A discussion with Rheumatologist and Researcher, Atul Deodhar, MD
How many people in the U.S. have spondyloarthritis? One might think this would be a simple task – determining an accurate, real-world number; but, as it turns out, that is far from simple.
We reached out to one of our foremost SpA experts to gain a full understanding of what is known, what is unknown, and the gray areas in between. We can’t thank him enough for granting us this enlightening interview! (Conducted just after Dr. Deodhar completed a hike up to a glacier on Mount Hood, in Oregon.)
Let’s start with what we think we know.
We’ve been using a figure of roughly 3.2 million as our estimated number of U.S. adults who have spondyloarthritis (ankylosing spondylitis and all related diseases combined). We’ve based this on a 2007 study by Helmick et al, which suggested that the prevalence of overall SpA in the U.S. is about 1.3% of the U.S. adult population. Hence, doing some math with the 2020 U.S. population numbers, we get to about 3.2 million.
Do you think this is an accurate and real-world realistic number? Or is it just the best we have?
Dr. Deodhar: The Helmick paper is a collection of different papers. It’s a review article about the burden of rheumatic diseases in the United States. Helmick and his coauthors looked at various published data about how common are, or the prevalence of, certain rheumatic diseases, including spondyloarthritis.
But that’s an older paper and in my opinion the estimates there are under-represented. When this was published, the NHANES study – which we view today as the gold standard – wasn’t yet done, so its data wasn’t available.
Now the NHANES study looked purely at axial spondyloarthritis prevalence, which as you know accounts for about half of the entire spondyloarthritis cases, or even less actually. The study reported that the prevalence of axial spondyloarthritis by itself was about 1%.
So now, if you add only psoriatic arthritis, which is thought to also be 1% – so another 3.2 million (by the way, this is a derived number for PsA, and I’ll get into that a bit later) you already have 2% of the U.S. population being impacted, and you still haven’t captured reactive arthritis, or those with IBD-associated arthritis. So realistically, we’re probably looking at a number at minimum roughly around 6.5 million to 7 million. But I don’t have a specific study to point you to. It just doesn’t exist. There hasn’t been a good quality study about the prevalence of total spondyloarthritis in the U.S.
Why do you think that is? Why do we lack a definitive study with clear numbers for all of SpA?
Dr. Deodhar: It’s a huge undertaking to do a study of that magnitude, and is very expensive. Someone has to take the initiative to do this population prevalence study, and secure the funding needed. Unless somebody takes the initiative… and SAA could do that. SAA could decide you want to do this study, or go after funding for it. So unless someone takes the initiative, it won’t get done.
So the gauntlet has been thrown down by Dr. Deodhar. Get this study done!
Dr. Deodhar: Yes! We should start a massive $10 million fundraising project and then we can do that study.
Let me also say there are two ways you can study prevalence. One is determining population prevalence, like the NHANES study. This kind of study investigates the general population to find the number of people who have the disease. This kind of study requires a lot of infrastructure, time, resources and a great desire to do it. It is challenging, and a very big undertaking.
The second way of studying prevalence is by looking at diagnostic prevalence. Diagnostic prevalence looks at an institution, say UCLA, asking them to provide the number of people they saw in the last year who had the various forms of spondyloarthritis (how many with PsA? With IBD-associated arthritis? With axial spondyloarthritis, etc.) And you add those all up.
And then you ask them to provide the total number of patients they had in that same year. So let’s say, just for example you find that out of the one-million people who came to UCLA Medical School that year, 10,000 of them had one of these conditions. So that’s a number of 10,000 out of 1-million.
And finally you extrapolate from that applying that same math to the total U.S. population. But this is diagnostic prevalence, which counts only those with SpA who came to UCLA, and were diagnosed with or treated for one of these conditions. This approach excludes people outside walking on the street in Los Angeles who have the disease, but would never go to a doctor because they don’t have the resources, or they don’t know where to go, or whatever the reason is. Or maybe they went once and were never diagnosed, or misdiagnosed. et cetera, et cetera. In short, diagnostic prevalence is always an underestimation of the real prevalence. The gold standard is a population prevalence study – like the NHANES study.
Your earlier question was why we don’t have a solid population prevalence study for all of SpA. Yes, it’s very expensive, but also it needs to be planned out very carefully, and many things need to be thought through: What kind of questions will be asked; what kind of examinations we will do, what kind of blood tests do we use, and start building and planning for it, and one day – get to it.
Can we also go back and elaborate on the 1% of the U.S. population with psoriatic arthritis figure being a derived number? What does this term signify?
Dr. Deodhar: Yes. Derived number just means that there was no direct study done for psoriatic arthritis prevalence. In this case, the study was done on psoriasis – so a very good, NHANES study was done which showed that the population prevalence of psoriasis in the U.S. is 3.2%.
So that is now a fact. And then another fact is that 30% of those with psoriasis have psoriatic arthritis. Doing some math, 30% of that would be roughly 1% again. We can say that 1% of the U.S. population probably has psoriatic arthritis. That’s what it means to be a derived number. It is not from a study. It is a calculated number, so we are less confident with it.
Do you personally think it’s important to have a high-quality population prevalence study completed, producing a definitive number for spondyloarthritis prevalence? Why or why not?
Dr. Deodhar: These things are important. Yes, absolutely. We need to understand the burden of the disease, and how many are impacted. The world’s governments need to have this information in deciding where to put resources, to plan for the accurate percentage of people with chronic illness, percentage who might become disabled. It’s important for public health planning, and for research, and for our patients to know as well.
To try and estimate a realistic number, in the absence of a definitive study is difficult, but we do have that 2% figure. Would you feel comfortable in saying that the population prevalence of the entire family of spondyloarthritis is at least 2% of the U.S. population, so at least 6.4 million adults?
Dr. Deodhar: Yes, I’m quite comfortable with that. Though we lack a precise study, I can defend the estimate stating some of the information I’ve already shared with you.
Thank you so much! Any closing thoughts on all of this?
Dr. Deodhar: I would say that spondyloarthritis has been a stepchild for rheumatologists in the U.S. It has been mostly rheumatoid arthritis, gout, and lupus to get the attention and funding for research, as well as from rheumatologists. Spondyloarthritis is slowly regaining its rightful place in large part because there are newer drugs on the market, and more treatment options coming to market. People are getting interested, and patients themselves as well as organizations like SAA are actively pushing for awareness and recognition.
The classification criteria has opened eyes about axial spondyloarthritis. New terminology and an ICD-10 code has come in: Non-radiographic axial spondyloarthritis – we didn’t even know about this term. We knew the existence of the condition and we called it generally undifferentiated spondyloarthritis. But now people are becoming aware of this condition as well. So my parting words – I believe that people are gaining interest in spondyloarthritis and once they start taking interest, then the type of study that you’re looking for – researching the true population prevalence of all of SpA, is more likely.
By Atul Deodhar, MD
Professor of Medicine
Oregon Health & Science University
Longstanding member of SAA’s Medical and Scientific Advisory Board