Samia, I strongly advise you to keep RCTs and cohorts separate, you canot combine their results as they two different designs. Are the cohorts all prospective? If not you should further subdivide them, if they are comparative that is.
Good luck with interpreting the observational studies, especially if they tell you a different story from the RCTs.
Best wishes,
Tom.--On 9 May 2013 02:41, Paul Elias <[log in to unmask]> wrote:
I would not combine them in a meta analysis if this is the first q..you should not combine controlled studies with uncontrolled studies.....I would combine the RCTs and observational designs separately and discuss them as such. If you did combine (though I argue no), I would then do a sensitivity type analysis and separate the observationals from the full set (7????) to show the impact on the summary estimate etc.
Best,Paul E. Alexander
--- On Wed, 5/8/13, Mohamed El Shayeb <[log in to unmask]> wrote:
From: Mohamed El Shayeb <[log in to unmask]>
Subject:
To: [log in to unmask]
Received: Wednesday, May 8, 2013, 11:05 PMIn an article that I think someone in the list distributed few weeks ago (unfortunately can't remember who and when) the issue of dealing with heterogeneity in reviews was discussed. No criticism was directed to the presence or absence of heterogeneity itself. In fact, the article criticized usage of an inappropriate model, and emphasized the importance of addressing and quantifying heterogeneity (Q and I2 statistics); interpreting it if you can and use the proper model. I will try to find the article and send it to you.Hi Samia,In my humble opinion, there are two notions in this. First, and most important, the conceptual notion. This should be determined by the researcher's discretion. I think the questions that really count rather than just the design will be: Do the studies measure the same clinical end point? Are the results overall consistent in regards to the overall effect magnitude and direction? Are patients characteristics somehow similar across the studies?
The second notion is the statistical notion of heterogeneity. There is no arbitrary line or threshold for heterogeneity that you may consider for such a decision. Further, the presence or absence of heterogeneity will then dictate which model to use (random versus mixed effects model). The first assumes similar distribution among all the studies. The latter assumes that each individual study distribution is part of a bigger distribution for the combined effect and does the calculation accordingly.
I hope this is helpful for you and your team .Cheers,Mohamed El ShayebHealth Technology and Policy UnitUniversity of Alberta
3025 Research Transition Facility8308 114 street
Edmonton, Alberta, CanadaT6G2V2Tel: +1 (780) 248 1524
On Wed, May 8, 2013 at 10:05 AM, samia Alhabib <[log in to unmask]> wrote:
Dear list;we are doing systematic review of the effectiveness of DC beads and conventional chemotherapy in treating hepatocellular carcinoma, and we found only 2 RCTs and 5 cohort studies that fulfill our inclusion criteria. My Q: is it appropriate to combine these studies although of different designs? bear in mind that it is uncommon disease. ..Another Q that we debated with the research team; what is the level of heterogeneity that could be acceptable to combine the effect size?Appreciate your feedbackSamia
Dr. Samia Alhabib, MD, PhDSent from Samsung Mobile
Dr Tom Jefferson
www.attentiallebufale.it