Diagnosis and Treatment of Refractory Chronic Cough: An American Broncho-Esophagological Association Expert Consensus Statement.

Governments today are grappling with increasing healthcare costs. The average healthcare spending as a percentage of overall government expenditure across Organisation for Economic Co‐operation and Development (OECD) countries stands at 15% and is projected to increase by 5% points by 2040 based on projected revenues []. But with limited policy options to address these costs, combined with growing demand for increasingly costly interventions, healthcare remains scarce. This scarcity has given rise to a core concern in population‐level bioethics, that is, how to distribute scarce health‐related resources within society. A range of thinkers have proposed and defended the use of a normative criterion for allocation such as cost‐effectiveness [], aggregate health maximization [], inequality aversion [] and priority for the worst‐off [, ], or even a hybrid combination of some of these criteria []. Others have focused on the procedural aspects of determining healthcare priorities and highlight the importance of public deliberation and engagement [, , , ]. In addition, a number of ethicists have argued for the inclusion of responsibility as a criterion in allocating scarce healthcare [, , ]. Importantly, these debates center around what types of health risk should be spread, that is, paid for by the public, and what types of risk should be personalized, that is, paid for by the individual. Despite extensive discussions on resource and burden distributions, none have addressed the “cost of information” required to make these important allocation decisions. To determine which health risks to spread and which to personalize, it is necessary to estimate the risk of specific diseases and the predicted cost of treatment for them. Problematically, the cost of information is likely to be uneven across disease states. A full genetic test to determine the risk of an individual developing a genetically related disease will be more costly than determining the risk of lung cancer from smoking. Beyond a certain point, paying the cost of information to determine the distribution of burdens becomes sub‐optimal for all members of society. Arbitrarily limiting the pursuit of information when costs are high results in inconsistent risk‐personalization across a population. This uneven application of risk‐personalization is a type of discrimination that I term “information discrimination”, which arises when the distribution of resources and burdens depends on arbitrarily varying levels of information on risk. Individuals whose risks are cheaply identified will have their risks personalized, whereas individuals whose risks are expensive to identify will have their risks spread across society. Information discrimination posits a significant challenge to the assignment of healthcare burdens based on predicted healthcare costs for any proposed healthcare allocation framework. The increased risk brought to the collective pool by an individual or group may be a reason to place more burdens on them. But the high cost of information to determine that increased risk may be reason to simply spread that increased risk across society. A theory of healthcare allocation must then adjudicate between these relevant reasons in such a way that the reason to discriminate based on the cost of information outweighs the reason to discriminate based on risk.