Wednesday, November 11, 2009

Sharing the Lessons of NHS Funded Research

I attended a conference at the Royal College of Physicians yesterday which reviewed progress with NHS funded research through the National Institute of Health Research (NIHR) and in particular its Health Technology Assessment (HTA) arm.

I am one of 3 lay members of the HTA's Pharmaceutical Panel - it is 1 of 6 such panels that filter out initial proposals for funding before they are considered by a Prioritisation Group and if approved by a Commissioning Board.

The process has been described as similar to the Grand National with a number of fences to be cleared before funding will be allocated. Using this analogy my role is therefore one of the twigs that make up the first hurdle. Also on the same panel are some very eminent practitioners with specialist clinical and pharmaceutical knowledge. One of these is Ben Goldacre - of Bad Science fame and another is Carl Henghan of the Centre for Evidence-Based Medicine, Dept of Primary Health Care, University of Oxford. Both are frequent commentators on Radio and TV on aspects of evidence based medicine.

The conference was chaired by Richard Smith, former editor of the BMJ and some 25 years or so ago an earlier TV medical pundit who sought to explain science to a lay audience.

He had just returned from a 5 day conference in Germany exploring how best to promote health literacy. Ben Goldacre was there as was Gerd Gigerenzer, whose book "Reckoning with Risk" I have almost finished. The book describes how it is almost impossible to achieve informed consent without understanding clearly the natural frequencies of possible outcomes.

Giggerenzer describes the fallacy of relying solely on percentages to express risk and describes workshops where clinicians have demonstrated themselves to be as innumerate as the general population and felt uncomfortable being asked to work out real risk. He describes how, once problems are restated in terms of natural frequencies- i.e. out of every 10,000 people (with a particular set of characteristics) how many would for example test positive when screened but not have a problem, or test negative but have the problem (false positives and false negatives) ,- the clinicians feel more comfortable and better able to share the risk with their patients. He also uncovers the hidden costs of false positive and false negative results - for example false positive HIV tests that have resulted in suicides. He makes the telling point that sometimes more people can die WITH a condition than OF it. For example prostate cancer in men. This is something that the "worried well" need to be more aware of if they are not going to harm themselves at great cost while benefiting only the growing band of today's equivalent of "snake oil" salesmen.

The first speaker at the conference was the Chief Medical Officer Sir Liam Donaldson. I first met him when he was The Regional Medical officer of the Northern Region and I was working at the Nuffield Institute Leeds University just after leaving the NHS after 20 years in a variety of management roles. I was a mentor to a trainee who was attached to Liam's department. I had "worked" with his predecessor as Chief Medical Officer Sir Kenneth Calman in the past. We had both spoken to clinicians in Scotland on clinical engagement with management . I had just completed a report for the then NHS Management Board based on some quick and dirty research and was able to share the outcome with those present. Ken who was at the time CMO Scotland while still practicing as a Cardiothoracic surgeon shortly after went on to be the CMO for England and Wales, succeeding Sir Donald Acheson (who was CMO at the time that I was seconded to the Department of Health). SIr Liam gets a bad press from clinical bloggers who see him as a tame patsy for the current government and their peddling of ersatz markets and false "choice". They are also upset with him over hte dumbing down of medical education and the rise of "noctors" (not doctors) , "Quacktitioners" (nurse practitioners) and similar attempts to dilute the skill levels of those providing clinical care more cheaply to the masses while ensuring that the political class and their hangers on continue to have access to the best that is available free of charge because of who they are and what they know about manipulating the system to their advantage. His recent inability to answer a straight question about the relationship between government and scientific advisers in light of the sacking by the Home Secretary of Professor Nutt underlined the difficulty he faces speaking with any credibility on such issues. His predecessor Ken Calman had no such difficulty - perhaps because he was what GPs and others would have regarded as a "real doctor" and not one of the band of "failed doctors" that they feel find their way into "Public Health". Ken Calman retained that sense of Independence of thought, speech and deed - or at least was able to get away with that impression. Sir Liam struggles in comparison.

However yesterday Liam was on firmer ground. He was returning to an area that he had made significant contributions to before he got embroiled first in General Management (as Chief Executive of the Northern Region) and later in national politics (that's the small "p" not the large "P" as in Party Politics) once elevated to the NHS Management Board as CMO.

He dug out notes he had written 10 to 15 years before and which were still relevant today. These included concerns he had about how best to disseminate the results of properly conducted research. He mentioned the concept coined by Rogers of the Diffusion of Innovations and remembered scribbling down the acronym NIIMBY (or as he pronounced it Nye Imby) - meaning Not Invented in My Back Yard. Others used have used the phrase "Not Invented Here" to describe this reluctance to give credence to evidence from elsewhere.

Sir Liam felt this barrier to dissemination might be based on feelings of humiliation and inadequacy. My own notes of his talk show 2 triangles at this point. The first with the apex at the top and a number of bands is Maslow's Hierarchy of Needs with "self actualisation" at the top. My second triangle was the inverse of this. The same bands but with self actualisation at the bottom and the "base" of the triangle at the top. Local managers and clinicians have increasingly been denied the space to experiment and pilot while at the same time the Dept of Health has moved away from disseminating good practice developed locally and concentrated instead on micro management. The long term consequences of this are spelt out in sections of Anthony Seldon's latest book "Trust: How we lost it and how to get it back"

The last 25 years in public service have seen this inversion of the hierarchy of needs magnified large with top down targets being driven on by bullying tactics based on the use of Dread, Envy, Fear, Greed and Hate. These motivational triggers have replaced the Altruism, Benevolence and Charity that were much more evident 25 years ago. Forget about "self actualisation" - survival is all that matters now - but don't breath a word of that to the press - remember what happens to whistle blowers! Perhaps these are areas ripe for the Chief Medical Officer to demonstrate his leadership qualities and rise to the challenge.

Nevertheless Sir Liam identified 6 ongoing concerns that needed to be addressed if the lessons of NHS Research were to be better diffused in the future.

The last 4 of these I managed to note down as:

3. Models of care are difficult to spread
4. NHS Leaders and managers are not evidence literate
5. Too much recent system change without piloting or evaluation
6 Clinical Behaviour Change is the hardest to achieve and is why technical changes appeal more because they offer more hope of success.

Sir Liam stressed his 4th point and I was reminded of evidence of this from my own past:

1) The time I was told by a Chief Executive that it would be a waste of everyone's time if I was to proceed to the panel interview after the preliminary series of exercises as I had demonstrated that I was "too data driven" for him and his team to feel comfortable with as his future deputy.

2) My attempt to set up an "Informed Management Unit" at the Nuffield Institute instead of the Information Management Unit that the NHS then seemed to want. It was difficult to "market" this concept when complexity of IT systems was in reality the greatest excuse for decision avoidance by managers and politicians at all levels and judging by the mess and waste of the latest National I.T. strategy it still is!

3) My failed attempt in 1973 when as a junior administrator at Pontefract General Infirmary I wanted to analyse hospital deaths and plot against staff rota's to see if there were any anomalies. (I had been sensitised to this by the response of staff who paid cremation certificate fees to junior doctors - known as "Ash Cash" - when I had asked if anyone took advantage of this.) One of the junior doctors at the hospital at the time was Harold Shipman. I was told that this was not something that hospital administrators should concern themselves about and to get back to balancing the petty cash!

4) My analysis of hospital death rates that was the front page story in the Times by Jeremy Laurance in April 1993. Responses from a Regional Medical Officer and a Hospital Chief Executive at the time indicated that they were both unaware of their relatively poor performance and what was more disturbing did not care anyway.

There are many more examples but these were what came to mind at the time.

20 odd years ago I was asked to take the lead nationally on a aspects of Organisational Development when seconded to the Dept of Health. The work was scuppered by the hasty reaction of the Thatcher Government to being embarrassed by clinicians over care being denied to children in A& E. This broke the unwritten compact between politicians and doctors and as a result, the gloves came off and successive governments have pushed through the choice and market agenda ever since. At the time I was asked to look at 2 aspects, (1) To what extent there was agreement on Key Values that underpinned the NHS and (2) What could be done to improve relationships between doctors and managers. I will return to this work in the future in another posting and when I do will edit this post to include a link to the findings and lessons.

Having conducted a survey of a couple of dozen hospitals where it was thought there were good relationships between managers and clinicians I had identified 30 actions that could contribute. What emerged as the top priority was "Agreement on what constitutes Quality".

If I was to repeat the work today I would want to include Gerd Giggerenzer's work and add in "Agreement on how to Reckon With Risk using Natural Frequencies", albeit with a flash card explaining terms with some examples.

Reflecting on Darell Huff's book How to Lie with Statistics, Giggerenzer has a chapter (12) devoted to "How innumeracy can be exploited". It mirrors much of what Ben Goldacre draws attention to in his articles, blog and book about Bad Science.

Perhaps a joint understanding of how innumeracy can be exploited would be a good shared starting point for overcoming barriers locally between researchers, clinicians and managers. Arising out of a series of workshops to achieve this, that would be delivered locally, researchers, clinicians, managers and - lets be really visionary - patient & carer representatives might be able to agree on some top priority evidence based plans for the future attention.

Perhaps this too is an area where Sir Liam could show his leadership qualities in the future and build on his past achievements in public health and evidence based medicine. A start could be made by the HTA insisting on the use of natural frequencies when is own panels consider prevalence rates and probabilities, they could also insist on research findings being expressed in a similar fashion. Monograms for public consumption could be published as a separate series. As was pointed out most members of the public reach for Google and do a search on any diagnosis. They are faced with a plethora of unsubstantiated advice. Perhaps the Dept of Health needs to divert some of its advertising, marketing and consultancy budget to paying Google ( and similar advertising funded search engines) for priority placement of Ads that highlight these public monograms. Again this is something that Sir Liam could take action on.

These are just a few thoughts - from a lay representative (who was able to attend the HTA conference) - on how to better disseminate research findings, engage local managers in the process and provide the worried well with some tools to overcome the exploitation of their own innumeracy.

No comments: