Adrian Midgley
One of our requirements and possibilities which has not yet been realised is to provide a knowledge service to every doctor as they work. Committing ourselves to bring knowledge, including evidence for clinical interventions, to the point where it is applied to the patient means that many organisations will wish to influence, control, shade and alter that knowledge. Parts of the NHS administration are included in these confounders, and one must maintain a healthy scepticism based on the known motives of the author.
Providing a knowledge service should be approached in layers. Medline is a credible source, though not free from biases, but it is inappropriate to go to Medline for most individual clinical problems. In order to see clearly one may stand on the shoulders of the giants responsible for the Cochrane collaboration. Even here the full Cochrane database is a bit daunting for a morning surgery, and access to the abstracts, via the World Wide Web is more appropriate. While the upper echelon of the peer reviewed journals are reliable sources of original work, the secondary journals which select, review, comment upon and amplify are again more useful from day to day. "Best Evidence" is a CD which should be added to every knowledge base, even where the Cochrane discs are not.
The move of print journals onto the Internet allows intelligent use of the various media, to complement each other. A review article or tutorial which cites original papers may be expected to give links allowing the reader who accesses its Web version to acquire first the summary and then the entire text, drilling down further if need be to the unpublished additional data and discussions of the original authors.
The Web accentuates the problem of ease of access to rubbish. Unlike a proliferation of heavily sponsored paper publications it also contains the antidote.
Evidence placed on the Web which is credible will point to independent sources giving a similar account. The same has always applied to print journals, but references take more effort to consult.
The evidence GPs may consult is the same that their patients may consult directly. This has always been possible in principle but electronic media as well as societies' mood of access to information have produced so large a quantitative change recently that it qualifies as qualitative. The question of whose evidence to trust is one which we must apply for and on behalf of our patients, by pointing to good sources of evidence, and encouraging accurate evaluation of the bad.
A final challenge is to avoid our own evidence being tainted by financial pressures. Fund-holding raised suspicions and unified cash flows in PCGs will at best place us all on the same tier. GPs must distinguish between presenting the evidence for a treatment and advising the patient that they may have it if they persuade their fellow citizens to pay more tax , which is ethical, and shading the evidence to persuade the patient that the treatment is not indicated, which is unethical.
Avoiding the information flow to our desktops being controlled in the hands of a single agency or exclusive network is one contribution to this.