Saturday, November 20, 2010

Using RevMan to Conduct the Meta Analysis

RevMan (or Review Manager) is designed as a review tool to facilitate the literature review and the meta analyses by the Cochrane Collaboration Group. RevMan can be downloaded from website for free. It can be installed into your system without requiring the system administer privilege. Thousands of systematic reviews and meta analyses published on the Cochrane Library are performed using RevMan. These systemic reviews and meta analyses have been one of the leading resources in evidence-based medicine.

RevMan can be easily used by the medical researchers who are non-statisticians. For statisticians who work in the medical research area, RevMan is an easy tool to perform the meta analyses and generate the graphs (forest plot, funnel plot) in publication standard.

The statistical method and statistical model are described in the document Standard statistical algorithms in Cochrane reviews by Jon Deeks and Julian Higgins and Cochrane Handbook for Systemic Review of Interventions. For statistical models, both fixed model and random model are included in the RevMan. For random models, DerSimonian and Laird random-effects models are used. This is most common random effects model used in Meta Analysis.  

RevMan 5 is extremely easy to use. Various tutorials, tips, webinars are provided in RevMan documentation website and The Cochrane Collaboration Open Learning Material. I find it is extremely useful to watch two webinars (especially the part 2 regarding the data and analyses. For

To perform a Meta analysis, RevMan is just a tool. There are a lot of works to be done prior to enter the information including data into the RevMan. Considerable time needs to be spent on the literature search. Since the data used in Meta analyses relies on the publications, some data needs to be converted first. For example, for outcomes measured in continuous variable, the published article may only provide the Standard Error or just the 95% confidence interval. The SE can be easily converted to the Standard Deviation by multiplying the square root of the sample size. If only the 95% confidence interval is available, the standard deviation can be approximated by normal approximation using upper bound = mean +/- 1.96 * SE.

11 comments:

Whispherer said...

Hello, I was wondering if non-Cochrane authors could use the RevMan program for conducting statistical analysis?

Web blog from Dr. Deng said...

I think so. Revman is a free software for everybody to use even though the Cochrane review is typically written by experts in the field.

Whispherer said...

ok, thanks. I downloaded it already but I was not sure. I am working on a master's thesis with two experts in the field but I was looking for a free and appropriate software that I could use, especially for the statistical analysis like forest plots, funnel plots etc.Your post was the only thing I found on the web about RevMan (that was not from the Cochrane Collab.)and it looked like it was only for Cochrane authors with an Archie sign in (which I obviously do not have).

Web blog from Dr. Deng said...

I think it is ok for use in your situation. If you are really worried, you can contact them.

Alternatively, if you know SAS, you can use SAS to perform Meta analysis. See a book https://support.sas.com/pubscat/bookdetails.jsp?pc=55810

Whispherer said...

Thanks so much for your help! I will contact them to be sure. Another question: Between SAS, Stata, RevMan and SPSS (I know there is SPSS macros for meta-analysis) which would you most recommend for an easy interface and completeness? I picked Revman because it was free and connected to Cochrane.I have the basic SPSS and can use it relatively easily. I am not as familiar with SAS and Stata. I heard of MIX as well which has scored high on usability and completeness. Generally I am looking for something that's cheap enough that maybe I can convince my school to fund and will be worth it. what do you think?

Web blog from Dr. Deng said...

I would say that Revman is more straightforward to use. If you know SAS pretty well, you can try SAS. SPSS AND STATA may also have macros for meta analysis, but we don't typically use these software.

There is a specialized softwarebfor performing meta analysis www.meta-analysis.com. I saw a lot of ads, but I have not tried it yet

InBetween said...

another question. Is it possible to do the data analysis and get the tables and plots without having to do the entire paper in RevMan?

Web blog from Dr. Deng said...

The answer is yes. But you have to type in some key information in order to get the table and graphl

frogie said...

Hello sir,

I was looking for online article on step by step instruction of doing a meta analysis. Please let me know.

Anonymous said...

Hello,
I read your useful comments and I took the liberty to ask you something.
I’m going to carried out a meta-analysis on the effectiveness of an intervention program both comparing pre and post treatment outcomes and the specific program with a control one. I was thinking to use the RevMan, however I’m not sure a) if RevMan calculate effect sizes and b) if it is appropriate for both type of effect sizes I need (pre-post contrasts and group contrasts). If not which meta-analytic software would you suggest?
Thank you in advance!

Anonymous said...

http://www.meta-analysis.com/index.php