Dr. Pantoja asks a very important question regarding clinical guidelines.
If I understand his question correctly, at the heart of it he is essentially
asking, "How do we obtain quality clinical guidelines as efficiently as
possible?"
I have several comments:
1) There are two "poles" on the spectrum: a) searching the primary
literature and building clinical guidelines from scratch and b) adapting
another group's work. To end up with valid, relevant, up-to-date guidelines
you will need to do a bit of both, because almost without exception the
guideline you choose to work from will be out of date.
2) For most groups the best way to start is to search for "seed" guidelines.
No matter where the guideline comes from, there is a need to evaluate it,
and like the critical appraisal of a published study, there is a need for a
checklist or tool (see below) to assist in the evaluation of guidelines.
1. Good places to start for evidence-based guidelines:
- New Zealand Guideline Group (I recommend this group as a place to start
because I personally know the quality of their methods which are the methods
I used at Group Health Cooperative, Seattle, based on the rigorous,
explicit, evidence-based principles recommended by David Eddy MD.)
http://www.nzgg.org.nz/tools.cfm
- Agency for Healthcare Research and Quality (and Evidence-base Practice
Center reports) http://www.ahrq.gov
2. Next, to "round up" other guidelines that are "out there":
-The National Guidelines Clearinghouse. www.guidelinesgovindex.asp
(Buyer beware. You need to carefully evaluate guidelines obtained from the
NGC, because the validity and relevance vary greatly.)
-Pubmed www.pubmed.gov
Set "Limits" to guidelines.
-For systematic reviews, use Clinical Evidence
http://www.clinicalevidence.org and Cochrane library
http://www.cochranelibrary.com/enter/
- If you still haven't found a seed guideline, you might then look at the
various EBM sites and start following the links. For example, Dr. Brian
Alper's EBM Web Page: A guide of evidence based and clinically relevant web
sites - is worth looking at: http://www.myhq.com/public/a/l/alper/
3. You will now need to evaluate the guideline for clinical significance,
relevance, "implementability." Further reading and tools are available:
- Rosser WW, Davis D, Gilbart. Assessing Guidelines for Use in Family
Practice. Journal of Family Practice 2001; 50: 969-973.
-Hayward RSA et al. User's Guides to the Medical Literature, VIII: How to
Use Clinical Practice Guidelines, A: Are the recommendations valid? JAMA.
1995; 274: 570-574.
- Wilson MC et al. User's Guides to the Medical Literature, VIII: How to
Use Clinical Practice Guidelines, B: What are the recommendations and will
they help you in caring for your patients? JAMA. 1995; 274: 1630-1632.
-Appraisal Instrument for Clinical Guidelines: St George's Hospital Medical
School. www.sgms.ac.uk/phs/hceu/form.html.
4) Finally, assuming you have found a good guideline, you will still need
to search for and critically appraise the studies published since the
issuance of the guideline and obtain permission to use the seed guideline.
I believe the above steps represent the most efficient way to obtain good
clinical guidelines. I would state a couple of caveats:
- It is worthwhile considering if the clinical area requires a clinical
guideline. It might be better to summarize a recent systematic review in
some cases. Here are some questions to ask before embarking: Is the problem
fixable? Is the improvement (in quality, cost and satisfaction) worth the
effort? Is it too complex an issue for us to predict successful change
practice? Can we measure the change?
- One of the most important issues in creating clinical guidelines is the
process by which the guideline was developed. The guideline documentation
should provide enough information for you to know how the literature search
was done, how the evidence was graded and how the key articles were
selected, what would happen if a group or organization implemented the
guideline (this requires a bit of work on your part using your own data and
the evidence to create projections). At the heart of all of this is the
strength of the evidence. You can have "luminaries" on the guideline
development team, a very "formal" consensus process and still end up with a
bad guideline (There are lots of bad guidelines out there).
I hope this is helpful.
Michael Stuart MD
President, Delfini Group, LLC
www.delfini.org
6831 31st Ave. N.E.
Seattle, WA 98115
206-854-3680
[log in to unmask]
----- Original Message -----
From: "Dr Tomas Pantoja" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Friday, November 29, 2002 5:13 AM
Subject: Quick & dirty guidelines
> Dear list members:
>
> We are a group of Family Doctors in Chile trying to develop Clinical
> Practice Guidelines for common conditions in general practice under the
> umbrella of our scientific society (Chilean Society of Family & General
> Medicine). We have some expertise in methodological issues regarding the
> development of CPG but we (as none of other specialty societies) don't
have
> specific and sufficient resources to allocate to this daunting task. Then,
> we have discussed the possibility to accomplish our objectives through the
> "adaptation" of guidelines developed by others (such as NZGG, NICE, SIGN,
> Clearinghouses,...), with a previous critical appraisal of those
guidelines
> (using instruments such as AGREE), and the use of formal consensus
> techniques such the RAND nominal group technique.
>
> We ask for opinions and thoughs from the list regarding
> does anyone know similar experiences of guidelines development in
developing
> countries?
> does anyone know how this "quick and (not so) dirty" guidelines
development
> process compare with more "formal" processes?
>
> Kind regards
>
> Tomas Pantoja MD MSc
> Assistant Professor
> Family & Community Medicine Programme
> Catholic University of Chile
>
> on behalf of the Chilean Society of Family & General Medicine guidelines
> development group
|