The dilemma of crossover versus long term results in randomised controlled trialsBMJ 2014; 348 doi: https://doi.org/10.1136/bmj.g2679 (Published 09 April 2014) Cite this as: BMJ 2014;348:g2679
- I Peng Thomas Soh, medical oncologist1
Anderson and colleagues’ study on the impact of a bodyweight and physical activity intervention targeted patients who have had a positive faecal occult blood test and colonoscopy—a highly motivated population.1
The intervention group lost an average of 2.7 kg versus 0.78 kg in the control group. Weight loss reduces the risk of colorectal adenomas and cancer,2 but assessment of the size of the risk reduction in this study would require an extremely long follow-up (years or even decades), given the long transition time from polyp to cancer.3
Ethically, a randomised controlled trial that shows marked benefit should allow for crossover in the interim or after the trial is concluded. This would allow those in the control group to reap the benefits of a formal weight loss programme, given that they are probably as motivated as the intervention group. However, this would affect the long term results. I believe the authors are aware of this dilemma, and it would be interesting to know if they have any plans to unblind the control group.
Cite this as: BMJ 2014;348:g2679
Competing interests: None declared.