Q Some of the new guidelines for clinicians, such as the National Service Framework (NSF) for coronary heart disease, seem to be including audit strategies within guidelines for clinical practice. To what extent should the processes of clinical management of the individual patient's problems and audit be combined in guideline design?
A It is now accepted that all guidelines should contain key audit measures so that those responsible for implementing the guideline can know whether their efforts have been successful.
To influence practice successfully, guidelines should concentrate on key clinical issues rather than rehearse everything that is known about a condition. Measures that accurately reflect these key issues should be clearly stated, and if the subject is worth the work of a guideline, these should be recorded as a routine part of normal clinical care.
Once these have been recorded, it should be easy to compare findings from a review of the records against an agreed standard (theoretically best practice, best observed practice, median practice, colleagues in the same practice etc.).
A well-designed guideline will help clinicians to make evidence-based decisions to ensure high standards of patient care, encourage records that help to communicate key facts to other professionals involved in care at the same time or at a future date, support audit to help individual clinicians determine whether there is still room for improvement, reduce litigation, and provide assurance to the NHS that quality of care is being maintained.
The NSF for heart disease goes some way towards this, but is very detailed and incorporates some measures that may owe more to the need to be seen to be complete than a realistic assessment of how much information the average person can take in and do something about.
Q We have been using the excellent SIGN (Scottish Intercollegiate Guidelines Network) guidelines in our practice for some time. They cover some areas, such as diabetes, which have not yet been addressed by the NSF. When the NSF is published, will it still be acceptable for us to work to our current system or must we adopt the NSF recommendations?
A SIGN guidelines have a well deserved reputation for excellence. The methodology for producing them has been refined through experience over several years.
They are based on thorough work, the type and strength of evidence on which the recommendations are based are very clear, and they take account of the environment in which they will be applied. This makes them robust and user-friendly, and they rightly command wide support.
The NSF for diabetes is likely to cover every aspect of diabetes care rather than focusing on selected issues. It will also include material about service configuration as well as clinical recommendations. I would be surprised if the SIGN guidelines and the NSF for diabetes were to differ significantly on the points covered by SIGN.
However, the way the NSF is constructed will almost inevitably lead to differences in emphasis, and perhaps in some of the measures used, and politicians will want to see the NHS perform in terms laid out in the NSF.
Q I have sometimes wasted precious time hunting unsuccessfully for guidelines on some clinical topics. Is there a definitive list of guidelines available or a list of topics for which guidelines are currently unavailable because of lack of evidence or consensus?
A Unfortunately, no comprehensive database of guidelines in the UK appears to exist. Although a UK national database has been discussed on more than one occasion over the past few years and there is general agreement that it would be valuable, no definite action appears to have been taken as yet. NICE was expected to take the lead on this, but has been busy dealing with higher profile issues.
One problem with any national database is deciding which guidelines are eligible for inclusion. A database confined to guidelines that have been endorsed nationally would be very small, but one which included all guidelines that had been produced locally, regardless of their quality, might be unhelpful.
Even more difficult, and therefore a more distant prospect, is a list of guidelines that are currently unavailable because of lack of evidence or consensus. It is always more difficult to prove a negative than a positive.
There are, however, some useful things you can do:
- A telephone call to the relevant professional body (such as a Royal College) may help: most such organisations have an individual or a unit who specialises in clinical audit and/or guidelines and/or effectiveness.
- Try contacting a local specialist, or a colleague with a special interest.
- A MEDLINE search will often turn up a reference to a guideline produced by a national body, even if the guideline itself has not been published in a peer-reviewed journal.
- The Healthcare Evaluation Unit at St George's Medical School (which has developed a widely used guidelines appraisal instrument) may be able to advise you.
- The National Institutes of Health in the US are developing the National Guidelines Clearinghouse: you may find their website (http://www.guidelines.gov/) useful.