The cusp: ethics of the learning curve



There is a moment in medical training when you think you are ready to go it alone. The difficulty is deciding when that moment has arrived. Independence, working without supervision, is a watershed moment.

Imagine this situation. A gastroenterology registrar who believes that she is ready to deal with bleeding ulcers receives a phone call. A patient is bleeding in the ER. She makes arrangements to bring the patient to the endoscopy unit. She decides not to call her consultant because he has said on a number of occasions that she is ready. He has ‘signed her off’. The patient now awaits her; she takes the endoscope and passes it into his mouth. She finds the ulcer quickly and knows what to do. But it is bleeding rapidly, and the views that she obtains are not very clear. She knows what to do. She washes the ulcer, tries to clean the blood away, but still it bleeds. She begins to feel nervous…even more nervous. She asks for a needle with which to inject adrenaline, hoping that this will slow the bleeding down. Then she might see enough to apply some definitive therapy, a clip or thermal coagulation.

She waits for the endoscopy nurse to get things ready, and watches the patient. The elderly man is sedated, but his pulse rate continues to climb despite the blood transfusion. The registrar knows that she would rather her supervisor was here. But then she reflects – this is what independence is about. Coming out of your comfort zone, absorbing the stress, dealing with the situation, making the decisions, …enlarging that zone… making yourself better so you can treat the next patient with even greater confidence and skill. But what if that process involves putting this frail man at risk? She readies herself for the next part of the procedure. She knows that if this goes well she will emerge from that room a better doctor.

The scenario can now go one of two ways…

A. She injects the adrenaline and as she had hoped it has a constricting effect on the blood vessels, causing the flow to slow down. Now when she washes the blood away it takes longer to ooze back, and she can see the culprit in the middle. A raw artery that has been eroded by acid. She chooses to use the heater probe and asks the nurse to make it ready. She passes it down the channel of the endoscope until she can see it emerge on the screen. With one hand she controls the wheels on the endoscope to optimise the position, and with the other she presses the heater probe onto the vessel. Then, with her right foot she presses a pedal on the floor and sends electricity into the probe, creating a tiny zone of intense heat until the vessel is ‘cooked’. Thinking that she has sealed artery she pulls the probe away. But the bleeding is even worse now. She must have torn the wall away. Quickly she calls for a clip and the nurse passes her the kit. The registrar pulls out the heater probe and quickly, calmly, replaces it with the clip delivery device. Soon she sees the metal jaws, grossly magnified, floating around the field of scarlet on the TV screen, and although the view is flooding with blood again she can still glimpse the artery. Before the window of opportunity has passed she pushes the clip onto the artery and asks the nurse to deploy. The clip closes down on the artery and stops the bleeding immediately. The field clears and she places two more clips above and below. The patient is stable.

“Well done,” says the nurse “that wasn’t easy. You could tackle anything now.”

B. She continues to wash away the blood, but the view is terrible. She injects adrenaline and it slows things down, improving the view. Now the time is right to use the heater probe. She places its tip on the ulcer, right on the vessel that is spurting blood, and presses the pedal. The heat dissipates into the pool of blood, and makes little difference. She knows that a clip is the next best thing, but she needs a better view. She readies the clip, and through the other channel of the endoscope she squirts more water. The view improves and when the time is right she deploys the clip. For a while she thinks that the job is done, she sighs in relief and smiles at the nurse, but 30 seconds later the bleeding erupts again and all views are lost. She takes out the camera, bleeps another member of staff to come and help look after the patient, and rings a surgeon. He arrives quickly, but argues that the patient is too frail to undergo an operation. The registrar argues back, saying that she can do no more within the scope. They debate the pros and cons of various other treatments, and in the end agree that surgery is the only hope. The patient is anaesthetised and in the operating theatre 40 minutes later. The ulcer is located and sealed. But he does badly after the operation, and develops a chest infection. He remains on a ventilator, and in the end, seven days later, dies.

The registrar describes all this to her consultant. From the description he can see no reason for her to blame herself.

“You did fine.” he says, “You knew when to give up, that’s half the skill.” He doubts that his presence would have made much of a difference. But the registrar knows that the patient’s greatest chance of survival would have been afforded if he had not had to have an operation, if the most experienced person had been there to treat the ulcer…if she had not proceeded on her own.


As a trainee approaches the top of a learning curve the moment comes when they have to decide if it is safe to go it alone – the cusp. The patient who comes into hospital on that day will have no idea that they represent a significant moment in the career of the doctor who is called to perform their procedure. They will have no idea where they lie on that learning curve, or that they might form a stepping stone to independence and immaturity. This would not matter if their risk of harm was no greater than that of any other patient having the procedure. But it is the result of this risk analysis that forms a perfect example of how we balance individual risks versus societal benefits in medicine.

The concept of the learning curve was introduced to many members of the public in a horribly vivid way during the Bristol Heart scandal. One of the paediatric heart surgeons involved said,

“I believe that the reality of the learning curve may be illustrated by the evolution of surgery for transposition of the Great Arteries in this country … in the late 80s and the very early 90s it was generally understood and accepted that when a unit introduced the Arterial Switch operation for neonates there would initially be a period of disappointing results.”

I am not concerned with such extreme examples here, but the essence pertains. In order to achieve complete expertise it is necessary to accept a degree of ‘trial and error’. Or is it?


A thorough enquiry into this subject was undertaken by a US paper, the Dallas News, following controversial reports coming out of Parkland Memorial Hospital, the primary teaching institution of University of Texas Southwestern Medical school. This hospital seemed to take a liberal attitude to the surgical training, crediting its juniors with autonomy to proceed with many operations unsupervised.

One faculty supervisor who quit in protest said the mainly poor, minority patients of Dallas County’s only public hospital had effectively become “clinical fodder.”


The head of UT Southwestern’s general surgery residency program once said it was “OK for residents to make mistakes” on patients “even if they could have been avoided with better faculty supervision,” according to notes taken by a faculty surgeon and later included in court records. Tim Doke, UT Southwestern’s spokesman, challenged the accuracy of that account. But Anderson has testified that some faculty believed “that’s how people learn,” though he said he disagreed with the philosophy.

In this case (and the newspaper report makes excellent reading, as does this graphic summarising mortality), one supervisor became uncomfortable, and complained after he was called into a gall bladder operation too later, after a irreversible damage had been done to the bile duct.

This controversy crystallises an ethical dilemma in medical training. As the journalists put it, “There’s good for the patient, and there’s a societal good. We can’t exist as a society without physicians learning on the ground.”

A questionnaire study published in the BMJ found that 86% of surgical trainees or young consultants had performed procedures for the first time without direct supervision. This appears to be the reality of medical education. Attempts have been made to resolve the dilemma, another BMJ paper seeking to lay out a framework based on respect for the individual, beneficence and non-maleficence. In their introduction Jagsi and Lehmann explain that…

The burdens of medical education are not currently distributed fairly. In one US study, students saw disproportionately high numbers of non-white patients and patients with Medicaid (public insurance for the indigent).Another study found that children of doctor parents were less likely to be seen by trainees than were other children.


Immanuel Kant (image from Wikipedia)

However, the approach laid out in this paper does not really equip trainees with a practical method of making decisions on the spot. Another paper (Journal of Medical Ethics) approaches the problem by applying Immanuel Kant’s Second Formulation of the Categorical Imperative,

‘‘Act so that you use humanity, as much in your own person as in the person of every other, always at the same time as end and never merely as means”

In reality however,

This conflict arises because, at least presently, medical practitioners can only acquire certain skills and abilities by practising on live, human patients, and given the inevitability and ubiquity of learning curves, this learning requires some patients to be treated only as a means to this end….Accordingly, until a way is found to reconcile them, we conclude that the Kantian ideal is inconsistent with the reality of medical practice.

To resolve this conflict,

…supervisors might undertake to delegate only under conditions where they can be as sure as possible that the procedure would be done as well as they could do it themselves. If this assurance can truly be given by the supervising doctor, then the conflict is solved.’


This seems unrealistic. So are the patients who take their place on our learning curves nothing more than a means to an end? The paper begins with a quotation from Atul Gawande’s book Complications: a surgeon’s notes on an imperfect science

‘To fail to adopt new techniques would mean denying patients meaningful medical advances. Yet the perils of the learning curve are inescapable—no less in practice than in residency’

Le Morvan and Stock seek to challenge the perception that patients are guinae pigs in four ways;

1) Discontinuing unnecessary use of patients without consent – they suggest that we introduce a consent process where possible. The example of pelvic examinations by students on anaesthetised patients is one such example.

2) Continuing to develop medical simulation models

3) Enhancing supervision, but…”We are sceptical that such an approach, applied stringently, is practical for all procedures. It is hard to imagine—for example, that an experienced surgeon can honestly say that his trainee’s first liver biopsy will be performed just as well as he would perform it himself. Moving in this direction, toward a more conservative educational model, would, however, reduce the extent to which patients are used as means only.”

4) Changing expectations, or universalising the problem. If the involvement of trainees is taken into account when the statistical outcome from a procedure is calculated, patients waiting for that procedure are not actually being disadvantaged by having performed by a trainee. This argument does have a whiff of sophistry about it, but I have found myself using it before. As a patient and a parent I would want the hospital’s best qualified person to treat me or my children (although I am probably too polite to demand as much), but as a trainee I often muttered to myself, in response to a patient’s underwhelmed expression, “Look, either I do this procedure or it’s another five hour wait…what will it be?” As the authors conclude,

It does, however, offer a useful way of approximating this ideal in light of the constraints imposed by the reality of medical practice.

I don’t think there is a way of truly resolving the Kantian conflict unless our patients accept that it is not possible to always see the most qualified person in the institution. But the deal must be reciprocated by trainees – they must ensure that every single clinical interaction is approached not from the point of view of ‘polishing their resume’ (as the Dallas News article put it), but from the point of view of the patient. The trainee may well be on the cusp, there may be a theoretically increased risk, but if the skills are embedded, if the trainers have given their blessing, if they feel ready on that particular day or night…no more can be asked.

cover to tweet real book

Click e-book or paperback icon to explore

2 thoughts on “The cusp: ethics of the learning curve

Add yours

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Blog at

Up ↑

%d bloggers like this: