Abstract
Colorectal cancer (CRC) accounts for nearly a million deaths worldwide each year. Endoscopy of the large intestine can reduce CRC incidence and mortality by removing polyps and by detecting CRC at an early stage. Screening programs across the world have switched the initial screening procedure from sigmoidoscopy to colonoscopy due to the assumption of a better colonoscopy screening effect in the proximal colon.
The aim of the thesis was to investigate comparative effectiveness and performance of endoscopic colorectal cancer screening and gastrointestinal endoscopy services. The analyses used data from four existing randomized trials comparing an invitation to sigmoidoscopy screening to usual care (i.e. no screening invitation), to estimate the effectiveness of endoscopic CRC screening. Also, baseline data from two trials on post-polypectomy surveillance were used to estimate the number of colonoscopies needed to achieve a colon free of polyps.
Pooled analysis of the four sigmoidoscopy screening trials showed that invitation to sigmoidoscopy screening reduced CRC incidence and mortality for at least 15 years, but the reduction was lower in women than men. Further, the additional benefits of switching to primary colonoscopy screening was limited – three quarters of the colonoscopy screening effect is already achieved by sigmoidoscopy screening. Lastly, there was substantial variations in number of colonoscopies needed to remove all colon polyps, before entering post-polypectomy surveillance. p>
The findings are important because they demonstrate long-term benefits of endoscopic screening, but also unwanted variation in gastrointestinal endoscopy services. CRC screening and surveillance are dependent on each other as parts of an individuals’ treatment path and put demand on the same endoscopy service resources. Although colonoscopy screening may provide the greatest benefits, it is too early to conclude that colonoscopy is the best screening approach at a population level.