Home › Blog › Ethical frameworks

The Four Principles of Medical Ethics: Autonomy, Beneficence, Non-maleficence and Justice, With Real UK Examples

The four principles of medical ethics are respect for autonomy, beneficence, non-maleficence and justice. They come from Tom Beauchamp and James Childress, whose book Principles of Biomedical Ethics was first published in 1979 and is now in its eighth edition. The framework, often called principlism, is taught in almost every UK medical school and used by clinical ethics committees, and it travels well: health tech companies, care providers, pharmaceutical firms and anyone handling patient data use it to test decisions. This page explains each principle, how UK law gives it teeth, where the principles clash, and how to use them in practice.

Where the four principles come from

Beauchamp and Childress set out the four principles as a common ground that people with different moral theories could agree on, whether they start from consequences, duties or virtues. Rather than deriving every answer from one master theory, the approach asks you to weigh four mid-level principles against the facts of a case. In the UK, the surgeon and ethicist Raanan Gillon did much to popularise the approach through the Journal of Medical Ethics and the BMJ, describing it as “four principles plus attention to scope”: you also have to ask to whom each principle is owed.

If you want the broader theory behind it, our guide to ethical theories in business covers utilitarianism, deontology and virtue ethics.

1. Respect for autonomy

Autonomy means respecting the right of a capable adult to make decisions about their own body and care, including decisions others think unwise. In practice it rests on three things: informed consent, confidentiality and honesty.

UK law has made this principle concrete. In Montgomery v Lanarkshire Health Board [2015] UKSC 11, the Supreme Court held that doctors must take reasonable care to make sure a patient knows about any material risks of a treatment and about reasonable alternatives. What counts as material is judged by what a reasonable person in the patient's position would want to know, not by what doctors usually tell patients. The Mental Capacity Act 2005 adds that an adult must be assumed to have capacity unless it is shown otherwise, and that making an unwise decision does not, by itself, mean someone lacks capacity.

Example: a patient with capacity refuses a blood transfusion on religious grounds. Respect for autonomy means the refusal stands, even though the clinicians believe the transfusion is in the patient's interest.

2. Beneficence

Beneficence is the duty to act for the patient's good: to provide benefit, prevent harm and promote welfare. It covers everyday clinical judgement, such as recommending the treatment with the best balance of outcomes, and wider duties such as keeping skills up to date. The General Medical Council's Good Medical Practice, the current version of which took effect on 30 January 2024, makes caring for patients and acting in their interests a core professional duty.

Example: a GP spots that an older patient's repeat prescriptions interact badly and arranges a medication review the patient did not ask for.

3. Non-maleficence

Non-maleficence, from the Latin maxim often summarised as “first, do no harm”, means not inflicting harm, or a risk of harm, that is out of proportion to the benefit. Almost every treatment carries some risk, so this principle is about weighing, not avoiding all harm. It underpins rules on competence, safe staffing, infection control and stopping treatments that no longer help.

Example: a surgical team decides not to operate on a frail patient because the risk of dying on the table outweighs the expected benefit, and offers palliative care instead.

4. Justice

Justice means fairness: treating like cases alike, distributing scarce resources fairly and respecting people's rights and the law. In the NHS it shows up in waiting list priorities, in the National Institute for Health and Care Excellence judging whether new medicines are cost-effective, and in organ allocation rules. It also covers equality: the Equality Act 2010 prohibits discrimination in the provision of services, including health and care.

Example: an integrated care board decides how many treatments of a costly new drug it can fund, knowing that money spent there is not available for other patients.

When the principles conflict

The principles are not ranked, so the hard cases are where they pull in different directions:

• Autonomy against beneficence: a capable patient declines a treatment that would very likely help. UK law generally sides with autonomy for adults with capacity.
• Confidentiality against harm to others: a patient with a condition that makes driving dangerous will not tell the DVLA. GMC guidance allows a doctor to disclose in the public interest after trying to persuade the patient.
• Beneficence against justice: an expensive treatment would help one patient, but funding it means fewer resources for others.
• Children and capacity: under the Gillick case (1985), a child under 16 can consent to treatment if they have enough understanding and intelligence to understand fully what is proposed.

The framework does not settle these clashes automatically. It gives a shared vocabulary for setting out what is at stake, so a decision can be explained and challenged.

How to apply the four principles: a five-step check

1. Get the facts: clinical options, risks, the person's wishes and capacity.
2. Test each principle: what does autonomy, beneficence, non-maleficence and justice each point to here?
3. Check scope: whose interests count: the patient, family, other patients, staff, the public?
4. Weigh the conflicts: where principles disagree, which considerations carry most weight in this case, and why?
5. Decide, record and review: write down the reasoning, and revisit it if the facts change.

This mirrors the general process in our ethical decision-making framework.

Why businesses outside medicine use them

The four principles are a useful test for any organisation whose products affect people's health or personal lives. A digital health app can ask whether users genuinely consent to how their data is used (autonomy), whether the product delivers real benefit (beneficence), what harm a wrong result could do (non-maleficence) and whether it works equally well for everyone it is sold to (justice). Care providers, insurers and employers running occupational health schemes face the same questions. Our guides to data ethics for business and what business ethics is take those ideas further.

Criticisms of principlism

Critics say the four principles are too abstract to settle real cases, that they reflect a Western emphasis on individual autonomy, and that they can be used as a checklist rather than a way of thinking. Supporters reply that no framework removes judgement, and that a shared language for disagreement is exactly what clinical teams and ethics committees need. Both points are worth keeping in mind: use the principles to structure a discussion, not to end one.

Sources: UK Supreme Court, Montgomery v Lanarkshire Health Board [2015] UKSC 11; Mental Capacity Act 2005, section 1; GMC, Good Medical Practice; GMC, confidentiality guidance.