Number needed to be screened in a study: a novel measure for disease screening effect / 中华流行病学杂志

ABSTRACT

<p><b>OBJECTIVE</b>To introduce the concept, methods for calculation and application of "number needed to be screened" (NNBS) in epidemiologic studies.</p><p><b>METHODS</b>The concept of "number needed to treat" (NNT) was extended for disease screening strategies. For the purpose of illustration, the values of number needed to invite for screening (NNI) and number needed to be screened (NNBS) were calculated on the basis of the results from two randomized controlled screening trials--Nottingham randomized controlled trial of faecal-occult-blood screening for colorectal cancer and Swedish mammographic screening trial for breast cancer in two counties.</p><p><b>RESULTS</b>In order to prevent one death from the colorectal cancer among local people aged from 45 to 74 during the 14 years of follow-up, the NNI and NNBS for faecal-occult-blood screening program were 1220 and 665, respectively. In addition, in order to prevent one death from breast cancer among local women aged 40-74 during 8 years of follow-up, the NNI and NNBS for mammographic screening program were 1961 and 1494, respectively.</p><p><b>CONCLUSION</b>Compared with the traditional indices, NNBS can evaluate the overall effectiveness of a screening program in an intuitively understandable manner so as to facilitate the communication among medical researchers, health workers, health policy makers and the public.</p>