Adjusting for treatment refusal in rationing decisions

Cost-utility analysis when not everyone wants the treatment: modeling split-choice bias

Not all clinically eligible patients will necessarily accept a new treatment. Cost-utility analysis recognizes this by multiplying the mean incremental expected utility (EU) by the participation rate to obtain the utility gain per head. However, the mean EU gain over all patients in a defined clinical category is traditionally used as a proxy for the mean EU gain over the subpopulation of acceptors. Even for clinically identical patients, this may lead to a biased assessment of total benefit because a patient motivated to accept the new treatment is likely to value its effects more favorably than a patient who declines. An analysis that ignores this tendency will be biased toward an underestimate of true benefits of a health technology (HT). The extent of this bias is described within a quality-adjusted life year-based utility model for a population of clinically indistinguishable patients who differ with respect to the values that they place on the possible health outcomes of an HT. The size of the bias is sensitive to the proportion of patients who accept the treatment, under both deterministic and probabilistic models of individual decision making. In all cases in which decision making is correlated with personal utility gain, the bias rises steeply as the proportion of acceptors declines.

Sample-size calculations for trials that inform individual treatment decisions: a 'true-choice' approach

BACKGROUND

Sample size decisions for clinical trials should be taken in such a way as to maximize informed choice by reducing scientific uncertainty about the consequences of an intervention.

PURPOSE

Recent approaches to trial design have focused on the potential decision impact of the trial when deciding whether the trial should be undertaken, and how large it ought to be. For the most part these approaches are concerned with the impact of trials either on clinical opinion or on collective reimbursement recommendations. Our purpose is to model the contribution of clinical trials to patient-level decision-making and to propose a way of assessing this contribution at the design stage.

METHODS

The model is developed within the framework of Bayesian decision theory. It is presumed that some patients make choices that they would not have made in the presence of perfect information about the likely consequences. These 'false' choices would be reversed in response to a fully informative (ie, very large) trial of the competing interventions. By contrast, choices that would not change in response to a fully informative trial are termed 'true' choices since they accurately reflect patient preferences.

RESULTS

An impact plot is proposed which maps how the expected numbers of 'true' and 'false' choices change in response to a trial of any given size. The approach is illustrated with reference to the choice of delivery mode for term breech presentation, using data obtained before the recent term breech trial. Applications in other contexts are indicated.

LIMITATIONS

No account is taken of the magnitude of expressed patient preferences for one treatment over another. The upside is that the need for detailed utility-elicitation is obviated.

CONCLUSIONS

The approach is a pragmatic aid to trial design in settings where patient preference drives the choice between alternative treatments.