The Four Principles Applied Inside an MMI Answer

There is a version of the ethics answer that every assessor has heard several hundred times. It goes: this scenario raises the four principles of medical ethics, which are autonomy, beneficence, non maleficence and justice. Autonomy means the patient has the right to make their own decisions. Beneficence means acting in the patient's best interests. And so on.

By the time the candidate reaches justice, forty seconds have gone and nothing has been said about the actual scenario. The framework is not the problem. Reciting it instead of using it is.

What the four principles are for

The four principles approach, most associated with the work of Beauchamp and Childress, was never intended as a definition list. It is a checklist for making sure a difficult case has been looked at from more than one direction before someone acts. Its value is coverage, not vocabulary.

That matters for an MMI because the failure mode of nervous candidates is tunnel vision. You see one obvious feature of a scenario, usually the patient's wishes, and build the entire answer on it. The four principles exist to widen the frame fast, which is exactly what you need in a two minute station.

Why naming them scores badly

Look at how MMI scoring works in Australia and you will notice that ethics rubrics tend to describe behaviours like identifying the competing interests, considering multiple perspectives, and reaching a defensible position. None of those say anything about terminology.

Definitions also make you sound like every other candidate on the day, which is a real cost when an assessor is scoring twelve people through the same door. Worse, they signal preparation rather than thinking, and an experienced assessor will usually respond by probing the one thing your recital did not cover.

Turn each principle into a question

The conversion is simple, and it happens in your head, not your mouth.

Autonomy: whose decision is this, and can they make it

Not just what does the person want, but who is entitled to decide, whether they have the capacity and information to decide, and whether the decision is genuinely theirs or is being shaped by someone else. A scenario involving a teenager, a family member translating, or a person under pressure from a partner is an autonomy question long before anyone refuses anything.

Beneficence: what would actually help this person

The useful word is actually. Not what is clinically available, but what would improve this person's life as they define it. This question is where good answers separate from adequate ones, because it forces you to consider that the beneficial option may not be the medical one.

Non maleficence: what could this cost, and to whom

Harm is rarely only physical. Consider the harm of being told, the harm of not being told, the harm to trust, the harm of a precedent, and the harm of doing nothing. Candidates routinely forget that inaction is a choice with consequences of its own.

Justice: who else is affected by this

The most neglected of the four and the easiest place to sound thoughtful. Other patients waiting. Colleagues who will carry the workload. People in similar situations who would expect the same treatment. Whether the rule you are proposing would survive being applied to everybody.

Running the four questions in reading time

Used as a silent scan, the framework costs about ten seconds and reliably produces two or three points you would otherwise have missed.

• Read the prompt, then ask the four questions in order, spending about two seconds on each.
• Notice which two produce the most friction. That friction is the tension the station is built around.
• Write down the two words that name them, and nothing else from the framework.
• Build your answer around that tension, using ordinary language rather than the technical terms.
• Keep the two principles you did not use in reserve. They are exactly what a follow up question will reach for.

What it sounds like when it works

Take a familiar shape: a competent adult refusing a treatment their family is desperate for them to have. A recital answer names four principles and defines them. A working answer sounds like a person thinking.

The central tension here is between respecting a decision that is genuinely hers to make and the fact that the decision will cause real harm to her and considerable distress to the people around her. My first question is whether the refusal is properly informed, because a refusal based on a misunderstanding is a communication problem rather than an ethical one. If she understands and still refuses, then the family's wishes, however sincere, do not override hers, and my job shifts to supporting both of them rather than changing her mind.

Every one of the four principles is present in that. Not one of them was named. That is the standard to aim for.

When is it fine to say the words

Occasionally, and briefly. If an assessor asks directly what framework you are using, name it and move on in one sentence. If a single term genuinely compresses an idea, such as saying a patient's autonomy is the decisive factor here, use it. The rule is that the word should replace a longer explanation, not substitute for one you have not done.

The mistakes that recur

• Treating the four as a scoring checklist and forcing all of them in, including the one that is irrelevant.
• Using autonomy as a trump card that ends the discussion rather than as one consideration among several.
• Assuming beneficence means whatever the doctor thinks is best, which is where paternalism sneaks back in wearing a friendly label.
• Skipping justice entirely, then being caught out when the follow up asks about the other people on the ward.
• Naming the principles and then never reaching a position, so the answer surveys the ethics without doing any.

How to drill it

Take a set of MMI ethical stations and, for each one, write only the two principles in tension. No answer, twenty seconds per scenario. Do fifteen. You are training the scan, and the scan is the part that has to be automatic on the day.

Then answer a smaller number of them out loud under a clock, drawn from a wider list of MMI questions used in Australia, with one rule: you are not allowed to say any of the four words. It feels impossible for about three attempts and then it rewires how you talk about ethics.

If you want the answer scored rather than praised, MasterMed's AI interviewer runs timed stations and marks against a rubric, and the first speaking station on the trial is free, no card, with no automatic conversion.

The framework is genuinely useful. Just keep it where it belongs, which is in your head during reading time, doing the quiet work of making sure you have not missed anything before you open your mouth.