In Reply Dr Holme and colleagues raise a question about how the results of NORCCAP were analyzed in our meta-analyses of flexible sigmoidoscopy trials. We purposely chose to use the unadjusted data from NORCCAP. Although this approach has limitations, for consistency across trials and because of how the data were reported in each trial, we used the raw unadjusted data from each of the flexible sigmoidoscopy trials to estimate the RR reduction in CRC mortality. When the meta-analyses were redone per their suggestion, no meaningful difference in pooled estimates for reduction in CRC mortality was found (RR, 0.73 [95% CI, 0.66-0.82] for the original estimate vs RR, 0.72 [95% CI, 0.65-0.80] for the recalculated estimate). In the full report,1 a discussion of the age-adjusted NORCCAP results was included.
Lin JS, Perdue LA, O’Connor E. Colorectal Cancer Screening—Reply. JAMA. 2016;316(16):1715. doi:10.1001/jama.2016.13861