Not so NICE?


I went to a pharmacy in Switzerland to get some codeine for a friend who had bruised his ribs skiing. The pharmacist was incredulous that I should want to use codeine because it was not, at least there, a painkiller.

In Switzerland, or perhaps that Canton, it is used for cough suppression. It is of course a pain killer but the incident serves as a reminder that what we may think of as medical facts are sometimes more an amalgam of data opinion and culture.

Reports suggest that NICE directives are soon to be mandatory, that practices and trusts will not be allowed to deviate from NICE directives under threat of punishment.

Unless the NICE approved standard of evidence is present, your doctor will not be able to offer you a particular treatment. It seems reasonable but is it?

The culture of contemporary modern medicine is that nothing can be used without evidence which on the face of it is of course quite correct. What is not correct is the assumption by almost everybody that evidence must come in the form of a randomised controlled trial, a RCT.

Indeed it has become a mantra now to chant, "where's the evidence?" whenever anything is suggested or proposed. It is as though years of training, exams and hard work count for nothing, that accumulated knowledge, intuition and careful clinical practice are of no value when compared to a trial.

A paper published in the BMJ a few years ago came to the, tongue in cheek, conclusion that "parachutes reduce the risk of injury after gravitational challenge, but their effectiveness has not been proven with randomised clinical trials", a self evident proposition but the emphasis on only one kind of evidence can hamper innovation, leading to belief in only one kind of truth.

Some years ago I was looking after a patient on an intensive care unit in London. We were struggling and after discussion with some colleagues we decided to use a drug called Prostacyclin. I looked up the relevant papers, and worked out the dose.

One of the staff was appalled that I could not produce a written protocol for it’s use, that when asked about the dose I said I was not sure but that after consulting and reading published work we would start with this dose, even without the evidence.

Now we are obliged to worship at the altar of evidence-based medicine, the high priest of which is NICE; the National Institute for Health and Clinical Excellence or as I once nicknamed it, the National Institute for Controlling Expenditure. There is nothing wrong with evidence or controlling expenditure.

The problem for some patients and some doctors is that NICE is not always that nice, the evidence is not always what it seems and the rigid insistence on only considering randomised clinical trials may in some cases mean that some patients may die who perhaps might have survived or that you will not be able to access potentially beneficial treatments.

To judge from many NHS publications you would think that that there is no disagreement about what constitutes best evidence, but this is far from the case.

NICE whilst being NICE is not perfect though. I am an advisor to NICE on interventional techniques for pain management (though if this article is published I probably won’t be for long).

Back pain affects 70% of the population at one point in their lives and costs millions of pounds a year in health resources. Treatments vary from surgery to drugs to manipulation. NICE produced a draft guideline for the management of back pain that is of less than one year’s duration. The Clinical Guidelines were published in May 2009.

In my speciality of pain medicine we run pain clinics which are often multidisciplinary but which may involve spinal injections. Now spinal injections are controversial. The profession is divided, there are those who say they are effective and those who say they are ineffective, a similar diversity of opinion exists for spinal surgery as well. Not surprisingly each camp has its own evidence base.

As a doctor myself I know that when opinions are divided like that, the truth is probably in the middle somewhere. Some patients will benefit greatly, some not at all. There is nothing wrong with that approach, sometimes treatments do not work and there may be multiple reasons for this, indeed, it may be near impossible to get to the supposed gold standard of evidence, the randomised trial.

Sir Michael Rawlings, the chair of NICE admitted as such in a recent oration to the Royal College of Physicians. The working party for NICE, which did not include a pain physician (I was not asked despite being on their books) came up with a different view.

The committee decided to look at one, only one, published study of 130 patients, which most people in the field would agree was not particularly well designed . On the basis of this they decided that spinal injections should not be offered to patients. They did however decide that manipulation and acupuncture should be offered, as should an opinion for spinal surgery. It will come as no surprise that the committee contained a spinal surgeon a number of manipulation practitioners and some complimentary therapists. It is not that surgery and manipulation are no help for back pain, they can be. However the evidence for both is not great and not much better (unless you have decided in advance what evidence you will accept) than that for some spinal injections.

The public and the profession need to guard themselves about accepting evidence, especially when cloaked in the respectability of NICE, which may be flawed. In 2007 I suggested an entirely plausible additional treatment for my rare and serious cancer to a leading oncologist in the UK. The reply was, "you cannot do that, we won't know if the chemotherapy has worked". Luckily I could take the treatment because I persuaded another doctor to prescribe it for me and he did. How do we decide what treatments to offer people and when? I have an aggressive, invariably fatal cancer of the lung called a mesothelioma. Treatments are not particularly effective and most patients diagnosed are dead within 18 months. There is a drug called Zolidinic Acid, it is in a class of drugs called bisphosphonates. It is used to treat cancers that have spread to the bone and is also used in patents with osteoporosis.

Over a year ago a paper was published documenting the effect of this drug on mesothelioma in mice, not on secondary mesothelioma , which had spread to bone, but on the primary tumour. This drug has been around for a while treating osteoporosis patients and its adverse effect profile is well documented and within acceptable limits.

I have been having the drug regularly. I am lucky though. I went to my local oncologist with a copy of the paper showing its effects in mice and asked to be given the drug, despite the lack of evidence in humans and despite the potential side effects. Why? Because with a disease like mine I did not want to wait for a few years till a clinical trial was published. That is what I call real informed patient choice. I doubt whether any other patients with mesothelioma will be offered or given the drug until they have secondary tumours in bone because there is "no evidence".

There is evidence though; the paper in mice, the use in cancers which have spread and the documented safety profile. To my mind that is enough when the alternative is a high probability of death.

To add to that evidence I have corresponded with the author of the paper, a professor in the USA. He feels that this drug has an effect on mesothelioma; recently he has found that it has an effect on human mesothelioma cells in the lab. Evidence enough for me! I am lucky- I am a senior doctor, understand the science and am able to persuade my colleagues if necessary that I am willing to try things without so called perfect evidence.

Why should this be controversial, when we fund homeopathic remedies on the NHS for which there is neither plausible scientific rationale nor evidence of any significant benefit?

The desire for high quality evidence, laudable in itself, has metamorphosed. If an RCT exists for a treatment, and demonstrates the treatment is highly effective then this may be true, provided the study is of high quality. If the study is poor or there is no study then we can only conclude that we do not have an effective RCT, not that the treatment does not work.

What is happening is that treatments are rationed on the basis of the lack of an RCT. It is worth considering whether in situations where death is almost certain whether it is ethically right to not offer treatments, which have some scientific plausibility. If patients can really choose then why can they not choose something un proven? It is deeply paternalistic to assume that "they" cannot make such choices?

There is another problem: in some studies in cancer drugs a number of patients show benefit but not enough to prove statistically that the majority do. The effectiveness of the drug is thus unproven – but it is only unproven for the population not for the individuals who benefit. There is an old adage in medicine that if you want the drug to work, be in the original trial – you might be a responder after all but if the drug does not cross the threshold only by being in the trial can you benefit, if the trial fails the treatment "does not work".

There is an inherent conflict between the desire to treat an individual and running a clinical trial for a population. In cancer or life threatening disease perhaps we need to lower the bar of proof – after all we have set the bar in the first place.

Patients and doctors also need to be aware that the methodology which is used to determine the "truth of trials" has been criticised in many academic papers from a statistical and philosophical position. It is not a universal truth, one highly respected journal stated in 2003 "by the rules of EBM itself , no definitive evidence whatsoever has accumulated over 15 years of research and debate on EBM to show that 'medicine by EBM' is superior to 'medicine as usual'".

Without the imprimatur from NICE, the Primary Care Trust or the doctor may say no to a given treatment. Why? Because there is no randomised clinical trial, which has become a trump card better than the Ace of Spades. If NICE says no then primary care trusts are under no obligation to provide the drug on the NHS.

A treatment that may help severely ill cancer sufferers may be refused. The ace of spades trumps all, and as in "Apocalypse Now" is left on your body with the words "No RCT…". You are in effect sacrificed on the altar of evidence-based medicine.

That is what is called bedside rationing. Clearly we need to allocate resources with a limited budget (rationing by any other name). There are however only 1500 mesothelioma patients a year, hardly an unlimited drain on resources, many more with back pain but spinal injections are a lot cheaper than surgery!

It is all very well being NICE but NICE’s decisions however are only as good as the evidence it obtains and from whom – as Tom Lehrer once said, "life is like a sewer, what you get out of it depends upon what you put in". We cannot apply the same rules across the board from in growing toenails to cancers, it is not morally correct.

It is a fine balance; the interests of the individual patient and the desire to find the most effective treatment for all but our desire for proof can go too far. People suffering with back pain should not be denied treatment because a committee has decided what evidence to look at and I have no intention of being sacrificed on the altar of evidence based medicine whilst waiting for a clinical trial for my cancer.

This article was written by Dr Andrew D Lawson FRCA FANZCA MSc, Medical Ethicist and Consultant in Pain Medicine and is reproduced with his direct permission.

Useful info:

News Archive

Get In Touch