Thank you so much all who replied to this scenario - I enjoyed your
replies, and returned last week to pull them together. However, disaster
had struck! All of my replies had been deleted.
I have now got the system working again, so can apologise for not getting
back to you all sooner. I have replies still from Ian Bowns and Paul
Glasziou who posted their replies to the list, but not from those of you
who replied direct to me.
If any of you have copies of the answers you sent me about 6 weeks ago (the
message was posted to the site on 3rd July) I would be hugely grateful if
you could re-send them. And just incase you missed the message the first
time I repeat it below.
Thanks and apologies all, all the evidence based best, Lee
Systematic review results scenario
Please tell us what you think!! In order to interpret the results of a
review effectively the details of the intervention and the condition have
been excluded from these results so that a blinded opinion can be obtained
from as many people as possible.
Introduction
A systematic review has been undertaken to investigate the effectiveness of
intervention A (compared with no intervention) to reduce the incidence of
condition B (rare but serious). Intervention A is not expensive, and only
very occasionally has serious side effects.
Methods
A comprehensive search strategy was developed (MEDLINE, EMBASE, Cochrane
Library and others) and articles in any language were considered.
Randomised controlled trials (RCTs) and controlled clinical trials
(CCTs),cohort studies and case controlled studies were considered.
Results
A total of 980 references were identified by the search strategy; reading
of the titles and abstracts reduced the number to 105, which were obtained
and read. No RCTs, CCTs or cohort studies were identified. Only two case
control studies were identified.The case control studies (including 114
people in total) were pooled using odds ratios and random effects
methodology. The forest plot showed apooled odds ratio of 0.32 (95% CI
0.04 to 2.81). (Heterogeneity chi-square2.89, df=1, p=0.089).
Based upon these results (and assuming this to be an area of practice which
is relevant to you) we would like to know what action you would take in
your own clinical practice (you may choose more than one).
I will:
1. always use intervention A to prevent condition B
2. look out for RCTs, cohort or further case-controlled studies on which to
base my practice in future
3. ask my patients which they prefer
4. keep on doing what I do now
5. examine the potential harms of intervention A before considering using it
6. never use intervention A to prevent condition B
7. use the expert guidance from my professional body
8. other – please tell us what this is:
Answers and discussion very welcome, I will let you know the consensus,
thanks for your help with this, Lee
|