HI,
Piersante raises an important topic. Do you know of any informative source
that compares risk differences and ratio measures. It has to do with control
group prevalence. But how?
Or is the argument against use of risk difference aimed at hiding away the
often tiny results seen when comparing differences. I would like to see some
good examples that illustrates the weaknesses in using differences compared to
ratios.
Best,
Bengt Brorsson
>----Ursprungligt meddelande----
>Från: [log in to unmask]
>Datum: 2012-03-05 00:17
>Till: <[log in to unmask]>
>Ärende: Re: Clinical utility of the Cochrane reviews
>
>Dear Piersante and all,
>
>The communication and usability of results is a really important issue.
>Speaking as the Cochrane Collaboration's Training Co-ordinator, I hope
>you'll be pleased to hear that we absolutely agree. Current Cochrane
>guidance to authors encourages the use of both relative and absolute
>effect measures, discussion of the clinical importance of the observed
>results, and the inclusion of Summary of Findings tables (as mentioned by
>others) that include both baseline and post-intervention values as well as
>the relative effect measure.
>
>Having said that, it's important to note that any health care provider
>will still need to use their own judgement to determine whether the
>patient in front of them has a similar baseline risk to the study
>participants included in the review - you might be dealing with patients
>in a very different context, or with very different underlying conditions
>than the included studies. If that's the case, then the relative effect
>measure may be more useful to estimate the likely effect for your patient.
>
>A brief answer to a complex issue. Although this guidance isn't yet
>reflected in every published Cochrane review, we hope you'll find things
>improving over time. Feedback from users of reviews is always helpful in
>our ongoing efforts to improve our reviews, so thanks.
>
>all the best,
>Miranda
>
>Miranda Cumpston
>Training Co-ordinator
>The Cochrane Collaboration
>
>[log in to unmask]
>http://training.cochrane.org
>Skype: miranda.cumpston
>p.+61 3 9903 0381 m.+61 423 550 833
>
>Australasian Cochrane Centre
>School of Public Health and Preventive Medicine
>Monash University
>The Alfred Centre
>Lvl 6, 99 Commercial Rd
>Melbourne VIC 3004 Australia
>
>-----Original Message-----
>From: Evidence based health (EBH)
>[mailto:[log in to unmask]] On Behalf Of Piersante
>Sestini
>Sent: Sunday, 4 March 2012 9:38 AM
>To: [log in to unmask]
>Subject: Clinical utility of the Cochrane reviews
>
>I find Cochrane reviews less and less useful in addressing clinical
>problems in practice.
>The point is that they insist in providing just odds ratios, which on one
>side are of limited value to a literate reader (me) without knowing the
>baseline risk, and are of no value at all for the other side of the
>problem (the patient).
>Why in the hell don't Cochrane review give properly the baseline and
>post-intervention rate, which is what we all agree is the more reasonable
>data on which one can make an informed choice?
>
> Is there a reason to it and/or a way to change this state of things?
>
>thanks,
>Piersante Sestini
>
|