Patient's global assessment of disease activity: What are we measuring?
Version of Record online: 27 AUG 2012
Copyright © 2012 by the American College of Rheumatology
Arthritis & Rheumatism
Volume 64, Issue 9, pages 2811–2813, September 2012
How to Cite
van Tuyl, L. H. D. and Boers, M. (2012), Patient's global assessment of disease activity: What are we measuring?. Arthritis & Rheumatism, 64: 2811–2813. doi: 10.1002/art.34540
- Issue online: 27 AUG 2012
- Version of Record online: 27 AUG 2012
- Accepted manuscript online: 21 MAY 2012 10:55AM EST
- Manuscript Accepted: 8 MAY 2012
- Manuscript Received: 24 APR 2012
There is a growing interest in the use of patient-reported outcomes in rheumatoid arthritis (RA) clinical trials as well as in clinical practice (1). More and more, patients actively participate in shared decision-making, and these processes require outcome measures that are relevant and understandable by patients (2). It has also been shown that many patient-reported outcomes can distinguish active treatments from control treatments as effectively as “objective” measures can (3–5). Most patient-reported outcomes data are much more easily collected than are joint counts or laboratory data (6).
The American College of Rheumatology (ACR) core set of disease activity measures in RA clinical trials contains 3 patient-reported outcomes: patient's global assessment of disease activity, pain, and physical function (7, 8). However, the Outcome Measures in Rheumatology (OMERACT) initiative, which included “expert patient” representatives as workshop participants, recommended in 2006 that all trials should additionally include a measure of fatigue (9, 10). This recommendation was endorsed by the European League Against Rheumatism (EULAR) and the ACR in their collaborative recommendations published in 2008 (11).
One of the available instruments for measuring fatigue is the recently validated Bristol RA Fatigue (BRAF) questionnaire, which was designed to measure fatigue in all its dimensions (12). Other new and promising patient-reported outcomes include the Rheumatoid Arthritis Impact of Disease (RAID) questionnaire, which measures the patient's perceived impact of RA on daily health (13), and the Routine Assessment of Patient Index Data 3 (RAPID-3), an index of 3 patient measures from the core set of disease activity measures in RA clinical trials (physical function, pain, and patient's global assessment) that was designed for use in a busy clinical setting (14).
Despite the growing interest in the development and use of patient-reported outcomes in the field of rheumatology, their interpretation is not always easy.
What are we measuring?
The most commonly used patient-reported outcome is possibly also the least straightforward in terms of interpretation. As an opener to the encounter with the patient in our clinic, we usually ask the patient something like, “How have you been?” In the global assessment question used in research, this question has been focused to something like, “Considering all the ways your arthritis affects you, how do you feel your arthritis is today?” This formulation invites the patient to take pain or constraints due to damaged joints into consideration, since this obviously is one of the ways that the arthritis affects the patient. However, in clinical trials in particular, researchers are mostly interested in the patient's opinion about current disease activity rather than the consequences of irreversible damage.
What does the patient's global assessment of disease activity measure? And, is that the same as what the physician's global assessment of disease activity measures? Considering the extent to which these measures are being used to guide treatment decisions, this is an important question. In this issue of Arthritis & Rheumatism, Studenic and colleagues (15) identify several factors that contribute to the patient's global assessment and the physician's global assessment, as well as the extent and reasons for the discordance. Specifically, they asked their patients to assess their disease activity during the previous week, and the investigators compared this result to their physicians' ratings of disease activity. This procedure was repeated after a mean of 4 months. Their study shows that by far the most important contributor to the patient's global assessment is pain (explaining 75.6% of the variability), with a very small contribution by the physical function score (1.3%) and the number of swollen joints (0.5%). In contrast, the swollen joint count is the most important contributor to the physician's global assessment (61%).
Although both tools aim to measure the same thing, it is not surprising that they do not. The patient is confronted with the disease on a daily basis, and pain is the most dominant symptom. Even though the authors specifically asked patients to rate their disease activity, pain due to comorbid conditions or RA joint damage cannot be excluded as contributors to their assessment. The physician sees the patient about once every 3 months and makes a judgment largely based on his or her medical experience and the visible signs of inflammation.
What worries us is that the patient and the physician rarely agree on the activity of the disease: in 76% of the observations, the patient's and the physician's global assessments were >5 mm apart and were thus classified as discordant. Five millimeters on a 100-mm scale can be considered a rather strict criterion, resulting in a high rate of discordance; however, in our calculation based on the data that were reported, ∼11% of assessments of change (>5 mm between assessments at a mean of 4 months apart) were maximally discordant, that is, where patients reported one direction of change (e.g., improvement) and physicians reported the other direction (e.g., deterioration). Studenic's analysis calls for awareness among physicians and researchers of the differences between specialists and patients on disease activity.
Studenic and colleagues are not the first to study the discrepancies between patients' and physicians' assessments of disease activity. In 2010, Barton et al (16) performed a very similar study, with the important distinction that pain was not one of the variables collected, and the clinically relevant discordance between patients and physicians was defined as >25 mm on a 100-mm visual analog scale. Hence, their results were different, with 35% discordance between the patients and their physicians and with the presence of greater depressive symptoms as the main independent predictor of a higher score by the patient than by the physician. Since depression has been shown to be associated with more pain and worse function, this finding supports those reported by Studenic et al. More recently, Khan et al (17) also reported that pain was the most important determinant of the patient's global assessment, followed by fatigue. This was in contrast to the physician's global assessment, which was influenced most by the swollen joint count, followed by the erythrocyte sedimentation rate and the tender joint count.
Although the patient's global assessment question was phrased differently in each of these 3 studies, all of them largely came to the same conclusion. However, a strength of the Studenic study is its longitudinal design, which allows for the analysis of discrepancies in assessments of change during treatment. In the clinic, it is useful to know that the patient's assessment of change is driven by the level of pain, whereas the physician's assessment of change is driven by the swollen joint count.
This is probably as far as statistics can help us to explain the difference between the patient's and the physician's global assessments. The next step in understanding factors that influence the two measures is through interviews or group discussions with patients and physicians who are experienced in evaluating global assessments and can share their ideas and motives behind their evaluations. That will possibly allow us to understand the remaining 25% of the variability in the perception of disease activity that is currently not explained by the core set of instruments.
The need to understand the patient's evaluation of disease activity becomes particularly important when the aim of treatment is remission. The new ACR/EULAR Boolean remission criteria require a patient's global assessment of ≤1 on a scale of 0–10 (18), a strict criterion that is hard to attain, especially for patients with comorbid conditions, such as a chronic pain syndrome, which is present in an estimated 17% of RA patients (19). It is nevertheless an important criterion, since it represents the patient's perception about the absence of disease activity. At least, that is what we assume, since no data are yet available on the patient's perception of remission in RA. It is possible that other factors are more important to patients when talking about the absence of disease activity. This is the main question of a qualitative study that we are currently conducting in cooperation with some of the Austrian group of investigators who conducted the Studenic et al study and colleagues from Bristol.
During the development process of the new remission criteria, it was highlighted that more information is needed concerning the patient's perspective on remission. At the tenth OMERACT meeting, we chaired a workshop dedicated to this subject, which resulted in a considerable research agenda, including clarification of the exact phrasing and meaning of the patient's global assessment (20).
The study by Studenic and colleagues (15) is an interesting contribution to understanding the patient global assessment and determinants of discrepancy between patient's global assessment and physician's global assessment. Next steps should include qualitative aspects to uncover the ideas and motives of patients when marking the stripe on a visual analog scale. This knowledge is particularly important now that therapies increasingly aim for remission, with treatment decisions being made based on criteria including the patient global assessment of disease activity.
Both authors drafted the article, revised it critically for important intellectual content, and approved the final version to be published.
- 12Measuring fatigue in rheumatoid arthritis: a cross-sectional study to evaluate the Bristol Rheumatoid Arthritis Fatigue Multi-Dimensional Questionnaire, visual analog scales, and numerical rating scales. Arthritis Care Res (Hoboken) 2010; 62: 1559–68., , , , , .
- 14A proposed continuous quality improvement approach to assessment and management of patients with rheumatoid arthritis without formal joint counts, based on quantitative Routine Assessment of Patient Index Data (RAPID) scores on a Multidimensional Health Assessment Questionnaire (MDHAQ). Best Pract Res Clin Rheumatol 2007; 21: 789–804., , , , .