top of page

New dashboard puts spotlight on research waste at German medical universities

A new dashboard released today shows how 35 German universities perform on key clinical trial transparency metrics. The new tool is designed to help institutions to improve their trial registration and reporting.


[Tagesschau und Sueddeutsche Zeitung and FAZ have reported on this in German language.]

[Abbreviated English language translation of a media report here.]


What does the dashboard show?


Research waste for non-drug trials


This is depressing. Only 69% of results of non-drug trials were made public within five years of trial completion. Since very few results are published later than that, this means that nearly a third of trials ended up as research waste.


Wurzburg performs best, with 20 of 23 trial results made public. The gap between top and bottom performers is large. While Wurzburg published 87% of trial results, Aachen only published 50%, meaning that half of its non-drug trials ended up as research waste.

Note: This metric includes only trials registered on ClinicalTrials.gov or DRKS, so it excludes investigative drug trials (CTIMPs) unless they were double registered. It counts both results published in journals and summary results posted on registries.


Slow publication of results for non-drug trials


German universities are publishing the results of non-drug trials far too slowly.


Only 41% of such trials made their results public within two years of trial completion. Homburg (58%) is the only university to publish more than half of its results within two years. Greifswald (20%) does so for only one in five trials.


Scientific and clinical progress could be significantly speeded up if German universities routinely uploaded all trial results onto registries within a year, as recommended by the WHO. Most German universities now do this for drug trials (see below), but so far none of them seems to have started systematically reporting non-drug trial results onto registries.

Note: This metric includes only trials registered on ClinicalTrials.gov or DRKS, so it excludes investigative drug trials (CTIMPs) unless they were double registered. It counts both results published in journals and summary results posted on registries.


Registry reporting for drug trials


Most German institutions have made huge progress on this metric in recent years.


The registry now contains 79% of their due results, and most universities are approaching 100%. Hannover (30%) is the only large sponsor that still owes the public over half of its results.


Uploading the results of drug trials onto the European trial registry is a regulatory requirement. Most German universities have put a lot of work into improving their drug trial reporting, and German regulators BfArM and Paul-Ehrlich-Institut have been actively following up on missing results.

Note: This metric is limited to investigative drug trials (CTIMPs) and is based on EU Trials Tracker data from a few months ago.


Prospective registration


The data show significant improvements in trial registration over time.


Registering a clinical trial before it starts is a global ethics requirement that German ethics committees have explicitly endorsed. Pre-registration acts as a safeguard against research waste and dubious statistical practices.


Assuming trends are constant, today around 80% of new non-drug trials are probably being prospectively registered. However, by extension, this means 20% of trials probably still violate ethics rules.

Note: This metric includes only trials registered on ClinicalTrials.gov or DRKS. Drug trials (CTIMPs) are directly registered by regulators, not by universities.


Are institutions improving?


Maybe the best feature of the new tool is that it allows individual universities to track whether they are improving over time.


For example, here you can see how prospective trial registration in ClinicalTrials.gov has gradually improved at the Charite in Berlin (and German universities overall) since 2006.


The chart below shows how the university of Cologne has consistently been ahead of its peers in uploading due drug trial results onto the European registry.


How does your university perform?


Visit the dashboard now and check how your own university performs.


Additional metrics include open access publication, and linkages between publications and registries. The creators are planning to regularly update the dashboard in future.


Explainer: Drug versus non-drug trials


Clinical trials fall into three regulatory categories in Germany: drug, device, and other.


German law mirrors European law in this regard.


Drug trials are directly registered by regulators on the European EudraCT/CTIS registry. Universities are legally obliged to make their results public on the registry within one year of trial completion. German regulators BfArM and Paul-Ehrlich-Institut are actively engaging with universities and pharma companies to make sure that they follow the rules.


Non-drug trials are registered by universities themselves on the DRKS or ClinicalTrials.gov registry.

  • In theory, there is a legal requirement to make the results of medical device trials public on a new European database. In practice, the European Commission messed up. The database (Eudamed) is years behind schedule and still not fully functional.

  • There is no legal requirement to make public the results of all other clinical trials. For example, it is legal (but unethical) to run a trial testing different surgical techniques on German patients but never publish the results.


Still confused about regulations? Check out the start of this video for a quick explanation (German transcript available). For even more more detail, check out this report (in English) and this paper (in German).



Note: The dashboard was developed by a team at the BIH-QUEST Center for Responsible Research in Berlin. TranspariMED was not involved in this work.


For details on the methodology, read the paper. All code and data have been posted online, so this could be implemented with different trial cohorts and in different countries. The lead researcher can be contacted on Twitter.


コメント


Subscribe
Subscribe
Recent Posts
bottom of page