He ran his private clinic for 30 years, from 1974 to 2004, and he collected electronic data on every patient he saw. I remember filling out a four-page questionnaire in the waiting area and watching one of his staff type everything into a computer in 30 seconds. When he saw me in the examination room, he had a dot matrix printout showing how my pain, physical function, gastrointestinal issues and more had changed during the past several visits. I just assumed this was standard practice, yet 40 years later I’ve learned how uncommon it still is now—and how very much ahead of his time Dr. Wolfe was.
In addition to computer software, Dr. Wolfe taught himself biostatistics. From SAS in the 1970s–90s to S+ and then Stata, R and WinBugs, there was always a new method to learn, a faster way to analyze the clinical data or a better graph to make. He jokingly mentioned that it was easier to publish papers using a brand-new statistical method because few understood enough to criticize. (Fred’s extensive vocabulary often had me digging into the largest dictionaries at the library for interpretation.)
I’m not sure at what moment the research bug hit, but Fred’s first publications echoed his vocal opinions, with letters to the editor calling out findings in JAMA in 1973, citing his own use of electronic medical records in 1977 and citing bias in 1982.8-10 Then, nine years after starting his clinic, he started publishing original research manuscripts—18 in his first two years. This led to more than 650 published papers and being listed in 2019 as one of the top 0.01% of all published scientists.11
He wrote to me about this: “A letter to me from John Ioannidis, one of my heroes. According to this ranking of citations I am in the 99.9th percentile of all scientists, alas beaten out by Tony Fauci and Steve Weinberg and, alas among rheumatologists, Dave Felson. We think this is all funny, but fun. Fred”
Fred conducted research in many areas, yet he is likely most known for his work with patient-reported outcome measures, patient registries and fibromyalgia criteria. By being a pioneer in collecting patient data early, he learned from and helped create more meaningful measures. He recognized the value of learning what the patient experienced at the clinic visit and away because he wanted to know what happened to his patients who didn’t return. This led to his involvement with the late Jim Fries’ ARAMIS registry and eventually to the creation of the National Data Bank for Rheumatic Diseases (NDB) in 1998, which paralleled national registries starting in Sweden and the U.K.