Blog :: All

Evidence based medicine has become the accepted modus operandi within medicine. And why not? The proposition that all medical decisions should be based on evidence is a no-brainer. The danger is in the definition of "evidence" that is at the heart of evidence based medicine (EBM).

EBM values evidence based upon good randomised control trials (RCTs), and especially systematic reviews of good randomised control trials. There's nothing wrong with this, but there are some caveats. The best known is the bias introduced by the non-reporting of negative results. The international community has made strenuous efforts to reduce this bias, but there are other remaining issues.

1. Systematic reviews produce an averaging effect, which may disguise the impact in sections of the populations, eg ethnic minorities.

2. I will call the second caveat "The feet are the wrong size for their shoes" in deference to the late great Douglas Adams. To establish a robust trial you need to exclude confounding factors and this means excluding those people who might introduce them. This generally means old people (who may have a range of other conditions) children ( whose systems may behave in different ways) and women who might be pregnant (for ethical and clinical reasons). These are the people whose feet are the wrong size for their shoes! Trials therefore often feature a strong representation of men aged 18-50. This group are not nearly so well represented in a typical patient population.

3. The third caveat may be characterised as "Making the feet fit even though they are the wrong size for their shoes". Because of the high regard in which RCTs are held, they are sometimes applied to problems to which they are wholly unsuited. This may be because of problems in defining a clear hypothesis for testing, or problems in defining the control and active groups for a trial. An example may be found at Lattimer V, George S, Thompson F, Thomas E, Mullee M, Turnbull J, et al. Safety and effectiveness of nurse telephone consultation in out of hours primary care: randomised controlled trial. BMJ 1998; 317: 1054-1059. available on-line at http://bmj.com. There are a range of problems for which RCTs are simply not amenable.

4. The final caveat relates to the undermining of other forms of knowledge and evidence. Tacit or implicit knowledge is highly valued in many spheres. It appears to have been undermined in the UK in medicine following a series of scandals in the 1990s which led to a public perception of paternalism and complacency.

So, yes, decision should be evidence-based, but the scope of that evidence may be wider than that deployed within orthodox EBM.

Informed consent is required in research, in treatment and for the storage use, disclosure or sharing of personal information. For this discussion, we shall consider consent to treatment as part of the delivery of health care. The concept came to the fore in the UK at a series of high profile inquiries into adverse events at Bristol and Liverpool. Those inquiries described situations where the health care professionals were deemed to have acted paternalistically and failed either to obtain valid consent or provide sufficient information to enable patients to make informed decisions. This was responsible for a shift in public opinion and reduced trust in doctors amongst some patients, although the longevity and extent of this shift remains to be proven.

In guidance to patients, published in response to these inquiries, the Department of Health states that:

“It does not matter so much how you show your consent: whether you sign or say you agree. What is important is that your consent is genuine or valid.
That means:
1. you must be able to give your consent
2. you must be given enough information to enable you to make a decision
3. you must be acting under your own free will and not, say, under the strong influence of another person”

The influence of this guidance is seen in every UK hospital, where strenuous efforts are made to ensure patient consent. However, it still appears problematic. Consider a patient journey of my own to illustrate some of the issues. A few years ago, I was afflicted with kidney stones. Before arriving at hospital, I demonstrated my incompetence to make informed decisions. In spite of being in extreme pain, I DROVE myself to hospital placing myself and other road users at risk. On arrival, my first priority was pain relief. My options were explained to me. I was unable in reality completely unable to reach an informed decision. My actual thoughts were,
“I don’t care. Just stop the pain. If you offered to stop the pain at the moment by chopping off my head with an axe, I would not only say “yes!”, I would grab the axe with both hands and chop my own head off”

A little later, morphine having been administered, I was asked if I would like more. Does the junkie refuse their next fix? No, and neither did I, although I had absolutely no idea if I needed it.

There can be other unintended side effects. Later in the same episode of care, I found myself awaiting surgery and was visited by a doctor seeking my consent for the procedure I was about to undergo. The first problem is that as in many cases, I had no real choice to make, there was no viable alternative. In such cases, too much information can harm the patient. If I have to undergo a procedure, is it helpful for me to know that the operation is low risk, but the required general anaesthetic is much more likely to kill me?, And if it is, is it helpful to me to reinforce this message with a second conversation shortly before entering the operating theatre? In this case, it seems to have more to do with watching the hospital’s back than looking after my welfare as a patient.

Informed or valid consent is a noble ideal, but appears a noble myth nevertheless.

The media often refers to the postcode lottery of access to health care delivery. The Guardian defines it as follows:
“The postcode lottery is shorthand for seemingly random countrywide variations in the provision and quality of public services - the huge gap between the best and the rest. Where you live defines the standard of services you can expect. So if you live in the "wrong" area, and, in extreme cases, on the "wrong" side of a road, you may get a poorer service than your neighbour or you may not get the service at all and have to pay for it privately. The postcode lottery is a big issue in the NHS, where the gap between the rhetoric of a comprehensive and universal "national" service and the reality is increasingly stretched.”

Even the Department of Health adopted the phrase. From the NHS Cancer Plan:
“There are widespread geographical inequalities in the quality and type of treatment patients receive, because of shortages of specialist staff, fragmentation of care, inadequate access to surgical facilities, a postcode lottery on prescribing and insufficient radiotherapy facilities.”

However, successive Governments have emphasised that local services must meet local priorities, from fund holding introduced by the Conservatives in 1990 to primary care groups and trusts introduced by Labour in 1997, and more recently, practice based commissioning and foundation hospitals. Variations in local services in attempt to meet local needs lead directly to what is known as the postcode lottery.

The postcode lottery arises when local capacity fails to meet local demands and patients lose out on the services or resources they require. Generally the media portrays this as a national issue, an attack on the rhetoric of a comprehensive and universal "national" service as the Guardian describes it, and national solutions are required and demanded by the media. However, a uniform national solution will not meet the variations in local demand satisfactorily and lead to either unsustainable surpluses in service level provided in those areas where demand is lower, or a failure to meet demand in high demand areas.

Perversely, the only solution to the “postcode lottery” is more and better local control and variation. If at a local level, the local health management can match local resources and services to demand, then patients will get the levels of service they need and expect. Unfortunately, it is the nature of Governments when criticised to be more controlling, more centralising, and this is evidenced through the wealth of targets, national service frameworks and plans that we now have in the NHS. Remarkably, in spite of this, the NHS continues to perform pretty well in most local areas.