The impact of inclusion criteria in health economic assessments

The debate surrounding whether the findings of efficacy studies are applicable to real-world treatment situations is ongoing. The issue of lack of applicability due to a lack of clinical heterogeneity could be addressed by employing less restrictive inclusion criteria. Given that health economic assessments based on cost-effectiveness measures are required by many governments and insurance providers, the impact of this choice may be far reaching. The objective of this article was to explore the use of a pilot study to examine the impact of inclusion criteria on cost-effectiveness results and clinical heterogeneity. A health economic assessment was conducted using QRISK®2 and simulation modelling of different population groups within the pilot study in Lower Austria. Patients were referred by their family physicians to ‘Active Prevention’ (Vorsorge Aktiv), a community-based lifestyle intervention focused on exercise and nutritional programmes. Cardiovascular risk factors were recorded before and after the intervention and translated to cardiovascular events. As expected, enforcing restrictive inclusion criteria produced stronger and more irrefutable computations — in the expected number of events, the number of deaths, the incremental cost per life-year saved and in the 95% confidence interval. These findings provide insight into the issues surrounding clinical heterogeneity and the need for restrictive inclusion criteria. This is not a full health economic assessment of the intervention. While inclusion criteria provide stronger results by limiting populations to those who would benefit the most, they must be enforced, both within and outside the clinical trial setting. Enforcement has costs, both monetary and arising from unintended negative consequences of enforcement mechanisms. All these considerations will affect the results realized by the payer organization. A pilot study can reveal whether an intervention may be cost effective ‘enough’ without restrictive inclusion criteria and can enable researchers to search for population subgroups in which the intervention remains cost effective. When the pilot study does not indicate sufficiently strong cost-effectiveness results, the broader trade-offs between clinical heterogeneity and the strength of the submission package to the reimbursement agency can be discussed by all parties. Payer concerns about the ability to generalize the results beyond the clinical trial can also be discussed at this time. Applicability then depends on the ability to enforce inclusion criteria similar to those used in the trials in the real world.