George Kernohan has twice asked for "evidence of the clinical 
effectiveness of the provision of evidence to nurses and PAMs" and 
includes the comment "Sorry to repeat this, but is the lack of 
replies a sign that we have a gap here?

Probably.  Is there any evidence that clinicians should practice EBM?
 
Although it's nice to see that the suggestion that medical students 
are becoming more enthusiastic about EBM, in my (limited) 
experience it still attracts great suspicion from established GPs.  
There's been a fair amount of recent discussion about making journal 
clubs evidence based and the importance of making questions focused.  
One of the risks though is that it becomes harder to ask background 
questions which affect all of clinical practice. An example would be:

"Should I practice EBM?"

or

"Compared to those who do not, do doctors who practice EBM have 
better outcomes for their patients?"

which is what established practitioners may want to ask before they 
get into focused questions.  This isn't easy to answer because the 
range of conditions is so large and the range of possible outcomes 
even more so.  It is a crucial question though.

There was a thread about the evidence for evidence based medicine in 
May (starts 13/5/98 with summary 18/5/98 from Jane Sandall) and a 
number of suggestions were made as to papers which provide such 
evidence.  I've looked at all the ones that I can easily access.  The 
list is below with some comments.  

If I was writing a guideline on the practice of EBM, I think I would 
be scraping to get a 'B' recommendation since none of them directly 
address the question.  Most of what's written about EBM including 
exhortations to practice it are basically enthusiastic 'C' 
recommendations based on its 'obvious' benefit or "compelling 
rationale" (eg JAMA 1192, 268(17), 2420-2425, quote from p2424).

Deciding to make the change from traditional practice to EBM is 
a big step and requires time and commitment.  Where's the 
evidence that EBM non-enthusiasts should invest their limited time 
with consequent opportunity costs?

Is there any better evidence?  Would it be possible to design (and 
get funded) a study to try and answer such a question.  Should 
we make large changes to our global clinical practice without 
compelling evidence of benefit?

Bruce Guthrie,
Department of General Practice,
University of Edinburgh.

1) Porter TF, Varner MW. Using evidence-based medicine to optimize  
cesarean section outcomes.  Clin Obstet Gynecol 1997;40:542-7

No available in my library nor any abstract on Medline, but it is a 
review rather than primary research.  Does it show anything 
different?

2) Balas,E.A.; Boren,S.A.; Hicks,L.L.; Chonko,A.M.; Stephenson,K. 
Effect of linking practice data to published evidence. A randomized 
controlled trial of clinical direct reports. Med Care 1998;36:79-87

Randomises US renal physicians to receive (or not) info about modes 
of dialysis and comparisons of their unit/personal practice with 
other local and national providers.  Intention was to increase CAPD 
rates, and it did seem to.  However, it's not directly about 
practicing EBM - it's about a method of information provision to 
passive physicians - and it doesn't measure a direct patient 
outcome (and is driven by cost containment).

3) Green,M.L.; Ellis,P.J. Impact of an evidence-based medicine 
curriculum based on adult learning theory. J Gen Intern Med 
1997;12:742-50

Not available to me but the abstract doesn't describe effect on any 
actual practice, and no patient outcomes.

 4) Flarey,D.L. Does evidence-based practice make a difference in 
outcome? Semin Nurse Manag 1997;5:160-1.

Not available to me and no abstract on Medline.

5) EFFECT OF CLINICAL GUIDELINES ON MEDICAL-PRACTICE - A
SYSTEMATIC REVIEW OF RIGOROUS EVALUATIONS.   GRIMSHAW_JM, RUSSELL_IT
JN: LANCET, 1993, Vol.342, No.8883, pp.1317-1322

Review of the impact of guidelines on practice.  A proportion show 
effects on practice (although patient outcomes aren't usually 
measured), but all the studies cited as showing effects are pre-1989. 
The guidelines are therefore unlikely to be "evidence based" in the 
sense this is now used.  Again it doesn't address the global 
question and is quite compatible with passive physicians being 
given information from on high rather than the practice of EBM which 
requires an active role.

6) where e-b docs use more ASA following MI's, deaths are reduced
by >20% [Circulation 1995;92:2841-7].

Observational study of elderly medicare recipients in USA.  No 
intervention, retrospective notes review.  Those getting aspirin did 
better than those not, but the latter group were much sicker to start 
with (eg 5% more had a terminal disease).  There's no way to identify 
whether or not the doctors involved practiced EBM or not, so it's 
really only information about the intervention (aspirin or not), not 
about the general mode of practice of the doctors.

7) where docs fail to use e-b indications for carotid surgery, 
peri-op stroke and death are many times higher than would have 
occurred if the patients had been left along [stroke 1997;28:891-8].

Again the same criticism as for 6).  It doesn't measure whether 
or not doctors are EBM practitioners, but whether or not this aspect 
of their practice follows an external standard defined by an EBM 
approach.  Therefore doesn't address the question of whether 
explicitly practicing EBM alters patient outcomes

8) where e-b docs use more warfarin and stroke unit referrals,
mortality declines by >20% [stroke 1996;27:1937-43].

Again the same criticism as for 6).  It's derived from routine data 
so there is no information about whether doctors are EBM 
practitioners.  Although it shows that involving neurologists alters 
care (probably for the better) the casemix information is limited and 
they say that they can't exclude clinical triaging on the basis of 
characteristics not captured from admin data as an explanation for 
the differences seen.

9)  Gill P, Dowell AC, Neal RD, Smith N, Heywood P, Wilson AE:
Evidence based general practice: a retrospective study of
interventions in one training practice. BMJ 1996;312:819-821.

No comparison group therefore doesn't address the question.

10)  Ellis J, Mulligan I, Rowe J, Sackett DL: Inpatient general 
medicine is evidence based. Lancet 1995;346:407-410.

Ditto

11)  Campbell NC et al.  Secondary prevention clinics for CHD: 
randomised trial of effect on health.  BMJ 1998, 316, 1434-1437

RCT randomising patients to be invited (or not) to attend a secondary 
prevention clinic for CHD aimed at increasing uptake of beneficial 
interventions (aspirin, lipid management, stopping smoking etc).  
They don't report the effect on these intermediate outcomes.  They 
do show some fairly substantial effects on various real outcomes 
(better 'health' as measured by SF36, reduced hospital admissions) in 
the first year.  It may be that this was due to changes in the 
intermediate outcomes they were targeting although the reduction in 
hospital admissions was for all conditions not just cardiac ones, but 
we won't know until the trial is complete and reported.  I saw this 
paper presented at a conference and my memory is that the 
intermediate outcomes which changed were not ones which would be 
expected to cause big changes in the first year (ie they were changes 
in statin therapy rather than fewer smokers or more aspirin).

Again though, it doesn't directly address the issue of whether or 
not I should globally practice EBM. 
Bruce Guthrie,
MRC Training Fellow in Health Services Research,
Department of General Practice,
University of Edinburgh,
20 West Richmond Street,
Edinburgh EH8 9DX
Tel 0131 650 9237
e-mail [log in to unmask]


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%