Methods for the cost-effectiveness modelling of screening interventions in an uncertain landscape
: Application to screening for prostate cancer

  • Edna Keeney

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)

Abstract

Decisions need to be made about who and how to screen for diseases to optimise health in
the population. Cost-effectiveness analyses of screening interventions can be associated
with many areas of uncertainty due to a constantly changing landscape in screening
methods, diagnostic tests, treatments and understanding of natural history. A failure to
account for such uncertainty may result in incorrect or poorly informed decisions. Prostate
cancer is an example of a disease where recent developments in the understanding of who
and how to screen have provided challenges to the analyst trying to make
recommendations on the most cost-effective screening strategy.
Using prostate cancer screening as a case study, this dissertation explores methods to
handle uncertainty when modelling the cost-effectiveness of screening interventions in an
uncertain landscape. The dissertation shows how a systematic review of previous models
can identify areas of parameter and structural uncertainty, how to gain expert consensus
with respect to relevant screening strategies, and how to appropriately adapt and calibrate
an existing natural history model to a new setting.  
It will demonstrate how the 22 studies identified in the systematic review informed the
structure and data parameters of the natural history model and how a modified-Delphi
process identified prostate cancer screening strategies that were deemed relevant by
experts, including risk-stratified and adaptive approaches. It will also show how a decision
model was adapted and calibrated to UK data to find that, of the strategies identified in the
Delphi, a once-off screening at age 50 years was most cost-effective.  
Many methods are available for dealing with uncertainty in cost-effectiveness modelling of
screening interventions. The dissertation will conclude with a discussion on the merits and
limitations of the methods used, with recommendations given for practice. The aim is to
provide a guide to identifying and addressing the uncertainty inherent in cost-effectiveness
analyses of screening strategies. 
Date of Award5 Dec 2023
Original languageEnglish
Awarding Institution
  • University of Bristol
SupervisorHoward H Z Thom (Supervisor), Richard M Martin (Supervisor), Emma L Turner (Supervisor) & Sabina Sanghera (Supervisor)

Cite this

'