Sigmoidoscopy screening: understanding the trade-off between detection of advanced neoplasia and diagnostic efficiency.

susceptibility by reductive metabolism gene variants. Sigmoidoscopy may reduce the incidence and mortality of colo-rectal cancer (CRC) in two ways. The first is by direct detection of distal (left-sided) early-stage CRC and removal of precancer-ous adenomatous polyps. The second takes advantage of the association between distal neoplastic findings and advanced neoplasia in the proximal colon. Several studies have shown a synchronous relationship between distal neoplastic findings and advanced prox-imal neoplasia (APN), in which the risk (or prevalence) of APN increases as the extent of distal neoplasia increases (1–4). Between 2009 and 2012, four trials of screening sigmoidoscopy were published (5–8), three of which reported outcomes of one-time sigmoidoscopy compared with usual care (5–7). Table 1 shows descriptive data and main results of the three trials. The trials are similar in the age range of invited and screened subjects, in intervention and control groups, and in the primary end points (CRC incidence and mortality). In all four trials, the effect of sigmoidos-copy on overall CRC mortality was due to its effect on distal CRC. Despite their similarities, the trials differ somewhat in their main findings. The differences in cancer outcomes may be due to differences in the underlying populations' risk profiles, in persons who were willing to participate and who were randomly assigned, in their adherence to the trial, and in follow-up duration (and the shorter duration for the Norwegian Colorectal Cancer Prevention Trial (NORCCAP) trial likely accounts for the lack of effect of screening on CRC incidence). These factors are beyond the control of study investigators and providers. Another reason for the differences in outcomes—one that is controllable—is the intervention itself: how well sigmoidos-copy was performed, and perhaps most important, the sigmoidoscopic findings that led to further evaluation with colonoscopy. Among the three trials, the proportion of patients referred for colonoscopy after sigmoidoscopy ranged from 4% in the Screening for COlon REctum Trial (SCORE) trial to 18% in the NORCCAP trial (Table 1). As countries and large health-care systems consider moving from trial evidence about the efficacy of sigmoidoscopy to its implementation , it is important to understand the effect of the varying referral rates for colonoscopy on CRC-specific outcomes in terms of detection of advanced neoplasia and the required effort and resources. In addition to understanding these parameters for implementation, they may be used for modeling and comparing the effectiveness and cost-effectiveness of sigmoidoscopy to other CRC screening strategies. …