Dr Mark Charny discusses what you need to consider before developing a local guideline, in the sixth article in this series

A useful principle is never develop your own guideline unless there really is no alternative. Developing a guideline takes a lot of time and skill, and if it is a subject worth considering, someone else has probably already considered it.

So, before you embark on developing a local guideline:

  • Make sure you know about relevant guidelines that have already been developed1
  • Assess the quality of any relevant guidelines you find2

If there are no existing guidelines, or the existing guidelines are not of adequate quality, you should take care before committing yourself. Ask yourself:

  • Are there no (good) existing guidelines because the subject is difficult to deal with?
  • Does the poor quality of any existing guideline reflect the poor quality of the evidence itself – and will you have the same problem?
  • Is the issue so pressing that it is worth taking on the task of developing your own guideline?
  • If a guideline is important, and nothing suitable is available, can you interest a national organisation to take on the work, or support you in developing something that will be useful to others?

The remainder of this article assumes that you have been through these questions.

To develop a good guideline, you need to access two different types of skills: assessing evidence and turning the conclusions of studies into useful recommendations to guide clinical practice.

Assessing evidence requires expertise in epidemiology, statistics, health services research, and information retrieval. Producing recommendations draws on the expertise of clinical opinion leaders, and good writing to express the evidence nd the recommendations in language that is suitable for a typical clinician working in a typical setting.

When identifying and assessing the evidence, and turning it into recommendations, you need to involve all stakeholders. Stakeholders are groups or individuals whose clinical activities are likely to be covered in the proposed guideline. You will also need to consider involving relevant non-clinical interests, such as local PCGs and trusts.

Clearly the group should involve all professions who provide care covered by the guideline. Apart from ownership, this also helps to ensure that the recommendations are balanced.

Orthopaedic surgeons' views about backache, or gastroenterologists' views about diarrhoea, inevitably have a different focus from GPs' views on the same subjects. And guidelines involving referrals to specialists or shared care between GPs and consultants cannot sensibly be drawn up by specialists or GPs alone.

Earlier articles1,2 have outlined the approach to searching for evidence. You will probably benefit from involving a specialist in information retrieval: you might approach your local postgraduate medical committee library, a national professional organisation (such as the Royal College of General Practitioners) or another national body (such as the National Institute for Clinical Excellence [NICE] or the NHS Centre for Reviews and Dissemination).

Mhe evidence then needs to be subjected to systematic review. This involves:

  • Collecting all the available evidence
  • Assessing whether the evidence is relevant to the issue of concern to you
  • Judging the extent to which the evidence is – or may be – biased
  • Summarising the conclusions that the evidence supports.

You then need to package the conclusions in a way that makes it easy for an average member of your target audience to understand and remember.

It is more likely that any work you do to develop guidelines will involve adapting guidelines you have obtained from elsewhere to meet local needs and circumstances. In this way you can benefit from the skilled work of others in systematically reviewing the evidence, and concentrate on applying this to your patch.

There are four reasons why a guideline that has passed your 'assessment tests' might need to be modified:

1. To develop a sense of ownership: Strictly, this is not a justification for modifying the guideline, but offering those involved an opportunity to talk through a guideline and express it in terms that seem familiar and feel comfortable is an important part of implementation (and the subject of the next article in this series).

2. To adapt the guideline to local circumstances: For example, a guideline on the management of acute myocardial infarction diagnosed at home might suggest that:

  • The GP should give appropriate analgesia
  • The patient should be transferred immediately by ambulance to the nearest A&E department
  • Thrombolytic therapy should be given within a maximum of 15 minutes after the time of the patient's arrival in A&E.

These recommendations may suit an urban area, but may not be appropriate in a sparsely populated rural area. In this case, it may be better for the GP to administer thrombolytic therapy before transfer of the patient to a district general hospital.

3. To develop more precise specifications: Two of the questions for assessing a guideline suggested in the previous article were whether the statements about the patients to which it applied and the recommendations it made were precise and unambiguous. An otherwise acceptable guideline may benefit from further refinement in this respect.

For example, a guideline on the use of skull X-rays after head injury may suggest that these apply when there is:

  • Discharge of blood from the nose or ear
  • Any history of unconsciousness since injury
  • Focal neurological signs or symptoms
  • Suspected skull penetration or a foreign body.

You might wish to clarify these conditions by:

  • Ruling out blood from the nose or ear when pressure on a bleeding cut stems the flow, in the absence of neurological signs
  • Ruling out a history of momentary memory loss, in the absence of focal neurological signs; and
  • Listing focal neurological signs and symptoms.

4. Choosing options: For example, a guideline on prophylaxis of venous thromboembolism may suggest that any one of the following may produce a beneficial effect:

  • Subcutaneous low molecular weight heparin
  • Adjusted dose subcutaneous heparin
  • Adjusted dose warfarin (international normalised ratio 2.0-3.0)
  • Intermittent pneumatic compression

If the facilities for intermittent pneumatic compression don't exist locally, they can be eliminated from the guideline. And by local agreement, clinicians may standardise on one of the first three options to simplify protocols and reduce pharmacy costs.

Don't take on lightly the work of developing your own guidelines from scratch without first making sure that a suitable guideline does not exist and being certain that the problem is so significant that the effort is worthwhile.

Even good guidelines from elsewhere may need to be adapted. But beware: major modifications may create a new guideline that is not supported by the evidence underpinning the original one.


  1. Charny M. How to find out about existing guidelines. Guidelines in Practice 1999; 2(May): 58-61.
  2. Charny M. Is this guideline right for your patients? Guidelines in Practice 1999; 2(June): 49-51.

Guidelines in Practice, July 1999, Volume 2
© 1999 MGP Ltd
further information | subscribe