6
Decisions, Decisions
Decades ago, patients pretty much did what doctors told them to do. The doctor, as expert, was trusted to know the right answer to a problem, and the patient, as a general rule, went along with the treatment prescribed by the expert. This is changing. The role of the passive patient has gradually shifted into that of an active consumer of health care. The catchphrase is ‘No decision about me without me’. This is an understandable and legitimate shift. In making healthcare decisions, the patient makes a choice between no treatment and what can be an array of different available treatments, each with its own set of risks and benefits, and such decisions require information and intelligent process. In fact, whether a certain patient receives a certain treatment is perhaps the most important question that doctors and patients ever tackle. From the doctor’s point of view, this is probably the one thing for which a medical education is essential. Pretty much everything else that is done by doctors and surgeons can be done by someone else with the right training.
Here are some of the things doctors do routinely: they take and record a medical history from patients, take blood samples, insert a catheter into the bladder, do an electrocardiogram, order a chest X-ray, prescribe medications, put up intravenous infusions, harvest veins from the leg for CABG, resuscitate a patient in cardiac arrest, and so on. Only 30 years ago, the mere idea of a non-doctor performing these duties would have been anathema to doctors and non-doctors alike. Nowadays, all of these tasks and more are performed by nurses and other healthcare professionals routinely and safely in many hospitals and in many parts of the world. The one thing that really, absolutely needs a full medical education is the thorny issue of selecting the right treatment for the disease, and offering it to the patient. In short, it is the decision-making. For that, doctors need a thorough education to make the right decisions, and patients need a reasonable level of knowledge to agree to or reject the treatment on offer. This is all the more important in surgery, as the consequences of the wrong decision (or the right decision that goes awry) can be disastrous. One day, you or a loved one will be faced with the offer of an operation. Deciding whether to accept that offer is not straightforward, but it can be done intelligently with a little understanding of the mechanisms necessary for decision-making. Before I explain the evidence-based way in which doctors go about this decision-making, we need to ask ourselves a simple yet crucially important question: why do doctors treat patients?
Why indeed? It is a question that I often put to my fourth-year medical students, and some of them get the answer right, but many get it wrong, so let’s begin by clarifying why doctors should not treat patients.
Poor reasons for treating patients:
All of these are not good reasons to administer medical treatment. There are only two good reasons for a doctor to treat a patient, and they are:
Any medical treatment that does not achieve at least one or the other of the above objectives should absolutely not be offered to the patient, because this would be unethical. A treatment that neither improves symptoms nor improves prognosis is at best useless and at worst both dangerous and a costly waste of resources.
The only exception to the above rule is in the field of immunisation, where on occasion it can be justified to offer vaccinations against a particular disease to all people (even those who are not susceptible to the disease) in order to increase the ‘herd immunity’ and thus eradicate or drastically reduce the impact of a nasty disease. Even then, receiving the vaccination can often have prognostic benefit to the individual, so that the rule remains intact in most vaccination programmes.
Surgery is no exception. The reason, or ‘indication’, for an operation is always either symptomatic (to reduce or abolish troublesome symptoms, such as pain, discomfort, breathlessness, itching, palpitations, and so forth) or prognostic (to improve the likelihood of survival). In other words, the symptomatic indication deals with quality of life, and the prognostic indication deals with quantity of life.
So our original question naturally branches out into two parts:
Let us deal first with the symptomatic indication, which is the simpler one of the two. In surgery (as opposed to general medicine), the symptomatic indication must fulfil an essential criterion that is always the same, regardless of the patient, the nature of the symptom, the surgical specialty, the surgeon in question, and the envisaged operation. It is a simple criterion that applies to all: there is no indication for surgery for symptoms until medical treatment has failed. The reason for this is obvious: patients in their right mind would never choose to have a hairy-armed surgeon cut them open with a knife if there is a tablet that achieves the same symptomatic relief. So the first question becomes: is the symptom successfully controlled by tablets? If the answer is ‘Yes’, keep taking the tablets. If the answer is ‘No’, consider surgery.
When medical treatment with tablets has failed, and surgery is being contemplated for symptoms, the next decision is whether to take the plunge and have the operation, and that is where risk comes into it. The patient must weigh up how troublesome the symptom is against the risk of the operation. For example, if a patient has angina, and the risk of a CABG to cure the angina, as calculated by a risk model (and perhaps adjusted to the surgeon’s own performance) is a mortality of 1 per cent, the patient needs to decide if he or she is prepared to accept a risk to life of 1 per cent to get rid of the angina. This is a relatively easy decision to make: the patient knows how troublesome the angina is, and its impact on quality of life, and most patients can easily understand what 1 per cent, or one in 100, means. The patient can therefore make an informed decision as to whether he or she thinks the benefit of the operation is worth the risk.
Deciding on having an operation for prognosis is much more problematic. On the face of it, nothing could be easier, as no symptoms are involved, only prognosis (survival) and risk. Because the operation is being contemplated for survival, we know that the condition or disease must carry a risk to life if not treated: in other words, not having the operation carries a risk. The operation itself also carries a risk. All we have to do is find out which risk is the smaller one, and, hey-presto, we have a decision.
Unfortunately, there is a catch.
The problem is that the risk of the operation is immediate, upfront, now. The risk of not having the operation is spread over time. If you have a condition that might kill you in the future, and you are thinking about having an operation to fix it tomorrow morning, you can be fairly certain that tomorrow night you are more likely to be alive if you do not have the operation. But you are not simply interested in tomorrow; you also care about next week, next month, and next year. You may also, depending on how young you are, be acutely interested in the next decade, or even the next 20 or 30 years. So how do you decide? And what information do you need to help you do so?
Let us assume that you are having a heart operation with a total risk to life of 5 per cent (3.5 per cent during the hospital stay, and another 1.5 per cent in the first three months after surgery), after which your survival becomes ‘normal’, in that it will follow the normal pattern for people of your age and sex, as shown by the graphs built from data used by insurance companies. The survival curve for a group of 100 patients just like you having surgery would look like this:
There is a sharp dip immediately after the operation, and thereafter the curve levels off and follows the normal survival pattern for people without the disease.
Now let us assume that you have decided not to have the operation, and that the disease kills about 5 per cent of the patients who carry it every year. Your survival graph without an operation would look something like this:
To compare the two, we superimpose the graphs one on top of the other, and end up with something like this:
It is pretty obvious that, up to about a year, you would be more likely to be alive if you have not had an operation. In fact, at about a year, the same percentage of patients is alive in both groups, but more years of life have been lost in the group who opted for surgery. From about a year onwards, things change, and the ones who have chosen to have an operation begin to outlive those who chose not to have an operation. Now look at the next graph:
It is identical to the previous one, except that two areas between the lines have been shaded. These areas are equal in size, and they are the difference in years of life lost between those who had surgery and those who had not. When the lines cross, an equal number of people are alive in the two groups, but those who had surgery lost more years of life. At just over 24 months, the ones who did not have surgery have lost as many years of life as the ones who did, and, from then onwards, surgery can only be good for your survival. That point is crucial, and the length of time it takes you to reach it after an operation is called ‘time until treatment equipoise’, or TUTE for short (Noorani 2014). When TUTE takes place is a vital piece of information that can help doctors and patients decide whether or not to have an operation purely for survival. The range can be astounding. In heart conditions where an operation is carried out to improve survival alone, TUTE can range from a few hours to a few decades.
Remember, however, that most operations are still carried out for symptoms, and the decision in such a scenario is easy: do the symptoms bother you enough to justify the risk? If so, have it. If not, tell the surgeon to put away the scalpel and leave you alone. Increasingly these days, however, patients who feel fine are being offered surgery for prognosis. When that happens, they (and their doctors) should know what the TUTE is, because it can be very, very different from one patient to the next and from one condition to the next. For example, an otherwise healthy middle-aged man with a narrowing in the left coronary artery has a TUTE of only a few weeks for a CABG operation: if he opts for surgery, a few weeks down the line he is more likely to be alive than if he opts for no surgery, so that it is virtually a no-brainer that he should take the plunge. On the other hand, for an 85-year-old woman with an aneurysm of the aorta, the TUTE may well be six years or more: that means it will be at least six years after the operation before she is more likely to be alive with than without surgery. Armed with knowledge, she may well politely decline the heroic surgeon’s plan for fixing her aorta.
The TUTE concept is relatively new, but I have tested it on many patients from all backgrounds and all levels of educational attainment, and they all understood it perfectly. Of course, we cannot calculate TUTE unless we know for sure the risk of an operation and the natural history of the disease without an operation. These are not fully and exactly determined for all conditions and all operations, but we already have enough data in the medical literature to work things out for many conditions and procedures, and, as time passes, we will know more and more. I would like the TUTE method to be used in decision-making by both patients and doctors when operations are planned for the sole purpose of improving survival, and I believe that, one day, when we know with reasonable certainty the value of TUTE for all types of conditions, it may well be. When that is done, the TUTE concept may also be applied to quality of life. Many operations are designed to improve quality of life, but there is no doubt that, immediately after surgery, an operation worsens quality of life. After a major operation, the patient needs to recover from the surgery itself, get over the pain and complications, be rehabilitated, and return to normal life before beginning to benefit from the promised improvement in quality of life. TUTE in such a scenario is equally important, especially if the patient is elderly and not expected to live very long in the first place. There have, to date, been no studies whatsoever examining the net balance of quality-of-life benefits and disadvantages when major surgery is carried out in the elderly, but this will soon change.