was available, the prevalence of H. pylori infection in survey data from Nova Scotia and Ontario was used to estimate population prevalence of infection. Due to concerns that these surveys could be underestimates of the true prevalence, the prevalence of H. pylori in stomach cancer cases was estimated from data from the USA and Australia from the published scientific literature.
17 To estimate population attributable risks we compared incident malignant melanoma cases on body parts exposed to natural UV and those not exposed. The analysis for the results presented here assumed that 60% of cases of malignant melanoma on the trunk were on unexposed areas of skin.
18 Air quality was measured as average concentrations of fine particulate matter (<2.5 µm in diameter; PM2.5). Data on PM2.5 levels were obtained for the year 2011 from the Clean Air Strategic Alliance (CASA) Data Warehouse for 27 communities across Alberta. Population-weighted PM2.5 concentrations were used to estimate the proportion of lung cancers attributable to air pollution.
19 Radon exposure levels for Alberta were estimated based on data from the Health Canada Cross-Country Survey of Radon Concentrations in Homes (CCSRH) conducted from 2009 – 2011 (http://www.hc-sc.gc.ca/ewh-semt/alt_formats/pdf/radiation/radon/survey-sondage-eng.pdf). This survey was conducted by Health Canada over the fall/winter seasons in 2009/2010 and 2010/2011. Radon measurements were obtained using an alpha track detector (measures alpha particles from radon gas) placed on the lowest lived-in level of the house over a three month period for participating homeowners.
20 Passive tobacco smoking (second hand smoke) exposure was defined as exposure at home, in a vehicle or a public place in Alberta in 2003 using data from the Canadian Community Health Survey, Cycle 2.1.