The value of reducing uncertainty: merging ethics and economics

Premenopausal women are generally not or scarcely included in pharmaceutical trials. At the same time, policy decisions about market access and reimbursement are made for women, based on evidence that is predominantly collected in men. It is deemed unethical to withhold women pharmaceuticals because there is a lack of evidence for them specifically. Currently, there is no formal and systematic assessment of whether the benefits of research in a specific vulnerable population such as women outweigh the risks. Especially in these financially challenging times, it becomes increasingly important to aim to reduce waste in the collecting and reporting of research evidence, to protect society and these vulnerable patients. In the proposed project we will develop guidelines to systematically and transparently weigh the risks and benefits of research, using a combined normative and economic approach. Value of information (VOI) analysis is increasingly used in healthcare to estimate the expected value, expressed in monetary terms, of additional research. However, since research only has value if the additional evidence impacts the decision whether or not to prescribe the pharmaceutical, the results of a VOI analysis are contingent upon what is considered an acceptable level of risk. To explore what may be considered an acceptable level of risk in this context, a normative analysis is needed. In the proposed study we will use Richardson’s method of specifying norms as a means to resolve complex ethical dilemmas. In this method, an ethical dilemma exists when mutually incompatible guidance for action results from two or more ethical norms to which we are committed. An attempt at resolution of the dilemma is made by specifying the relevant norms. Specifying norms, here, means refining them, qualifying them, restricting their scope. It is considered a necessary step for general norms to provide practical guidance. Additionally, VOI analysis can inform what type of research provides most value for money. However, the decision which research design is optimal should not be based on efficiency alone, but should also take into account ethical considerations. Hence, both in deciding on whether research is needed, and what is the optimal research design, a combined approach using VOI analysis and normative analysis allow for more systematic and accountable decisions. The question whether specific research is needed in premenopausal women is apparent in the case of novel oral anticoagulants (NOACs). NOACs comprise a new generation of thromboprophylaxis for stroke prevention in the comprehensive management of atrial fibrillation (AF). The number of NOACs is rising and pressure from industry to prescribe NOACs is increasing. At the same time, the prevalence of AF in premenopausal women is growing. While NOACs are currently increasingly being prescribed for premenopausal women, they are not tested in this younger population. There may be serious harms however that may jeopardize their safety and effectiveness. Some studies have found a higher risk of gastrointestinal bleeding for NOACs than their alternative, vitamin K antagonists. This may also imply heavy menstrual bleeding (menorrhagia) associated with NOACs in these females. To date, this risk is not acknowledged. This raises the question whether we need to reduce the uncertainty about the safety and effectiveness of NOACs in premenopausal women. The relatively low prevalence in premenopausal women combined with the relatively low number of events further complicate research in this population, and raise the question whether registries may be preferred over randomized trials. To answer these questions, we need a formal assessment of the value of additional research and the optimal research design. The proposed interdisciplinary research has both a methodological and a clinical objective. Besides demonstrating the benefits of a combined economic and normative approach and developing guidelines for systematic and accountable decision making upon research in vulnerable populations, we will answer the question whether further research on NOACs in premenopausal women is needed, and how this should be done. The proposed project will result in two tangible products. First, based on the results of the project we will write a research proposal describing the design of the research that is necessary to improve the decision whether or not to prescribe and reimburse NOACs in premenopausal women. Second, the proposed project will provide guidelines how to inform trade-offs between risks and benefits of clinical research in vulnerable populations using a combined economic and normative approach. Eventually the results of the proposed project will lead to more systematic and transparent decision making upon clinical research in vulnerable populations.