The Health Professionals Follow-up Study (HPFS) was initiated in 1986 when 51,529 male health professionals aged 40 to 75 years returned a mailed questionnaire. We excluded participants who had cancer, left extensive food items blank, or had implausible energy intake at baseline in 1986. As a result, 47,302 men were included. This study was approved by the Institutional Review Boards of the Harvard School of Public Health (Boston, MA).
Dietary information was collected from participants using validated semi-quantitative food frequency questionnaires in 1986 and every four years thereafter. Responses on frequencies of a specified serving size for each food item were converted to average serving per day. The quantities of choline and betaine from foods were calculated by multiplying average serving per day by the nutrient content of one serving of that food (5
). Because supplements contributed little to total intake, we presented dietary choline and betaine intake (from foods only) as main results. For total intake (from foods and supplements), we took into account current use of supplements and brand and type of multivitamins (asked every 2 years).
Information on weight, physical activity, aspirin use, current multivitamin use, family history of colorectal cancer in parents and siblings, and endoscopy history were updated every two to four years. Duration of multivitamin use, average number of cigarettes smoked per day in the age ranges of <15, 15-19, and 20-29 years, and height were assessed at baseline.
We obtained self-reported information on the occurrence of colorectal cancer on each follow-up questionnaire and asked participants (or next-of-kin for those who had died) for permission to access medical records to confirm the cancer diagnosis. In addition, the National Death Index (8
) was used to identify fatalities.
Participants were categorized using quintiles on the basis of the distribution of intakes. Intakes of choline, betaine, and other nutrients were energy-adjusted using the residuals from the regression of nutrient intake on total energy intake (9
). Relative risk (RR) and 95% confidence interval (95% CI) were calculated using the Cox proportional hazards model (10
), stratified by age and calendar year, using SAS 9.1 (SAS institute, Cary, NC). Person-years of follow-up were estimated from the date that the baseline questionnaire was returned to the date of colorectal cancer diagnosis, date of death, or end of follow-up (January 31, 2004), whichever came first. In the multivariate models, we adjusted for possible risk factors listed in the footnote to . As the main analytic strategy, the cumulative average choline and betaine intakes were calculated from all available dietary questionnaires (11
). We also examined baseline intake and latency between intake and colorectal cancer diagnosis, methods of which are described in footnotes to .
Relative risks (RR) and 95% confidence intervals (CIs) of colorectal cancer according to quintiles of energy-adjusted dietary choline and betaine intake in the Health Professionals Follow-up Study
For the test for trend, participants were assigned the median value of their intake category, and this variable was tested with the Wald test. A test for interaction was conducted using the likelihood ratio test by comparing the model fit including the cross-product term of a continuous exposure variable with a modifier variable with the model fit excluding the cross-protect term.