What is real world evidence?

What is real world data? What is the difference between real world data and real world evidence?

Real world data are messy. You know, data, plural data data in the singular. So real world data are messy because, you know, in real life, life is messy. You know, patients often, you know, miss a dose because they were on vacation or in the hospital, or they may have gone to urgent care for an infection. And those records don't show up in the main centers' data. So, real world data are representative. Again, the strength of them is that they represent more patients getting treatment where they are, you know, regardless of how healthy they are, how frail they are, what language they speak, etc.

Real world data are messier, though, because of all the complexities inherent to life. Real world evidence is what we do when we take those real world data and try to analyze them in a way that adjusts for some of the complexities inherent to them, to kind of figure out what makes the most sense for patients. And so we'll discuss several examples of that where we use real world data, for example, in this case through HealthTree Cure Hub to help identify a research question and to answer that question. And the answer to that question will be real world evidence.

Real world data would be just, you know, looking at a map and real world evidence would be using that map to figure out the direction where we want to go. You know, for with our trip or something along those lines.

How is real world evidence generated from real world data?

I think we should have a question in advance first. So the scientific method that all of you hearing this talk, you know, learned in middle school or high school, you may know your children and grandchildren who've learned the same thing, is that you always start with a hypothesis. So before just seeing what happens, have a hypothesis to say, look, this is what I want to look at. And then from there, use the real world data to analyze that question. You know, those are your methods and have your results in your conclusion. And even though that sounds like something out of a high school science project, that is very literally what we still do today as clinician researchers using real world data.

So in my case, for example, you know, one recent project that we'll talk about with the HealthTree Cure Hub, we're looking for example, at this idea of dexamethasone, which many patients with myeloma are very familiar with being on Dex or dexamethasone, you know, once a week for weeks, months, years, and what are the outcomes of it? And there are data around that that were available and Cure Hub. And so we asked a question and we said, well, is there any link between dexamethasone exposure and eye issues, for example, cataracts. And so we use the real world data to actually investigate that. And we actually did find a link that more, you know, milligrams and more doses of dex someone had received in their lifetime milligrams per week, weeks per month, months per year and so forth, the higher their risk of cataracts.

And so we use that to generate real world evidence as our conclusion, saying that in the real world setting, dexamethasone in myeloma is associated with a higher risk of visually significant cataracts, and therefore future studies should look at can we get rid of the Dex earlier from our patients or reduce the dose?

So that would be a good example of where we took already available real world data. And we mined the data. We had a specific question in mind. We went and worked with the HealthTree team in this case. To answer that question, we submitted to a conference, we got it published, and now it's public domain literature that hopefully will inform future studies based on that real world evidence.