Resource Allocation: Fairness Frameworks Explained

One bed, two patients. One donor organ, three candidates. One hour of your clinic, four people who need it. These stations are common because they are clean: the scarcity is stipulated, so you cannot escape into finding more resources.

Most candidates answer them with a gut call and some justification stapled on afterwards. The answer that scores is the one that notices something first: the word fair means at least four different things, those meanings disagree with each other, and picking between them is the actual work.

Four definitions of fair

Learn these as competing criteria, not as a checklist. Each is genuinely defensible and each produces a different answer to the same prompt.

• Need. The sickest person first. It matches most people's moral instinct and it underpins triage in an emergency. Its weakness is that the person in greatest need is sometimes the person least able to be helped.
• Benefit. Whoever will gain the most from the resource. This is how you get the most health out of a fixed budget. Its weakness is that it can quietly abandon people who are hard to treat, and it treats individuals as means to a total.
• Equity. Attention to who has been disadvantaged, and to gaps between groups. It asks not just who is sickest today but why they are sicker, and whether the allocation entrenches an existing inequality. Its weakness is that it is harder to apply case by case and can feel arbitrary at the bedside.
• Lottery, or first come first served. Everyone gets an equal chance and nobody is judged. It is transparent and incorruptible. Its weakness is obvious: it ignores information that clearly matters, and queueing itself favours people with time, transport and confidence.

The move that impresses is not knowing the four. It is showing that they conflict. Need and benefit pull apart in exactly the cases these stations are built from, and saying so out loud is the sentence that opens a real answer.

The criteria people reach for and should not

Under pressure, candidates smuggle in criteria that will not survive a follow up. Know them so you do not use them by accident.

Social worth is the big one: the surgeon over the unemployed man, the mother of three over the person with no dependants. It feels intuitive and it is a road nobody should be on, because it makes a clinician the judge of whose life counts. Say explicitly that you would not use it.

Responsibility for illness is the subtler one: the smoker, the drinker, the person who did not follow advice. It sounds like fairness and it usually collapses under two questions. How would you assess it consistently across every patient, and how much of that behaviour was genuinely chosen given the person's circumstances? There is a narrow, non moral version where a behaviour affects the likely success of a treatment, and that belongs under benefit rather than desert. Draw that distinction explicitly and you will handle the follow up well.

Age is the third. There is a difference between age as a proxy for prognosis, which can be legitimate clinical reasoning, and age as a measure of how much a life is worth. Keep them separate and say which one you are using.

A structure that works in four minutes

State the scarcity and refuse the escape hatch. Yes, the real answer is more beds, and the prompt has told me there is one, so I will work with that. Assessors mark down candidates who spend two minutes wishing the constraint away.

Name the criteria in play, and which two are pulling against each other here. Ask what you would want to know, since these prompts are usually missing exactly the information that would decide it. Commit to a criterion and say why it fits this decision. Then name what your choice costs and who bears it.

Finish on process. Decisions like this should not rest on one person's view in a corridor, so I would want agreed criteria set in advance, a second opinion, and a record of the reasoning. That last move is important, because in real systems the fairness of the procedure carries as much weight as the fairness of the outcome.

Say who pays

Every allocation answer has a loser, and candidates instinctively hurry past them. Do the opposite. If I allocate on benefit, the person I am not treating is the one who is sickest, and I do not think that is a comfortable thing to say. I think it is still the right call here, for these reasons.

This is the difference between an answer that has been weighed and one that has been chosen. Our guide to MMI ethical stations in Australia goes into why acknowledging the cost of your position tends to strengthen rather than weaken it.

You should also say what you would do for the person who missed out. Allocation is not only about who gets the resource; it is about whether the other person is still cared for, told honestly what is happening, and not simply dropped.

The two levels these questions live on

Some prompts put you at the bedside choosing between two people in front of you. Others put you on a committee deciding a policy for thousands you will never meet. The criteria are the same; the right emphasis is not.

At the bedside, advocacy for the patient in front of you is part of the role, and pure benefit maximising can feel like a betrayal of it. At policy level, the maths is unavoidable and consistency matters more than individual advocacy. Noticing which level a prompt is set at, and saying so, is a strong opening move.

It also gives you a clean answer to a classic follow up: would you feel differently if you knew the patient? Yes, honestly, and that is exactly why decisions like this are made against criteria set in advance rather than in the moment.

How to practise it

Take one allocation scenario and answer it four times, once under each criterion, arguing each properly. It takes about fifteen minutes and it does something no amount of reading does: it makes the criteria feel like tools you can pick up rather than words you can name. Broader sets of MMI interview questions used in Australia will give you enough scenarios to rotate through.

Then practise defending one position against an interrogator who keeps changing a variable: now the younger patient has a worse prognosis, now the older one has dependants. How many rounds of that you get depends on format, and our comparison of MMI and panel interviews in Australia sets out the difference. Formats vary between universities and change over time, so check the university's current admissions page for what you are sitting.

The measure of whether you have learned this is whether your reasoning survives being pushed, which is hard to test alone. MasterMed's live AI interviewer runs timed MMI stations and marks you against a rubric, so you can find out whether your criterion held or whether you quietly switched to a different one halfway through. The first speaking station is free on the trial, no card, and the trial does not convert by itself.

There is no correct answer waiting on the assessor's sheet. There is a set of defensible answers and a much larger set of undefended ones. Name your criterion, apply it consistently, admit its cost, and say what would make you change it. That is the whole exercise.