A team of researchers recently evaluated whether the Health Assessment Questionnaire‐Disability Index (HAQ-DI) is interchangeable with the Patient‐Reported Outcomes Measurement Information System‐Physical Function (PROMIS-PF) when measuring minimal disease activity (MDA) among patients with psoriatic arthritis (PsA). The findings showed that there is “excellent agreement” between HAQ‐DI and PROMIS‐based MDA definitions statically and longitudinally. In this podcast, study coauthor Ana-Maria Orbai, MD, MHS, talks about her team’s findings and how they may influence the way you calculate MDA. Full transcript below.
Ana-Maria Orbai, MD, MHS, is an assistant professor of medicine at the Johns Hopkins University School of Medicine. There, she is also director of the Division of Rheumatology’s Psoriatic Arthritis Program.
Additional Resources:
1. Chew E, Perin J, Grader‐Beck T, Orbai AM. Measurement of minimal disease activity in psoriatic arthritis using PROMIS‐Physical Function or the Health Assessment Questionnaire‐Disability Index. Arthritis Care Res (Hoboken). Published online August 29, 2020. doi:10.1002/acr.24433
2. Schalet BD, Revicki DA, Cook KF, Krishnan E, Fries JF, Cella D. Establishing a common metric for physical function: linking the HAQ-DI and SF-36 PF subscale to PROMIS(®) Physical Function. J Gen Intern Med. 2015;30(10):1517-1523. doi:10.1007/s11606-015-3360-0
TRANSCRIPT:
Announcer: The content for this topic is independent and is made available with support from Amgen.
Colleen Murphy: Hello, everyone. Welcome to another installment of “Podcasts360,” your go‑to resource for medical news and clinical updates. I’m your moderator, Colleen Murphy, with Consultant360 Specialty Network.
Today, we’re continuing our special podcast series on psoriatic arthritis. Minimal disease activity is a treatment goal that you may be helping your patients with psoriatic arthritis achieve. How do you calculate for such activity?
A team of researchers at Johns Hopkins University recently evaluated whether the Health Assessment Questionnaire Disability Index is interchangeable with the Patient‑Reported Outcomes Measurement Information System-Physical Function when measuring minimal disease activity among patients with psoriatic arthritis.
Dr Ana‑Maria Orbai was one of the members of that research team. Dr Orbai is an assistant professor of medicine at the Johns Hopkins University School of Medicine. There, she’s also director of the Division of Rheumatology’s Psoriatic Arthritis Program.
She joins me today to talk about her findings and how they may impact the future calculation of minimal disease activity in psoriatic arthritis. Thank you for talking with me today, Dr Orbai.
Ana‑Maria Orbai: My pleasure to be here, Colleen. Thank you for picking our study for this podcast.
CM: Of course. Now, before we get into the specifics of your study, would you actually be able to provide some background information on minimal disease activity, maybe the potential that the state of disease activity has as a treatment target and in a treat‑to‑target strategy?
AO: Absolutely. Psoriatic arthritis is a multidimensional rheumatologic disease. Its complexity lies in the fact that it affects the skin, the joints, the enthesis, as well as causing symptoms and having quite a lot of impact on the patient’s function and well‑being.
It has become important, especially with the availability of new therapies, to be able to get patients to a state where they truly are doing well in terms of disease activity as well as disease impact. Minimal disease activity is a target that is specific to psoriatic arthritis. At least in the clinic, it’s relatively easy to implement because it works like a checklist. It evaluates the number of swollen and tender joint counts, and there’s a threshold.
To be in minimal disease activity, a patient with psoriatic arthritis should have at most 1 swollen joint, at most 1 tender joint, no more than 1 tender enthesitis point. It goes through a list of 7 criteria, and the state is achieved if at least 5 out of 7 criteria are met.
One of the minimal disease activity criteria is physical function, which traditionally has been measured using the Health Assessment Questionnaire Disability Index. This was developed for rheumatoid arthritis at Stanford a long time ago and has become the measure of choice for physical function in rheumatology. It’s also used in rheumatoid arthritis, psoriatic arthritis. It has become an objective to achieve on its own in clinical trials, so obviously, function is very important.
To give you some background about why we may want to measure physical function differently: The PROMIS system of measures was developed more recently and in collaboration with the NIH. It has several advantages compared to older measures. For example, PROMIS measures are questionnaires that were developed with patient input. They are normed to the general population. The scoring system is based on T‑score metric where a score of 50 represents the general population mean, and 10 points represents one standard deviation. When we get a score on the PROMIS measure of physical function, for example, let’s say someone’s score is 60, we know that the population range is 50. If someone has this score, they are doing one standard deviation better than the population average. They are doing well. If we are in the other direction from the other score of 40, we know that they are one standard deviation below the population mean. Hopefully, we can improve things. It turns out that, on average, people with psoriatic arthritis tend to have that average score around 40 on the PROMIS measure. It’s easy to interpret. PROMIS scores have been implemented in electronic medical records like Epic, which we use at Hopkins, and it’s used at multiple other institutions. It’s easy to assess in regular clinical care.
Our question was, do we have to continue to administer the health assessment questionnaire when we have this newer, shorter measure, potentially more feasible, patients can complete it on a tablet? It’s a generation too of the physical function measures that existed. That was the reason for our work. The impact is going to be that in regular clinical care, we’ll be able to calculate minimal disease activity with which patient‑reported outcome that we collect at the clinic visit. These can be available during the clinic visit. Treatment decisions can be taken directly with the patient at the time of the clinic visit.
CM: That’s great background. Thank you so much for that. We went over the why. I guess we can go over the how. Can you talk a little bit about how you conducted your research?
AO: The Psoriatic Arthritis Cohort is a longitudinal study that is approved by the Institutional Review Boards at Johns Hopkins. People who consent to be participants in this study complete questionnaires in conjunction with their rheumatology clinical visit. They also consent to us being able to analyze aggregate data from their clinical status as well as the study questionnaires, as well as blood samples. The study focused on analyzing the scores. We collected concomitantly HAQ‑DI scores—the old physical function measure—as well as PROMIS measures.
We’ve compared definitions of minimal disease activity when we used the traditional cutoff of the HAQ‑DI that used to calculate MDA. We compared that with a definition that incorporates the PROMIS score. We were fortunate that the team of Schalet, et al. from Northwestern developed a walkway between HAQ‑DI score and PROMIS score. We had available this cutoff of 41.3 which was equivalent in the general population to the cutoff that we used for MDA for the HAQ‑DI score. We simply compared 3 sets of definition. The MDA, calculated using the HAQ‑DI; the MDA, calculated using the PROMIS physical function short form, which is only 4 items, 4 questions to assess physical function; and the MDA definition, using the PROMIS Computer Adaptive Tests for physical function, which was programmed to administer a maximum of 8 items.
Having collected these measures on every patient, we were able to analyze the agreement between MDA using the HAQ as well as MDA using these 2 PROMIS measures that more comprehensive computer adaptive test, as well as the 4r‑item short form. We found excellent agreement between these measures and almost overlap between the groups.
We went into more detail to also analyze where there are some differences between relevant groups. For example, were these definitions less likely to agree in men vs women, or are they less likely to agree in people with higher levels of pain? We were rigorous about doing subgroup analysis to make sure that the correspondence between definitions remains true. We were able to show that this walkway holds even in different groups based on gender, levels of symptoms, or disease activity.
More so, we wanted to see if the PROMIS thresholds using ROC curve analysis were able to predict the HAQ‑DI definition. The answer there was that the performance of the PROMIS measures was very good compared to the HAQ‑DI. If researchers or clinicians, for example, want to determine MDA, and they use PROMIS in the clinic, there is no need to continue to collect the HAQ‑DI. They can actually use this threshold of a PROMIS score, more than 41.3, to check off the criteria of minimal disease activity for the physical function item because these definitions are almost equivalent.
CM: That’s great. Thank you so much for explaining how our listeners can take your findings and implement them into their management decision. Can you envision these findings influencing management decisions in any other ways besides the ones that you just mentioned?
AO: This will be very helpful for our clinicians and patients when they discuss treatment targets at the clinic visit. I always like to work with an objective—and I know my colleagues as well—it’s much easier to have a goal when we work with a patient. Completing the checklist, doing the PROMIS scores, will give the clinician a very good idea of where we are, how well the treatment is working, and so on.
Moving on to treatment. We can assess on a larger scale. If institutions collect these measures uniformly, we can assess at a larger scale in longitudinal studies how well treatments are working.
Simply by using, perhaps, higher quality measures or population norm measures, we’ll be able to learn more about psoriatic arthritis and if we are truly able to change the course of this disease with the new therapies that are available.
CM: Now, for any clinician who might be thinking about making that switch from the HAQ‑DI to PROMIS, is there anything that they should keep in mind when doing that switch to make sure that it's a successful and smooth transition?
AO: One thing that is difficult in, I guess, clinical practice and when patient volumes are high, and our visit times are not getting longer. It’s becoming difficult to collect the HAQ‑DI, which is the 20‑item questionnaire consistently, especially if there is not a computer‑based method or EMR support, right?
If the patients have to complete this questionnaire on a piece of paper, and then you’d have to calculate the score manually, being able to do this for the HAQ‑DI becomes very, very difficult in clinical practice. I doubt that it could be done in real time. With the PROMIS questionnaire especially that the PROMIS short form is only 4 items, you’ll have a huge feasibility advantage. Even if you’re in a paper‑based clinic, it’s much easier to administer a 4‑item questionnaire, to use a conversion table—which gives you exactly the T‑score—and then plug it into the minimal disease activity definition, and see whether the patient is a target or not. It is much more feasible.
With the electronic medical record…the PROMIS measures are the preferred ones and that are available. The electronic medical records usually calculate the PROMIS score instantly. It can be imported into clinical notes as soon as the patient completes it, and the score is calculated. You can actually discuss it with patients during the visit.
CM: Now, taking a step back and, again, looking at this more at a macro level of minimal disease activity: What future work is needed to even better understand or measure the state of minimal disease activity among patients with psoriatic arthritis?
AO: Currently, minimal disease activity is centered around the joint count, the enthesis, the physical function, a patient global assessment, and this is a data‑driven definition that has established itself as the treatment target.
Several of the measures that are included in them or this activity, particularly enthesitis, do have some challenges. For example, we know that enthesitis is not uniformly assessing clinical practices. There are constraints on time.
Most people assess joint counts. For minimal disease activity, we really need to assess the 66/68 joint count, who even have enthesitis. Some of these measures that are included in the minimal disease activity definition could be improved.
Our study is just a step towards improving minimal disease activity. Showing that PROMIS measures and the HAQ‑DI is interchangeable. Similar things could be done on improving the other definitions, and potentially making this minimal disease activity definition more representative of the spectrum of psoriatic arthritis, in terms of how it affects patients.
We’ll need better measure of how we measure enthesitis. We’ll probably need to understand better what the vision global assessment is. That’s where the direction is. These definitions may change as we learn more about psoriatic arthritis and about what is truly meaningful to patients. It’s a great definition to work with and quite feasible to implement in clinical care.
CM: Dr Orbai, I wanted to thank you again for your time. I hope our listeners have found what you presented here today to be beneficial.
AO: Thank you.
For more content on psoriatic arthritis, visit our Resource Center.
