by Dr. Femida Gwadry-Sridhar
The FDA recently released guidance on how to characterize real-world data (RWD). I am really glad they have now explained data standards are important and why ontologies are the underlying framework to characterize data elements. What does this mean for rare disease organizations and pharma?
As people flock to the RWD excitement, they may sometimes forget the tenets of what it takes for data to become evidence. This requires scientific rigour, thoughtful discussion with stakeholders—especially patients—to determine what is important to them, and, finally, well-articulated data variables with standardized data collection to ensure that the data are both reliable and reproducible.
The Rare Disease Journey Is Uphill
In September, I was hiking the base of Mount Etna and exploring its caves, and it reminded me of a common but nevertheless true metaphor: that the rare disease journey is filled with steep inclines and dark caves as patients, their families, healthcare professionals, and researchers search for effective treatments and cures. Capturing their experiences as RWD can accelerate the way to treatments, but only if it is done by following rigorous scientific methods. This includes following standardized procedures for collecting and characterizing RWD.
RWD are not just about scraping data from the electronic health records (EHR) and creating a commodity from these data. While different data sources collectively contribute to the creation of evidence, each of them bringing a signal to the otherwise noisy landscape of data, we need to remember that data do not just turn into evidence: they require care, consideration, structuring and science. Without the rigour that is required, we are doing a disservice to patients and caregivers, researchers and industry who need confidence in the data that are provided to them. Patients’ lives rely on the integrity of the data and the scientific approach taken to collect the data.
Choosing the Proper Path to Structure Data Properly
A common misconception when rare disease communities set up a patient registry is that they need to collect all possible data just in case the community might miss something important. I believe this is wrong.
Science is about asking the right questions. When selecting what information to collect in your patient registry, working with data scientists can help you decide what information would most benefit your rare disease community.
This includes setting the right terminology so data can be compared to other data. For example, a doctor might say “hypertension,” but a patient might say “high blood pressure.” If data continue to be recorded by using different words for the same condition, then researchers have to interpret the data, which increases bias. Properly structured data avoids this problem.
Real-World Evidence Guidelines Benefit Rare Disease Communities
Rare disease communities will only benefit from this new guidance because larger subsets of data will begin to carry meaning across diseases. They will therefore require less interpretation (as explained in the hypertension vs. high blood pressure example), can be used in more studies, and can therefore pave an easier path toward treatment.
What does this ultimately mean? Easier collaboration, which can help us find treatments faster.
