Background Diet is considered a modifiable risk factor in the aetiology of colorectal cancer (CRC) but evidence on adherence to diet related cancer prevention guidelines and associations with CRC risk is limited and conflicting. The aim of this cohort analysis is to evaluate associations between adherence to recommendations from the World Cancer Research Fund/American Institute of Cancer Research (WCRF/AICR) and incident CRC.

Methods The UK Women’s Cohort Study (UKWCS) comprises over 35,372 middle-aged women, for whom dietary, anthropometric and lifestyle information was collected between 1995 and 1998, using a 217-item food frequency questionnaire. A score depicting adherence to eight of the WCRF/AICR cancer prevention guidelines was constructed for each woman in the cohort. The cohort was followed up for CRC incidence for a median of 18.7 y. Cases were identified as patients who were cancer free at baseline and who developed CRC a minimum of 12 months after the start of dietary assessment. Participants with missing data on Body Mass Index (BMI) were excluded. Cox proportional hazards regression was used to provide hazard ratios (HRs) and 95% confidence intervals (CIs) for the estimation of cancer risk. Associations were estimated for CRC, and then for colon and rectal cancer separately, adjusting for age, smoking status, family history of cancer in a first degree relative and socio-economic status.

Results The UK Women’s Cohort Study (UKWCS) comprises over 35,372 middle-aged women, for whom dietary, anthropometric and lifestyle information was collected between 1995 and 1998, using a 217-item food frequency questionnaire. A score depicting adherence to eight of the WCRF/AICR cancer prevention guidelines was constructed for each woman in the cohort. The cohort was followed up for CRC incidence for a median of 18.7 y. Cases were identified as patients who were cancer free at baseline and who developed CRC a minimum of 12 months after the start of dietary assessment. Participants with missing data on Body Mass Index (BMI) were excluded. Cox proportional hazards regression was used to provide hazard ratios (HRs) and 95% confidence intervals (CIs) for the estimation of cancer risk. Associations were estimated for CRC, and then for colon and rectal cancer separately, adjusting for age, smoking status, family history of cancer in a first degree relative and socio-economic status.

Conclusion Although no statistically significant trends were shown between adherence to WCRF/AICR cancer prevention guidelines and risk of CRC, results were all in a protective direction. There was no evidence that following these guidelines would be harmful.