Design issues in cancer screening trials
Abstract
Randomised controlled trials avoid many of the potential biases associated with the evaluation of cancer screening. Nevertheless there are many issues concerning the design of such trials that require careful consideration and that will influence interpretation of the results. This article discusses issues related to recruitment and randomisation, which will affect the extent to which the population studied, is representative of the eventual target population of a screening programme. It addresses sample size considerations, the use of appropriate outcome measures and the timing of the intervention. Finally, issues related to ensuring appropriate analyses are discussed.
Collections
Research team
Cancer Screening Evaluation Unit (DoH)
Language
eng
License start date
2010-10
Citation
STATISTICAL METHODS IN MEDICAL RESEARCH, 2010, 19 pp. 451 - 461
Publisher
SAGE PUBLICATIONS LTD