Public clinical trial registers were designed to comply with regulations (and make the public aware of past, ongoing, and future clinical trials), not for the purposes of clinical trial feasibility. This becomes abundantly clear once you try to gather the data you need for site selection or calculate recruitment rate for multiple sites in multiple countries.
Public registers have many limitations that go beyond incomplete or missing data. For one thing, they are intimidating to patients and don’t help much to enhance clinical trial awareness and participation – an issue my team and I are tackling through FindMeCure, our patient-friendly search platform for clinical trials.
On the other side of the coin are feasibility and startup experts. Public registers were not designed to serve their needs and in many cases, the data they do find can be misleading without the proper context. Let’s take a look at the 3 limits of public registers that routinely slow down clinical research… and why public registers are still the go-to for clinical trial data.
Limitation 1: Too many missing pieces.
While CT.gov is by far the best source of data for site feasibility, there is always some data that is missing due to one of many possible reasons: the climate of competition in the industry that tempts companies to hide information, the updates that have not been included yet, contacts no longer relevant or completely missing, etc.
Many times public registers have not been updated for more than six months and even now the WHO register, which is the worldwide clinical trial register, is not open for the public. Despite these setbacks, TrialHub updates its database in real-time because what you can see on the WHO register (with the search option completely missing) is not what it actually contains inside.
Considering all the ways data can be missing, incomplete, outdated or unavailable, feasibility experts find themselves in a strange position – they have to select the best sites and investigators while always keeping in mind that no one has the full picture and some pieces of the puzzle will never make it to them.
For European clinical trials not on CT.gov (although now more and more European trials get registered there), national registers provide data without any sites and investigators and even when they do, the data is for the local country only and the rest of the trial is left out.
Limitation 2: Searching for data is a nightmare.
If the missing pieces were not enough, feasibility experts soon encounter another issue: public clinical trial registers were not designed with their needs in mind. Although great for site-level detail, CT.gov was not created for the purposes of clinical trial feasibility and the search options reflect just that.
The search can help you find clinical trials but a feasibility expert cannot search and compare multiple countries and if she has to find the best sites in 5-10 countries this means hours upon hours of manual research. The inconvenient search options and the missing data mean you have to either rely on previous knowledge you have about the indication and the country or cross-reference dozens of sources in a matter of just a few days.
Things get even trickier if you rely on CT.gov for competition analysis – the depth and detail are simply not there or take a lot of manual calculation. This is why when designing the TrialHub algorithm our main priority was to provide you with fast and convenient calculations that go in-depth and show detail.
It’s not enough to know that there are 5 competing trials in Germany – something you can find out on national registers as well as CT.gov with some research. Instead, you want to know that they are run in 15 out of the 25 experienced sites, leaving you with 10 experienced sites without competition. This is the kind of detailed and precise calculation that can mean the world of difference for a clinical trial.
Limitation 3: Calculating recruitment rate is a pain.
Ct.gov is the only public clinical trial register that gives you the relevant data to predict recruitment rate (RR). Other public registers simply don’t provide this level of detail as they lack site-level data. However, even on CT.gov, you don’t have the RR readily available and it takes a lot of time (and manual calculations) to predict it for more than one trial.
We use Natural Language Processing to extract the data and automate the process – TrialHub analyzes every single trial to predict the RR and give you the data ready to use in a matter of seconds. To be able to do this, however, we need to rely on CT.gov data: it’s still the only public register that includes the information needed because it proves data on a site, not just country level.
Which leads us to the question – with so many missing pieces, why are public clinical trial registers still the go-to for feasibility?
Why are public registers still the go-to?
The short answer is that there is no real alternative. Public registers contain in one place the information about past, ongoing, and future clinical trials. If you were to search for this data elsewhere, you could easily spend weeks just gathering enough information about one trial.
Even with all the missing pieces, public registers have the advantage of being generally reliable (as the regulators require for the information to be publicly available) and providing the data in one place.
Of course, not all registers are created equal. As I already mentioned above, you cannot find data on sites and investigators on national registries, the WHO register, or the European register. So far, CT.gov is the only place where such data can be found. Most European clinical trials are also available there and this trend shows no signs of slowing.
Now, to answer a question feasibility experts will probably have – is it better to work with all available registers or to solely rely on CT.gov? As I said earlier, CT.gov is so far the only public register that provides the site-level data you need for competition analysis. By including other registers in your calculations, you don’t add clarity to the picture – quite the opposite, you are skewing the calculations.
However, by not including other registers and solely relying on CT.gov, you are missing out on a lot of clinical trials not yet registered on CT.gov. Those trials, though, are published on national registers where site-level detail is unavailable and competition analysis or calculating RR are impossible.
Ultimately, as a feasibility expert, you have to make a choice – do you prioritize detailed in-depth analysis or do you value having more data? On the surface, more is always better – it’s only when you need precise analysis that you realize that by adding more variables the calculations don’t add up.
For more on the topic, I recommend Maya Zlatanova’s video about “The More of Less”.