Follow us on

Big Thinkers: Thomas Beauchamp & James Childress on medical ethics

by The Ethics Centre
15 August 2017
BIG THINKERS
Two people brought the ethics of medical consent and care into the Victorian Supreme Court by challenging their forced shock therapy treatment today and yesterday. They inspired us to feature a double Big Thinker act: Beauchamp and Childress, the creators of the core principles of medical ethics.

Thomas L Beauchamp (1939-present) and James F Childress (1940-present) are American philosophers who are best-known for their work in medical ethics. Their book Principles of Biomedical Ethics was first published in 1985, where it quickly became a must-read for medical students, researchers and academics.

Authors Beauchamp and Childress worked on the first edition of the book (it’s now in its seventh edition) in the wake of some horrific biomedical experiments – most notably the Tuskegee Study, where hundreds of black American men were not informed they had syphilis, nor were they given treatment for it.

By the end of the study, over 100 people had died from the disease or related complications, many had infected their partners and a number of children were born with a congenital version of the illness.

The lack of ethical governance over medical research led to the Belmont Report, produced by the US National Commission for the Protection of Human Subjects of Biomedical and Behavioural Research. They were tasked with identifying “the basic ethical principles that should underlie the conduct of biomedical and behavioural research”.  

They identified three key ethical principles for research ethics: respect for persons, beneficence and justice.

This is significant, because one member of the National Commission was Beauchamp, who was at the same time writing Principles of Biomedical Ethics with Childress, which would become just as influential as the Belmont Report itself.

Like the Belmont Report, Beauchamp and Childress were looking for basic ethical principles for healthcare. They described their project as identifying “a common morality” – “shared by all persons committed to morality”, applicable to everyone at all times and all places.

What are the cornerstones of this common morality? How do they apply to medical ethics? For Beauchamp and Childress, it all boils down to four principles:

 
  1. Respect for autonomy
  2. Beneficence
  3. Non-maleficence
  4. Justice

These principles are often in tension with one another, but all healthcare workers and researchers need to factor each into their reflections on what to do in a situation.

Respect for autonomy

Philosophers usually talk about autonomy as a fact of human existence. We are responsible for what we do and ultimately any action we take is the product of our own choice. Recognising this basic freedom at the heart of humanity is a starting point for Beauchamp and Childress.

By itself, the idea human beings are free and in control of themselves isn’t especially interesting. However in a healthcare setting, where patients are often vulnerable and surrounded by experts, it is easy for a patient’s autonomous decision to be disrespected.

Beauchamp and Childress were writing at a time when the expertise of doctors meant they often took extreme measures in doing what they had decided was in the best interests of their patient. They adopted a paternalistic approach, treating their patients like uninformed children rather than autonomous, capable adults. This went as far as performing involuntary sterilisations. In one widely discussed court case in bioethics, Madrigal v Quillian, ten Latina women in the US successfully sued after doctors performed hysterectomies on them without their informed consent.

Are there any boundaries if consent is there? | The Ethics of Consent | Sydney | 30 August | Buy tickets here

Strictly speaking, the women in Madrigal v Quillian had provided consent. However, Beauchamp and Childress explain clearly why the kind of consent they provided isn’t adequate. The women – who spoke Spanish as a first language – were all being given emergency caesareans. They were asked to sign consent forms written in English which empowered doctors to do what they deemed medically necessary.

In doing so, they weren’t being given the ability to exercise their autonomy. The consent they provided was essentially meaningless.

To address this issue, Beauchamp and Childress encourage us to think about autonomy as creating both ‘negative’ and ‘positive’ duties. The negative duty influences what we must not do: “autonomous actions should not be subject to controlling constraints by others”, they write. But positively, autonomy also requires “respectful treatment in disclosing information” so people can make their own decisions.

Basically, respecting autonomy isn’t just about waiting for someone to give you the OK. It’s about empowering their decision making so you’re confident they’re as free as possible under the circumstances.

Non-maleficence: ‘First do no harm’

The origins of medical ethics lie in the Hippocratic Oath, which although it includes a lot of different ideas, is often condensed to ‘first do no harm’. This principle, which captures what Beauchamp and Childress mean by non-maleficence, seems sensible on one level and almost impossible to do in practice on another.

Medicine routinely involves doing things most people would consider harmful. Surgeons cut people open, doctors write prescriptions for medicines with a range of side-effects, researchers give sick people experimental drugs – the list goes on. If the first thing you did in medicine was to do no harm, it’s hard to see what you might do second.

This is clearly too broad a definition of harm to be useful. Instead, Beauchamp and Childress provide some helpful nuance, suggesting in practice, ‘first do no harm’ means avoiding anything which is unnecessarily or unjustifiably harmful. All medicine has some risk. The relevant question is whether the level of harm is proportionate to the good it might achieve and whether there are other procedures that might achieve the same result without causing as much harm.

Beneficence: Do as much good as you can

Some people have suggested Beauchamp and Childress’s four principles are three principles. They suggest beneficence and non-maleficence are two sides of the same coin.
Beneficence refers to acts of kindness, charity and altruism. A beneficent person does more than the bare minimum. In a medical context, this means not only ensuring you don’t treat a patient badly but ensuring you treat them well.

The applications of beneficence in healthcare are wide reaching. On an individual level, beneficence will require doctors to be compassionate, empathetic and sensitive in their ‘bedside manner’. On a larger level, beneficence can determine how a national health system approaches a problem like organ donation.

In Australia, organ donation is voluntary. Unless you sign up, the health system assumes you don’t want your organs removed and given to someone else when you die. However, beneficence might suggest reversing this process. Our healthcare system would do more good if we had more organs to give. One way to do that is assume people do consent unless they said otherwise – an ‘opt out’ instead of ‘opt in’ system.

The principle of beneficence can often clash with the principle of autonomy. If a patient hasn’t consented to a procedure which could be in their best interests, what should a doctor do?

Beauchamp and Childress think autonomy can only be violated in the most extreme circumstances: when there is risk of serious and preventable harm, the benefits of a procedure outweigh the risks and the path of action empowers autonomy as much as possible whilst still administering treatment.

However, given the administration of medical procedures without consent can result in legal charges of assault or battery in Australia, there is clearly still debate around how to best balance these two principles.

Justice: Distribute health resources fairly

Healthcare often operates with limited resources. As much as we would like to treat everyone, sometimes there aren’t enough beds, doctors, nurses or medications to go around. Justice is the principle that helps us determine who gets priority in these cases.

However, rather than providing their own theory, Beauchamp and Childress pointed out the various different philosophical theories of justice in circulation. They observe how resources are distributed will depend on which theory of justice a society subscribes to.

For example, a consequentialist approach to justice will distribute resources in the way that generates the best outcomes or most happiness. This might mean leaving an elderly patient with no dependents to die in order to save a parent with young children.

By contrast, they suggest someone like John Rawls would want the access to health resources to be allocated according to principles every person could agree to. This might suggest we allocate resources on the basis of who needs treatment the most, which is the way paramedics and emergency workers think when performing triage.

Beauchamp and Childress’s treatment of justice highlights one of the major criticisms of their work: it isn’t precise enough to help people decide what to do. If somebody wants to work out how to distribute resources, they might not want to be shown several theories to choose between. They want to be given a framework for answering the question. Of course when it comes to life and death decisions, it’s quite likely there are no easy answers.

Follow The Ethics Centre on TwitterFacebookInstagram and LinkedIn.

Twitter-Logo.png Facebook-Logo.png instagram-logo-sketch-copy.png linkedin-logo.jpg