Evaluating diagnostic strategies for early detection of cancer: the CanTest framework

Published on Dec 1, 2019in BMC Cancer2.93
· DOI :10.1186/s12885-019-5746-6
Fiona M Walter33
Estimated H-index: 33
(University of Cambridge),
Matthew Thompson43
Estimated H-index: 43
(UW: University of Washington)
+ 12 AuthorsJon Emery39
Estimated H-index: 39
(University of Melbourne)
Novel diagnostic triage and testing strategies to support early detection of cancer could improve clinical outcomes. Most apparently promising diagnostic tests ultimately fail because of inadequate performance in real-world, low prevalence populations such as primary care or general community populations. They should therefore be systematically evaluated before implementation to determine whether they lead to earlier detection, are cost-effective, and improve patient safety and quality of care, while minimising over-investigation and over-diagnosis. We performed a systematic scoping review of frameworks for the evaluation of tests and diagnostic approaches. We identified 16 frameworks: none addressed the entire continuum from test development to impact on diagnosis and patient outcomes in the intended population, nor the way in which tests may be used for triage purposes as part of a wider diagnostic strategy. Informed by these findings, we developed a new framework, the ‘CanTest Framework’, which proposes five iterative research phases forming a clear translational pathway from new test development to health system implementation and evaluation. This framework is suitable for testing in low prevalence populations, where tests are often applied for triage testing and incorporated into a wider diagnostic strategy. It has relevance for a wide range of stakeholders including patients, policymakers, purchasers, healthcare providers and industry.
  • References (33)
  • Citations (0)
#1Joshua D. Cohen (Johns Hopkins University)H-Index: 7
#2Lu Li (Johns Hopkins University)H-Index: 7
Last.Fay Wong (Johns Hopkins University)H-Index: 3
view all 42 authors...
#1Peter Ström (KI: Karolinska Institutet)H-Index: 7
#2Tobias Nordström (KI: Karolinska Institutet)H-Index: 10
Last.Martin Eklund (KI: Karolinska Institutet)H-Index: 20
view all 4 authors...
#1Bjørn Hofmann (NTNU: Norwegian University of Science and Technology)H-Index: 23
#2H. Gilbert Welch (Dartmouth College)H-Index: 61
#1Juliet Usher-Smith (University of Cambridge)H-Index: 15
#2Stephen J. Sharp (University of Cambridge)H-Index: 64
Last.Simon J. Griffin (University of Cambridge)H-Index: 64
view all 3 authors...
Cited By0
View next paperA public health approach to pharmacogenomics and gene-based diagnostic tests