Handling Uncertainty in Cost-Effectiveness Models

Abstract

The use of modelling in economic evaluation is widespread, and it most often involves synthesising data from a number of sources. However, even when economic evaluations are conducted alongside clinical trials, some form of modelling is usually essential. The aim of this article is to review the handling of uncertainty in the cost-effectiveness results that are generated by the use of decision-analytic-type modelling. The modelling process is split into a number of stages: (i) a set of methods to be employed in a study are defined, which should include a 'reference case' of agreed methods to enhance the comparability of results; (ii) the clinical and demographic characteristics of the patients the model relates to should be specified as carefully as in any experimental study; and (iii) the data requirements of the model should be estimated using the principles of Bayesian statistics, such that prior distributions are specified for unknown model parameters. Monte Carlo simulation can then be employed to sample from these prior distributions to obtain a distribution of the cost effectiveness of the intervention. Such probabilistic analyses are related to parameter uncertainty. In addition, modelling uncertainty is likely to add a further layer of uncertainty to the results of particular analyses.

Is it cost‐effective to provide internet‐based interventions to complement the current provision of smoking cessation services in the Netherlands? An analysis based on the EQUIPTMOD

“…This may lead to productivity gains which decrease the societal burden. Fourthly, a probabilistic sensitivity analysis (PSA), that addresses the uncertainty around the model results by defining probability distributions for the input estimates instead of point estimates 60, could not be conducted, as the EQUIPTMOD was developed primarily to underpin an ROI tool for decision‐making purposes 22. This objective inevitably required the model to have the restricted PSA functionality (allowing a comparison between current and no investment only), thereby leading us to conduct a deterministic sensitivity analysis instead for this purpose.…”

Development and application of an economic model (EQUIPTMOD) to assess the impact of smoking cessation

“…Within EQUIPTMOD, the effect of parameter uncertainty on the calculated outcomes can be assessed through probabilistic sensitivity analyses involving Monte Carlo simulation (MCS) 59. For the MCS, probability distributions related to natural history parameters, relative risks and odds ratios, costs and utilities are incorporated into the model.…”

“…Analysis adopts standard methods for defining uncertainty around parameters 59. Transition probabilities are characterized by beta distributions and relative risks and odds ratios are characterized by log‐normal distributions.…”

“…a set of 1000 outcome estimates are obtained. The results are displayed by a scatterplot of costs versus QALYs and by cost‐effectiveness acceptability curves (CEACs) that present graphically the probability of being cost‐effective as a function of decision‐makers’ willingness to pay for a QALY 59.…”

The economic case for precision medicine

“…Parameter uncertainty refers to the uncertainty in the true value of each input of an economic evaluation; for example, the accuracy and clinical effectiveness of the strategies to test for activating mutations of EGFR may be uncertain (Case Study 1). A probabilistic analysis is, therefore, necessary to handle parameter uncertainty by characterizing input parameters as distributions and simulating expected outcomes over these distributions [87,88]. …”