The UK magazine Update recently carried a story on how health information provision is being coordinated in the UK National Health Service. It contains a striking quote about the value of information - the 'application of what we know' - from Dr Muir Grey who has been very involved in this area for several years.

The visionary’s view of information as a key resource in clinical practice was set out in a recent consultation document, Best Current Evidence Strategy [.doc]. In it Muir (as he is known in health information circles) makes a challenging proposition: ‘The application of what we know will have a bigger impact on health and disease than any single drug or technology likely to be introduced in the next decade.’
According to Muir, ‘by putting knowledge into practice, we can prevent or minimise the seven universal problems of healthcare’. In his model there are three strands to working with knowledge: the Best Current Evidence Service (explored in detail in the report); the NLH [National Library for Health], which he sees as being charged with ‘organising and mobilising the evidence’; and the National Knowledge Infrastructure, which contains technical standards, tools and services. [CILIP | New roles for information professionals in the NHS - Stephen Singleton]

At the same time, Adam Bosworth of Google has been writing about how patients need better access to organized health information, better ways of controlling their own health information, and better ways of connecting to others with similar health interests. He highlights some of the ways in which inefficiencies in information access and use complicate diagnosis, treatment and choice of care-giver and goes on to say:

These are some of the health-related problems we're thinking through at Google. We don't have any products or services to announce yet and may not for quite some time, but we thought we'd share a bit about the problems we're interested in helping out on even before we introduce solutions. As we explore these problems and continue to work on them, we hope to share more about our efforts along the way. [Official Google Blog: Health care information matters]
In each case we see the value of 'evidence' and the importance of mobilizing it more effectively by practitioners and patients alike.

Comments: 2

Dec 04, 2006
Ben Toth

We met the team at Google Health recently and had a very fruitful discussion. We're aiming to expose the NLH metadata to Google (and to anyone else who can use it)

Dec 06, 2006
Chris Rusbridge

I was at a Research Information Nework seminar on "Getting the most out of data" yesterday. During Q&A, Peter Dukes from MRC (which is undertaking a thoughtful and measured approach to data curation, see MRC’s data sharing and preservation initiative) relayed a concern raised by some of medical researchers they had spoken with: that too much information in a comparatively raw form will allow "inappropriate data mining". This could result in wrong conclusions being reached, potentially leading to harmful publicity and even loss of trust by members of the cohorts they are studying, with significant long term damage to research (not to mention health effects if anyone acted on mis-guided conclusions).

This does feel uncomfortably paternalistic ("don't confuse the children"), and I'm afraid I unfairly accused him of that. On the other hand, scientists spend years learning sophisticated techniques to try to distinguish coincidence from causality, so there is plenty of room for mis-interpretation. I think sufficient controls can be put around this to (at least) warn of the risks of mis-interpretation, but I would certainly want to continue nevertheless.

So is "a little knowledge ... a dangerous thing"?