Guest Post: Samuel Altman, University of Oxford

Full Article: Against Proportional Shortfall as a Priority-Setting Principle

Past health is regularly considered irrelevant in priority-setting decisions. Often, people mistakenly think of past health, or rather past ill-health, as a ‘sunk’ cost which can be ignored when making decisions about present and future health. However, past health is not always sunk and the continuing effects of past ill-health can impact present and future health. Therefore, even though it is the continuing effects of past health that are relevant for priority-setting decisions, past health should be taken into account as a proxy for these continuing effects which are often too broad to be properly considered from a practical standpoint.

In my recent article I argue that the much-lauded ‘Proportional Shortfall’ priority-setting principle, which ignores anything from before a patient got ill, is not the silver bullet it was made out to be. Proportional Shortfall has gained attention in recent years as various health systems have been considering formally adopting the principle for their priority-setting decisions. Justification for ignoring past health typically runs along the lines that it is sunk in the past. Past good health is considered a ‘sunk good’ and past illnesses a ‘sunk cost’. It then follows that when it comes to priority-setting decisions we should only take into account the things we can influence – present and future health.

There are multiple occasions when past health is not ‘sunk’ in the past, and the continuing effects of past health have a direct impact on future health. This means that past health is at least sometimes relevant for priority-setting decisions. It is worth considering, here, when and why past health is relevant for priority-setting decisions.

Consider the following common example: Suppose we have a single cochlear implant that we can give to one of two deaf people – call them patients A and B – and we must decide which individual deserves the greater priority. Suppose both individuals are aged 40 and both would like the cochlear implant. Patient A went deaf as an adult aged 30, while patient B went deaf as a child aged 5. The question is whether we should take into account patient B’s additional 25 years of past deafness in our decision. These additional years of deafness have an effect on patient B’s future. Going deaf as a child has been shown to affect language and social development, thereby affecting their social skills as an adult. Therefore patient B is, in some sense, worse off than patient A as they will face this additional disadvantage caused by their childhood illness. For example, their potentially worse social skills will most likely make it harder to find gainful employment and harder to get on with their peers. We could argue, therefore, that this past health ought to be taken into account, given its affect on their present and future.

That said, it is not patient B’s relatively poorer past health that is affecting their future. Rather, it is the state of their language and social skills in the present that affects their future. Another way to put this: It isn’t their past health that is affecting their future health, it is the continuing effects of their past health (in the present) which affects their future.

Instead of considering past health, we could instead make the case that we ought to take into account these kinds of broader factor when making our priority-setting decisions; if we could to take into account all the possible effects of their past health there would be no need to take into account past health itself. However this approach is impractical. It is not feasible to take into account all the factors in the present that might affect an individual’s future on account of their past health. Which factors should be taken into account? We could try to consider their language skills, possible disabilities, years of schooling and so on. Meanwhile accounting for lost opportunities is even more problematic. Taking into account what has been lost through past health, relative to what might have been, runs into difficulty.

We often justify focusing on health when it comes to priority-setting because health acts as a strong proxy for welfare, rather than considering any broader welfare-relevant factors directly. We could therefore extend this logic to overcome the difficulty of identifying opportunities lost and assume that past health works as an effective proxy for the effects of past health. That is, we may as well take past health into account in our priority-setting decisions in order to indirectly take into account the continuing effects of past health. This is one possible way of justifying taking into account past health when it comes to priority-setting – as a means of indirectly taking into account whether an individual is likely to be worse off in some other relevant way.

Competing interests: None

Author affiliations: The University of Oxford