gms | German Medical Science

A high-quality systematic review, meta-analysis, and Cochrane community – all this represents a big challenge, especially if you do not have access to the data for analysis. Here I present my experience with the Vivli platform who helped me a lot to get the rough data for analysis. Also, in my presentation, I will describe several tips and lessons learned for using clinical trial data from the industry.

I was very thrilled when I started a collaboration with the Cochrane community, more specifically the Inflammatory Bowel Disease (IBD) group as an author of the Cochrane systematic review. I had an idea, made a protocol, published it and everything went well until I did not realize that almost all the data which I had to include in my review were difficult to reach. The first attempt was to contact the primary authors of potentially included papers. And the first obstacle was that about 3% of contacted authors responded to email and 1% of them gave me the information I needed. I was desperate, but I did not want to give up. The little hope emerged when one of the investigators (he was, of course, the first author of many articles in the field which I studied) advised me that I should seek help at the Vivli platform. He shared a link to Vivli. I read the document “How to-Requesting studies on Vivli”, searching for information that I needed. I admit, it seemed to be very complicated and I didn't hope to succeed, but I tried. The primary outcome was change in health-related quality of life scores as an efficacy endpoint. Mostly the primary outcome was the results of the IBD Questionnaire. So, the first request was sent on 5 November 2018 and Vivli and I signed Data Use Agreement (DUA) on 3 April 2019 for one year. One month later, I had access to all the requested studies! My statistician and I analyzed all the data using R within the Vivli platform online. We had some issues to resolve in the process of analysis – for example, we need to harmonize data from different studies so we contacted the Vivli support via Vivli chat available online. The whole process took longer than I was planned and in the meantime, I made a second request to Vivli (as an amendment to the first one) aiming to include a few more studies. I also extended DUA for 6 months. Finally, I had access to data from 8 studies, all high-quality phase 3, multicenter, randomized controlled trials sponsored by two different pharmaceutical companies.

In the last part of my talk, I will provide a number of lessons learned, which should make secondary analyses projects of clinical trial data a smooth process, if planned accordingly. From my experience, Vivli could be a very useful tool for research projects.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.