Online Supporting Material (OSM)

Figure 1: Flow chart

1. Detailed dietary assessment methods

Subjects were provided a previously validated semi-quantitative Spanish Food Frequency Questionnaire (FFQ) [1], which was modified to include regional products. The FFQ included 140 food items, and assessed usual dietary intake during the previous year. The FFQ included the specific cooking methods for meat and some pictures to establish the personal preference of doneness. The FFQ was self-administered and returned by mail or filled out face to face (global response rate 88%). The food composition table has been a compiled table from the Centre for Higher Studies in Nutrition and Dietetics (CESNID) and other specific sources [2].

Cross-check questions on food groups intakes were used to adjust the frequency of foods eaten and reduce misreporting of food groups with large numbers of items [3,4]. Further assessment of misreporting included the identification of potential under- and over-reporting subjects using the predicted total energy expenditure (pTEE) method [5]. pTEE was estimated using Dietary Reference Intakes prediction equations, and implausible reporters were identified on the basis of the ratio of reported intakes to estimated requirements (rEI:pTEE).

Frequency data were used to derive amount (g/day, g/1000kcal) of each of the individual meat types and summary variables. Meat-related variables were grouped as follows: (i) red meat included all types of beef, pork, lamb and duck; (ii) white meat included chicken, turkey and rabbit; (iii) organ meat included all types of liver and offal; (iv) cured meat included cooked ham, Spanish cured ham, bacon and other Spanish cured sausages (chorizo, fuet, salchichón and others); (v) processed meat included all types of hamburgers and to-be-cooked sausages, hot dogs, meat balls and other meat products. Doneness preference was grouped into 3 categories: rare, medium, and well-done. Cooking methods included: griddle-grilling, barbequing, pan-frying, bread-coated frying, stewing, oven-baking, and others. These categories were non-exclusive, so for each meat subtype (i.e. white meat) a question on the frequency of each cooking method was asked. Frequencies for each method and meat subtype were then given a score as follows: never or almost never=0, sometimes=1, half of the times=2, most of the times=3, and always=4. The proportion of meat cooked in a given cooking method for each subject was then computed by dividing the score for the given method by the sum of the scores for each method. Finally, these proportions were used to compute the grams of each given meat subtype cooked with each cooking method. For the manuscript tables cooking methods were further grouped into 4 groups: Griddle-grilled / barbecued, Pan-fried / bread-coated fried, Stewed, and Oven-baked / other.

References 1. García-Closas R, García-Closas M, Kogevinas M, Malats N, Silverman D, Serra C, Tardón A, Carrato A, Castaño-Vinyals G, Dosemeci M, Moore L, Rothman N, Sinha R. Food, nutrient and heterocyclic amine intake and the risk of bladder cancer. Eur J Cancer 2007;43:1731–40.

2. Tablas de composición de alimentos CESNID. Taules de composició dels aliments CESNID. Barcelona: Ediciones de la Universitat de Barcelona, McGraw Hill-Interamericana de España S.A. 2008.

3. Calvert C, Cade J, Barrett J H, Woodhouse A, and UKWCS Steering Group. Using crosscheck questions to address the problem of mis-reporting of specific food groups on Food Frequency Questionnaires. European Journal of Clinical Nutrition (1997) 51, 708-712.

4. Nothlings U., Hoffmann K., Boeing H. Do cross-check questions improve food frequency questionnaire data? IARC Scientific Publications, IARC press, Lyon 2002.

5. Mendez MA, Popkin BM, Buckland G, Schroder H, Amiano P, Barricarte A, Huerta JM, Quirós JR, Sánchez MJ, González CA. Alternative Methods of Accounting for Underreporting and Overreporting When Measuring Dietary Intake-Obesity Relations. Am J Epidemiol 2011;173:448–458. Online supplement table 1. OR (95% CI) for the associations between tertiles of relative meat intake (gr/1000kcal) and colorectal cancer in the MCC study, according to tumour location. Colon tumours

All Men (813 cases / 2231 controls) (519 cases / 1190 controls)

Model 1 Model 2 Model 1 Model 2 White meat T1 ref ref ref ref 0.96 (0.73- T2 1.02 (0.83-1.25) 1.04 (0.84-1.29) 0.95 (0.74-1.24) 1.26) 1.09 (0.82- T3 1.06 (0.85-1.31) 1.12 (0.89-1.41) 1.03 (0.78-1.36) 1.47)

Ptrend 0.596 0.323 0.836 Red meat T1 ref ref ref ref 1.09 (0.82- T2 1.14 (0.93-1.40) 1.10 (0.89-1.36) 1.12 (0.86-1.46) 1.43) 1.28 (0.96- T3 1.28 (1.04-1.59) 1.16 (0.93-1.45) 1.40 (1.06-1.83) 1.71)

Ptrend 0.021 0.195 0.016 Processed meat a T1 ref ref ref ref 0.87 (0.65- T2 1.06 (0.86-1.31) 0.98 (0.79-1.21) 0.98 (0.75-1.28) 1.15) 1.12 (0.83- T3 1.35 (1.10-1.67) 1.16 (0.92-1.45) 1.34 (1.02-1.76) 1.49)

Ptrend 0.005 0.205 0.034 Total meat T1 ref ref ref ref 0.94 (0.72- T2 1.05 (0.85-1.30) 0.99 (0.79-1.23) 1.03 (0.79-1.34) 1.25) 1.38 (1.03- T3 1.45 (1.17-1.79) 1.30 (1.04-1.63) 1.53 (1.16-2.01) 1.85)

Ptrend 0.001 0.024 0.002 Rectum tumours

All Men (372 cases / 2231 controls) (253 cases / 1190 controls)

Model 1 Model 2 Model 1 Model 2 White meat T1 ref ref ref ref 1.22 (0.84- T2 1.26 (0.95-1.67) 1.36 (1.01-1.84) 1.11 (0.79-1.57) 1.76) 1.35 (0.91- T3 1.18 (0.87-1.59) 1.34 (0.97-1.84) 1.14 (0.79-1.63) 1.99)

Ptrend 0.318 0.084 0.499 Red meat T1 ref ref ref ref 1.02 (0.70- T2 1.18 (0.88-1.59) 1.12 (0.83-1.52) 1.04 (0.73-1.49) 1.49) 1.65 (1.13- T3 1.63 (1.22-2.18) 1.53 (1.13-2.08) 1.63 (1.15-2.32) 2.40)

Ptrend 0.001 0.005 0.005 Processed meat a T1 ref ref ref ref 1.34 (0.91- T2 1.36 (1.02-1.82) 1.21 (0.89-1.64) 1.48 (1.03-2.14) 1.97) 1.65 (1.10- T3 1.59 (1.18-2.13) 1.38 (1.00-1.89) 1.84 (1.27-2.66) 2.47)

Ptrend 0.002 0.048 0.001 Total meat T1 ref ref ref ref 1.57 (1.08- T2 1.63 (1.22-2.17) 1.59 (1.18-2.16) 1.53 (1.07-2.19) 2.30) 1.83 (1.22- T3 1.75 (1.29-2.37) 1.67 (1.20-2.31) 1.80 (1.24-2.62) 2.75)

Ptrend <0.001 0.003 0.002 Model 1 adjusted by age, area, gender (only in "All" model) and educational level. Model 2 adjusted by age, area, gender (only in "All" model), BMI one year before recruitment, educational level, total energy intake, plausibility of reported intake, smoking status one year before recruitment, physical activity during the previous 10 years (excluding 2 years before recruitment), intakes of: fruits, vegetables, nuts, dairy products and fibre, alcohol intake at age 30-40, family history of colorectal cancer, use of anti-inflammatory drugs, use of hormone replacement therapy (only in women) and season of FFQ. a Including processed meat, cured meat and organ meat.