Globally Optimal Societal Decision Maker Trials

Introduction

Within a jurisdiction adopting and trialling (AT) where a new therapy has positive, while uncertain net clinical benefit and overall net benefit is usually infeasible and unethical, as informed patients prefer certainty of treatment outside trial to chance of new therapy in a trial setting. Hence, ‘within jurisdiction’ feasible options as considered in Chap. 5 for such promising therapies or strategies will often be restricted to delaying and trialling (DT) or adopting now (AN) in optimising joint research and reimbursement decisions. However, in designing optimal trials across jurisdictions, trials don’t need to be undertaken within jurisdiction, and hence adopting and trialling in another jurisdiction becomes a feasible (and often valuable) option for promising therapies or strategies of interest with expected positive while uncertain INB. That is, a jurisdiction can adopt and translate trial evidence from other jurisdictions as part of optimising joint research and reimbursement decision and trial design across jurisdictions, and ideally globally, as this chapter considers following Eckermann and Willan (2009).

Indeed, in this chapter it becomes clear that within jurisdiction, VOI analysis is inconsistent to the extent that synthesising all translatable evidence arising external to jurisdiction in estimating the prior distribution for INB recognises all relevant trial evidence retrospectively yet only considers evidence arising within jurisdiction as having prospective value. Evidence arising from publicly available trials is nonrival and non-excludable across jurisdictions. Hence, provided evidence is translatable, evidence arising in one jurisdiction has value across jurisdictions.

Where prospective VOI from trials across jurisdictions is considered, an additional viable option to undertaking a trial within jurisdiction is for a side payment to influence trial design in another jurisdiction. This avoids replication of fixed trial costs and reduces heterogeneity of evidence, the potential for cherry picking of evidence or Frankenstein’s monster issues in evidence synthesis that frequently arise with multiple trials. Hence, provided trial evidence translates, a single optimal

© Springer International Publishing AG 2017 153

S. Eckermann, Health Economics from Theory to Practice,

DOI 10.1007/978-3-319-50613-5_6

trial across two jurisdictions improves on separate trials within each jurisdiction. Extending this principle across all jurisdictions raises the question: what is the globally optimal trial design?