Friday, April 16, 2010

Evidence, Ethics and Responsible Care

A radically innovative “treatment” approach to MS has sparked worldwide interest and huge demand for a surgery that has not been “proven” to bring a benefit for MS sufferers. I will not go into the details of the proposed “treatment” here but rather look at the issues of ethics, evidence and responsibility that are raised by this situation and situations like it.

Evidence

In most (all) areas of human endeavour we do things for reasons. So, if I perform some action or another you could ask “Why?” and could reasonably expect a reply that would give my reasons. Typically those reasons would consist of something like an outcome I wished to achieve and some method for reaching that outcome. “Why did you go to the store?” “To buy curry powder.” “Why did you want curry powder?” “To make curry for dinner.” And so on. Indeed in health care one of the ways that a professional would evaluate a person’s capacity to make health care decisions would be on the basis of the person’s ability to undertake this sort of means-ends reasoning; identifying intelligible goals and appropriate methods of achieving those goals. In everyday life we accept quite a wide range of goals as intelligible and quite a wide range of types as reasons as possible methods of achieving those goals. In health care in general, however, and more specifically in medicine, the goals are more restricted – health or physical outcomes or states, and the reasons given have come to be called “evidence.” “Why did you prescribe that drug?” “Because it increases blood flow.” “Why did you want to increase blood flow?” “Because it will increase oxygen uptake which will ease the patient’s symptoms of shortness of breath, which will allow him to perform more of his everyday activities.” The key question here, however, is what would count as evidence for the claim that this drug, or other form of treatment, will have this outcome. What counts as evidence (or good evidence) that this “X” causes that “Y?”
It is worth reiterating that the standards of evidence in medicine (and indeed the standards of evidence in professional life generally) are quite different from the standards, or indeed what might even be counted as evidence, that might be acceptable in ordinary life. That is because the professional – in this case the physician or other health care professional stands in a special “fiduciary” relationship to the patient (or client.) Professionals – and let’s use the example of physicians, have undergone a specialised training and licensing process which confers on them an expert status. They have been trained in the methods of their profession and they are committed to practising within the professional guidelines and standards of that profession. In medicine the standards of professional practice require that medical decision-making be “evidence based.” A physician – as physician cannot prescribe a drug because he or she thinks the stars are in appropriate alignment, or because he or she had a dream which suggested this drug, or because the drug manufacturer has paid him or her to do so. The physician is bound by standards of good professional practice only to prescribe the drug or other treatment, which, on the evidence, is likely to cause the outcome which is desired.
So far so good. But what counts as evidence of the causal efficacy of a drug or other treatment? This is an area for some debate. The so called “gold standard” for evidence in medicine is a randomised clinical trial performed on a suitably large group of suitably diverse subjects for an appropriate length of time. The reason for randomised clinical trials is that this is the best method we have devised of isolating the causal effects of the drug or other intervention on the outcome we are measuring. The assumption is that randomisation, over sufficiently large numbers will even out the background noise of naturally occurring variation to allow the true causal effect to shine through. The idea is that with a well-designed clinical trial we will be able to say with a certain degree of confidence (typically 95%) that the effect “Y” is caused by the intervention “X.” (More precisely that the effect “Y” is not attributable to chance.) Perhaps we should add a word on confidence. Certainty, the absolute, cannot- be- wrong, kind that can be found in mathematics, is not to be found anywhere in medicine. The best we get are probabilities. Clinical trials are designed so that we can make causal statements with a high degree of probability. A clinical trial is designed so that we can make causal statements with a high degree of confidence that the effects we have observed are very unlikely to have been caused by chance. (Typically only a 1 in 20, 1 in 100 or 1 in a 1000 possibility that the observed effect was caused by chance and not the effect of the drug or intervention under study. )
The other significant component of a clinical trial in medicine is the search for possible side effects. Not only does the drug or intervention have to be effective, but its side effects must be sufficiently slight or unlikely that the anticipated benefit outweighs the possible negative side effects. Again, trials are designed to be able to maximise the chances of identifying possible negative side effects.
So, the good clinician will only prescribe treatments or interventions for which there is good evidence.

Radical Innovations

But what happens when something new comes along? What happens when someone proposes or discovers a radically new way of thinking about – and approaching a disease or condition? In the first instance the demand will be for evidence. What are the reasons to believe that this proposed approach will bring about the effects that are intended? Why should we think this will work? In some cases the initial answer to this question is that the new approach builds on ways that the disease is understood, and works by extending or developing existing approaches. In cases like this the development is incremental, we would still need evidence in order to proceed, but the search for that evidence would typically fit into the research programmes that already exist. What happens when the approach is radically different and requires a new way of thinking about the disease and is treatment. This situation is far more difficult, because on top of the standard demands for evidence there is the added layer of what could be thought of as the sociology and psychology of medical and scientific practice. If you have spent your life and research efforts exploring a disease using one approach you may not be the best person to evaluate a radically new approach which basically undercuts your life’s work. That is not to impute bad faith to anyone – just to recognise that we approach problems from a particular perspective and within a particular framework or paradigm. Those working within that framework or paradigm may well find it extremely difficult to appreciate, understand, accept or evaluate an approach to the problem that employs a different framework or paradigm. On the flip side, those advocating the revolutionary approach may well – at this stage, be doing so without sufficient evidence to make the new paradigm or framework compelling. What plays out is a scientific (and sometimes small “p” political) debate and the search for more and better evidence. But that is the scientific debate. What about the physician – the practitioner who faces demands from her patients for an unproven therapy?

Responsible Practice

Responsible practice requires evidence. In the cases we are discussing, by hypothesis, the evidence is insufficient to warrant a change in the way in which care is standardly provided. That is, there is not a consensus on the standards of good practice that has changed the current approaches. In practical terms this means that the expert bodies in the relevant areas are not recommending a change in practice. The “evidence” for the new approach may be anecdotal or theoretical but is not yet validated or accepted. The good practitioner should monitor the evidence and the debate as it emerges and may well wish to participate in research and clinical trials to seek evidence one way or the other. The responsible practitioner recognises that standards of good care are community decisions – the community being the expert practitioners engaged in the collaborative enterprises that are science and medicine. Does this mean that sometimes individuals will be harmed because of the relatively slow adoption of new treatment approaches (that in fact turn out to be successful) as we wait for the evidence of effectiveness and safety to accumulate? Undoubtedly yes. But the flip side is that if we introduce new approaches too soon we will harm people either by performing treatments and interventions that are simply ineffective, or by causing side effects both known and unknown.

The patient’s perspective

So far we have looked at this issue from the perspective of the medical practitioner and the scientist behind that practice. The perspective of the patient may well be completely different. That is, if the disease I face is sufficiently debilitating it may well make sense to me to try anything that holds any prospect of success. It may well be rational for the drowning man to clutch at straws. If I am going to die I may (quite rationally) wish to try anything I think may save my life. And, if we go back to the point at which we started I am under no compulsion to have “evidence” for my belief or desired. I may wish to try shark’s cartilage, or grapes or I may wish to pray or bathe in a particular pool. I am entitled to have my own beliefs and my own reasons for those beliefs, and those beliefs or reasons do not need to fit within the canons of contemporary science. I may also be quite willing to undergo any risk of known or unknown side effects. And, if the treatment or intervention I seek is under the control of my medical practitioner – which it may well be (for instance if it is a drug or controlled substance or a surgical intervention) I may seek to pressure him or her to provide me with that treatment. What should the patient-focussed , caring practitioner do under those circumstances?
I think this question takes us right back to questions of responsible practice. I believe the physician or other medical practitioner practices within a community of good care and professional standards. I believe that community determines what constitutes good care, professional practice and appropriate types of evidence. Which leaves the practitioner in a really tough spot, having to say no to a desperate patient as he or she clutches at straws.

I realise that this leaves many things unsaid – we have not discussed the ways in which responsible researchers may choose, in effect on ethical grounds, to stop a clinical trial early (either because the results are especially good, or because the side effects appear unacceptably bad). Nor have we discussed the added pressures which arise within a publicly funded health care system. In such a system the medical practitioner has an additional commitment not to waste scarce resources (for instance by pursuing unproven or highly unlikely treatments.) But we will leave those topics for another time.

What do you think?