The Application and Implications of Novel Deterministic Sensitivity Analysis Methods

Rick A. Vreman (), Joost W. Geenen, Saskia Knies, Aukje K. Mantel-Teeuwisse, Hubert G. M. Leufkens and Wim G. Goettsch
Additional contact information
Rick A. Vreman: Utrecht University
Joost W. Geenen: Utrecht University
Saskia Knies: The National Health Care Institute (ZIN)
Aukje K. Mantel-Teeuwisse: Utrecht University
Hubert G. M. Leufkens: Utrecht University
Wim G. Goettsch: Utrecht University

PharmacoEconomics, 2021, vol. 39, issue 1, No 1, 17 pages

Abstract: Abstract Deterministic sensitivity analyses (DSA) remain important to interpret the effect of uncertainties in individual parameters on results of cost-effectiveness analyses. Classic DSA methodologies may lead to wrong conclusions due to a lack of or misleading information regarding marginal effects, non-linearity, likelihood and correlations. In addition, tornado diagrams are misleading in some situations. Recent advances in DSA methods have the potential to provide decision makers with more reliable information regarding the effects of uncertainties in individual parameters. This practical application discusses advances to classic DSA methods and their implications. Three methods are discussed: stepwise DSA, distributional DSA and probabilistic DSA. For each method, the technical specifications, options for presenting results, and its implications for decision making are discussed. Options for visualizing DSA results in incremental cost-effectiveness ratios and in incremental net benefits are presented. The use of stepwise DSA increases interpretability of marginal effects and non-linearities in the model, which is especially relevant when arbitrary ranges are implemented. Using the probability distribution of each parameter in distributional DSA provides insight on the likelihood of model outcomes while probabilistic DSA also includes the effects of correlations between parameters. Probabilistic DSA, preferably expressed in incremental net benefit, is the most appropriate method for providing insight on the effect of uncertainty in individual parameters on the estimate of cost effectiveness. However, the opportunities provided by probabilistic DSA may not always be needed for decision making. Other DSA methods, in particular distributional DSA, can sometimes be sufficient depending on model features. Decision makers must determine to which extent they will accept and implement these new and improved DSA methodologies and adjust guidelines accordingly.

Date: 2021
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s40273-020-00979-3 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:pharme:v:39:y:2021:i:1:d:10.1007_s40273-020-00979-3

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/40273

DOI: 10.1007/s40273-020-00979-3

Access Statistics for this article

PharmacoEconomics is currently edited by Timothy Wrightson and Christopher I. Carswell

More articles in PharmacoEconomics from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().