I'd be very wary of designating every systematic review as level 1
evidence and every RCT as level 2 evidence. It would be very important
to appraise the quality of research whatever its methodology.
A poorly conducted systematic review could be riddled with bias and
error, for example, and I wouldn't want to assume that it was better
evidence than a well thought out, rigourously conducted RCT, without
looking closely at their design and execution.
Liz Payne
----Original Message----
From: [log in to unmask]
Date: 08/02/2007 23:10
To:
Subj: Re: Level of evidence
Hi Craig,
I’d be going with level 1 because for me the inherent value of a
systematic review is that you can be confident that you have (at least
in theory) all the evidence.
For that reason, a systematic review which contains only one RCT is
still higher level evidence than that one RCT alone.
I’ll be interested to hear what others think,
Tari Turner
Senior Project Officer
Monash Institute of Health Services Research
Monash Medical Centre
Locked Bag 29
Clayton VIC
Australia 3168
Ph: +61 3 9594 7568
Fx: +61 3 9594 7554
From: Evidence based health (EBH) [mailto:EVIDENCE-BASED-
[log in to unmask]] On Behalf Of Craig Lockwood
Sent: Friday, 3 August 2007 4:02 PM
To: [log in to unmask]
Subject: Level of evidence
Dear All,
Just had a scenario brought to my attention and hoped to benefit from
your collective wisdom.
If a systematic review with meta analysis is level 1 evidence, and
If a well designed RCT with adequate power is level 2
What is a systematic review of 2 or more RCTs that were not amendable
to meta synthesis?
Is it level 1 because they come from a systematic search and have been
narratively summarised
Or is it level 2 because although the search was systematic, narrative
summary does not increase the point estimate or power?
With thanks
Craig
|