How are we defining Real World in RWE?
Given the history of the all things real and reality, we need to ask, how close to reality is what we call “Real World Evidence”?
Current definitions include information that does not come from a typical clinical trial, and can include EHR data, claims data, genomic data, and, mortality data. In reality, this data doesn’t come directly from patients, there’s whole system in the middle. Still EHRs are known to be inaccurate. We all know a claim is not necessarily and event, and a non-claim is not necessarily a non-event. Many things happen in the lives of a patient that would never become apparent. When it comes to patient data, all data is imperfect of course. The job is to get as close to reality as possible, and start to understand trends both for populations, but also for individuals. We want an understanding of the patient journey and what things impact the course or trajectory of their journey. But the name begs the question, “Are these kinds of data really ‘real world’ data?”
Oftentimes, it is the best we have. Most “real world” data does not include any at-home data, but that’s changing.
A recent twitter thread including some oncologists gets at the heart of the matter.
Same thing @SABCSSanAntonio where lots of posters have titles like "Real world outcomes for XX therapy." Of course, any given clinic/hospital in any given part of the US or rest of world is not in any way representative of the "real world," arguably even less so that large RCTs https://t.co/bktxVljg2l— Hal Burstein, MD (@DrHBurstein) December 13, 2018
Clinics are rarely the real world, even if they are the sometimes the best measurements we have had in the past. We are entering a new era of real possibility with digital medicine, with real science. The FDA has approved 11 algorithms (link). The FDA has issued new guidance on “real world medicine. We’ll be following up with detailed posts on each of these in the coming days, and I hope you’ll join us.
As we move toward more at-home monitoring and at-home diagnostics in real time and with measurements taken sometimes continuously over very long periods of time, we can get close and closer to what the “real world” looks like. Our goal here is to help build consensus on what the “RW” in RWE and RWD really mean today, and what they can mean tomorrow..