The initial findings of the colonoscopic bowel cancer screening initiative in Germany are important and could give direction to the use of screening programs.
However, it seems sensible to present the results and validity of this screening program by using the terms usually applied in prevention—positive predictive value and numbers needed to screen.
In 2.82 million performed (or rather, projected) screening colonoscopies in members of statutory health insurance schemes aged 55–84, 98 734 precursor lesions and 47 168 preclinical (asymptomatic) cases of bowel cancer were treated successfully (projected).
The positive predictive value is the measure for a true positive test result (true positives/negatives). If colonoscopy screening has 90% sensitivity and specificity, the positive predictive value is 3%.
The number needed to screen is the number of screening rounds needed to prevent one death or adverse event. On the basis of the available data, the number needed to screen is 19.
If the results are even only approximately correctly interpreted and if the number needed to screen is compared with mammography screening, for example, the authors' demand for improved uptake of the screening program should be urgently supported.
References
- 1.Brenner H, Altenhofen L, Hoffmeister M. Eight years of colonoscopic bowel cancer screening in Germany: Initial findings and projections. Dtsch Arztebl Int. 2010;107(43):753–760. doi: 10.3238/arztebl.2010.0753. [DOI] [PMC free article] [PubMed] [Google Scholar]