 |
 |
Hi Judy, For a few years, I was working on Comparative Effectiveness Reviews for AHRQ Evidence-base Practice Centres (EPC). During that time, I came to the stark realization that large reviews are both difficult to manage (especially when you have a large staff) and that the commercially-available products were expensive and did not really help me manage all aspects of the project. If you are looking for commercial solutions, here is a brief list of some available on the Cochrane website (http://ims.cochrane.org/revman/other-resources). For me the solution was to start writing my own program, module by module. The core concept is that if it can be achieved without human interaction then it should be automated to work without human interaction. For example, I used to spend a lot of time running search strategies in online databases (e.g. copy search term, paste it in textbox in Ovid Embase (or other database), click on the search button, wait for the page to refresh, and do it again). Now I feed my program the search strategy and take a break for a few minutes while it runs the search for me. Same thing for removing duplicates, etc. Also exporting citations for individual reviews, retrieving their decisions and comparing them for conflicts has become automated as well. Currently I am finalizing the data extraction module in order to allow reviews to extract data in an array of possible formats (e.g. means, medians, ranges, SD, SE, Var, computed effect sizes, etc.) with automatic conversion to a standard format by the program. The final product should be able to create a RevMan file without any copying or pasting of data. Most of the modules are still in beta-testing, but I actively use them in all my current reviews. This allows me to see first hand any bugs, and more often than not, jot down ideas for further improvements. I do have an earlier stable version that is only used for screening titles/ abstracts/ key words, but does allow the creation of independent file so that reviews can work independently and in duplicate. Also it does file the results and tells you which are included, excluded and where the conflicts lie for further discussion. This version has been used by at least three active centres in Canada for the past few years including an EPC and a Cochrane Field. If anyone wants you want to discuss my program further, or wants to see a demonstration, please feel free to email me off list. Sincerely, Ahmed > Date: Wed, 21 Aug 2013 13:13:09 +0100 > From: [log in to unmask] > Subject: Software for large systematic reviews > To: [log in to unmask] > > *apologies for cross posting* > > Dear all, > > I'm involved in planning a large systematic review of outcome instruments/measures (not effectiveness of interventions). I have run searches for a similar review in a smaller area so have scaled up my estimation of numbers. Searches may find approx. 90,000 records which will reduce by about 1/3 to approx. 60,000 when duplicates are removed. Several thousand full text papers will be read and I guess data will be extracted from 500+ papers. At least 2 reviewers will work on the review. > > please can anyone advise me with the following questions? > > Reference Management > How does Endnote handle 100,000 records in one library? the largest Endnote library I have encountered is 40,000 and it was slow to process these but got there in the end. > Would you recommend other reference management software for collating records together and de-duplicating them? > > Review Management software > What review management software would you recommend for handling a large review bearing in mind: > - its not an intervention or DAT review so RevMan is probably not suitable > - it will have different data extraction forms, different data collected for different types of measurements (probably 6 different forms) It needs to be flexible enough to have a large number of custom fields for data extraction for the variety of data being collected. > - It needs to allow access for multiple reviewers. > > would EPPI-reviewer 4 be suitable? I've never used it so I'd be very interested to hear your of your experiences if you have used it for a large review/synthesis. > > > Thank you very much for your help. I'm happy to circulate responses, > > best wishes > > Judy > > Judy Wright > Senior Information Specialist to LIHS and the NIHR Research Design Service Yorkshire & the Humber Leeds Institute of Health Sciences University of Leeds Room 1.14 Charles Thackrah Building > 101 Clarendon Road > Leeds LS2 9LJ > +44 (0)113 343 0876 > > http://medhealth.leeds.ac.uk/auhe/is > > www.facebook.com/HealthEconomicsLeeds<http://www.facebook.com/HealthEconomicsLeeds> > > www.rds-yh.nihr.ac.uk<http://www.rds-yh.nihr.ac.uk> > > > > To unsubscribe from this group and stop receiving emails from it, send an email to [log in to unmask]