When I was a BMT (blood and marrow transplant) coordinator, I was intimately involved in every aspect of a transplant recipient's care—and in their lives as well. I could have told you my patients' medical histories, allergies, past and upcoming treatments, family members' names—probably even their favorite places to vacation.
I acquired as much knowledge about my patients as I possibly could to ensure the best chances for transplant success. At this micro level of patient knowledge, I was an over achiever.
But at the macro level, my tools were woefully inadequate. I can remember feeling incredibly frustrated when physicians or other transplant team members would ask me how past patients of mine had fared on a proposed treatment. Anecdotally, I might have been able to speak to one or two recent outcomes, but I had no way of definitely identifying trends that might influence a proposed treatment's success or failure.
The desire to find a way to aggregate and correlate patient data in order to identify trends and improve outcomes drove me to a new role at my hospital: quality improvement specialist. And while I did very meaningful work in this role, it was here that I discovered that the data needed to impact outcomes on a large scale often didn't exist. Or, if it did, it wasn't easily accessible.
I also discovered that this problem wasn't unique to my hospital—or the BMT field.