See acknowledgments for members of the LUMINA Study Group
Systemic lupus erythematosus in three ethnic groups. XI. Sources of discrepancy in perception of disease activity: A comparison of physician and patient visual analog scale scores†
Version of Record online: 12 AUG 2002
Copyright © 2002 by the American College of Rheumatology
Arthritis Care & Research
Volume 47, Issue 4, pages 408–413, 15 August 2002
How to Cite
Alarcón, G. S., McGwin, G., Brooks, K., Roseman, J. M., Fessler, B. J., Sanchez, M. L., Bastian, H. M., Friedman, A. W., Baethge, B. A., Reveille, J. D. and LUMINA STUDY GROUP (2002), Systemic lupus erythematosus in three ethnic groups. XI. Sources of discrepancy in perception of disease activity: A comparison of physician and patient visual analog scale scores. Arthritis & Rheumatism, 47: 408–413. doi: 10.1002/art.10512
- Issue online: 12 AUG 2002
- Version of Record online: 12 AUG 2002
- Manuscript Accepted: 24 OCT 2001
- Manuscript Received: 1 JUN 2001
- NIH, Institute of Arthritis and Musculoskeletal and Skin Diseases. Grant Number: R01-AR-42503
- General Clinical Center Research Grants. Grant Numbers: M01-RR-02558, M01-RR-00073, M01-RR-00032
- Systemic lupus erythematosus;
- Disease activity;
- Visual analog scale;
To compare patient's and physician's assessment of disease activity in a multiethnic (Hispanic, African American, and Caucasian) cohort of systemic lupus erythematosus (SLE) patients.
Three hundred patients with SLE from the LUMINA (Lupus in Minority populations: Nature versus nurture) cohort were included. Disease activity was assessed with the Systemic Lupus Activity Measure (SLAM); patients and physicians assessed disease activity using a 10-cm anchored visual analog scale (VAS). The difference between VAS scores was termed discrepancy (>1 cm was considered a priori clinically relevant). Selected sociodemographic, clinical, behavioral, and psychological variables were examined in relation to discrepancy in univariable and multivariable models adjusting for the physician global VAS score in order to eliminate ceiling and floor effects.
A discrepancy was exhibited by 58% of the patients. Abnormal laboratory findings were negatively associated with discrepancy, and poor self-perceived functioning and joint involvement were positively associated with discrepancy. Ethnicity did not account for discrepant perception of disease activity.
Patients and physicians rate disease activity in SLE differently. Physicians appear to place more emphasis on laboratory features while patients place more emphasis on function.
Systemic lupus erythematosus (SLE) is a multisystemic autoimmune disease that primarily affects women during their reproductive years (1). It occurs in ethnically and socioeconomically diverse populations. There is evidence, however, that disease and outcome are differentially affected by age, ethnicity, and socioeconomic status (2–4).
The disease course of SLE is variable, but for the most part it is marked by periods of relatively quiescent clinical manifestations and periods of disease exacerbation or “flares.” Patients may not be fully aware of the significance of particular clinical manifestations as an indication of the activity or latency of their disease. The possible negative result is that patients may become complacent in the management of their disease. The challenge is not only to quantify the difference between the patient's and physician's perception of disease activity, but also to create a profile of the reasons why this discrepancy occurs and to minimize its impact on the disease course and outcome (5, 6).
Many factors cause patients' perception of disease activity to either coincide with or diverge from the opinion of the treating physician. Because each culture has its own model of disease causation, course, and outcome, one might suspect that ethnicity could be a significant influence in the way patients perceive their own health in general, and the degree of disease activity in SLE, in particular (7). Ethnicity, therefore, could be the underlying factor accounting for discrepant perception of disease activity among patients with SLE. It is quite possible, however, that such a discrepancy between patients and physicians may not be related to ethnicity per se, but rather result from the type of clinical manifestations present or from other patient-related attributes, such as their psychosocial background, which may or may not be related to ethnicity. This discrepant perception of disease activity and its possible causes has been examined in a primarily Caucasian SLE cohort (5, 6). We are now reporting observations encompassing a multiethnic (Hispanic, African American, and Caucasian) SLE cohort from a longitudinal study of disease outcome.
PATIENTS AND METHODS
Patients were from the Lupus in Minority populations: Nature versus nurture (LUMINA) cohort, a longitudinal study of outcome in SLE, with centers in Alabama and Texas. The constitution of this cohort, as well as the baseline sociodemographic, clinical, immunologic, immunogenetic, psychosocial, behavioral, and overall functioning data have been previously published (2–4, 8–10). Three hundred LUMINA patients were included in this study (81 Hispanic, 119 African American, and 100 Caucasian).
Patients were enrolled and followed at the Clinical Research Center (CRC) at each of the 3 participating institutions: The University of Alabama at Birmingham, The University of Texas Houston-Health Sciences Center, and the University of Texas Medical Branch in Galveston, Texas. Study visits consisted of a physical examination by a study physician, administration of questionnaires, and an interview by the center study coordinator. Followup visits occurred initially at 6 months and yearly thereafter. For the most part, study visits for each patient were conducted by the same study physician.
Variables for the sociodemographic, clinical, immunologic, immunogenetic, psychological, behavioral, and functioning domains were obtained at study visits. Only the variables of particular interest for this study will be described. Disease activity was measured with the Systemic Lupus Activity Measure (SLAM). The SLAM is a validated instrument consisting of various clinical manifestations that are characteristic of organ systems affected by SLE, grouped by domains (11). A separate section scores laboratory tests including urine sediment, hematocrit, erythrocyte sedimentation rate, and serum creatinine, as well as white blood cell, lymphocyte, and platelet counts. The severity of each of these manifestations or abnormal laboratory tests is graded on a scale of 0–3 with the score indicating magnitude or degree of severity. As per the SLAM, patients are asked to rate their level of disease activity over the preceding month on a 10-cm anchored visual analog scale (VAS) where 0 is no disease activity and 10 is the most disease activity possible. The study physician rates the patient's level of disease activity on a separate VAS only after completing the patient's clinical evaluation and after all laboratory data are available (SLAM global VAS scores, herein). Given the fact that these laboratory tests are readily available, the SLAM including the physician's VAS is completed within 24 hours of the visit in the vast majority of patients. Study physicians were trained in the application of the SLAM (and other instruments) prior to study start, and a recalibration exercise was given each subsequent year.
To quantify the differences in the patients' and physicians' assessment of disease activity, the construct called “discrepancy” was computed from the patient and physician SLAM global VAS scores. This discrepancy can be in the positive direction if the patient SLAM global VAS score is greater than the physician's, or negative if otherwise. A positive discrepancy indicates the patient perceives greater disease activity than the physician, and a negative indicates the physician perceives greater disease activity than the patient.
Social support was measured with the Interpersonal Support Evaluation List (ISEL), a 40-item instrument composed of 4 subscales that provide a total summary measure; higher scores indicate better social support (12). Abnormal illness-related behaviors were measured with the Illness Behavior Questionnaire (IBQ), a 62-item questionnaire that ascertains how patients cope with their disease; higher scores indicate more abnormal illness-related behaviors (13). The construct of learned helplessness was evaluated by the Rheumatology Attitude Index, a validated instrument containing 15 items scored individually; a higher score indicates a higher degree of helplessness (14, 15). Self-perceived functioning was assessed with the Medical Outcomes Study Short Form-36 (SF-36), a validated instrument which yields a total score and 2 summary measures, the mental and physical component summary measures or MCS and PCS, respectively (16).
In order to determine the general direction of the discrepancy construct, its frequency distribution was first examined. Next, the frequency distribution of selected variables from the sociodemographic, clinical, and behavior and psychological domains for the 3 ethnic groups were examined. The association between the different components of the SLAM, the psychological and behavioral variables, and the discrepancy score was then examined; the association with the number of visits per patient was also examined. Finally, SLAM specific clinical manifestations, ethnicity, self-perceived functioning, social support, abnormal illness-related behaviors, and helplessness were entered into a stepwise linear regression model to examine the independent contribution of these variables to discrepancy. In both univariable and multivariable analyses, an adjustment for the physician SLAM global VAS score was done in order to eliminate the ceiling and floor effects that these scores may have exerted (that is, if a physician SLAM global was either 10 or 0, no discrepancy could have been observed). All analyses were done with SAS Statistical Software (SAS, Carey, NC).
Table 1 depicts selected socioeconomic demographic, and clinical variables in all patients studied. In general, Caucasian patients were older and had less active disease at diagnosis and at entry into the study. They also tended to have better social support and more appropriate illness-related behaviors. These data are consistent with data we have previously reported for our LUMINA cohort (2–4).
|Variable||Hispanic n = 81||African American n = 119||Caucasian n = 101||P*|
|Age, years ± SD||32.9 ± 12.3||35.0 ± 11.8||40.7 ± 14.2||0.0001|
|Sex, % women||92.6||90.8||82.2|
|Disease duration, months ± SD||57.9 ± 28.3||52.6 ± 31.2||54.7 ± 33.6|
|Disease onset, % acute||57.5||40.7||34.3||0.0065|
|SLAM (T0), mean ± SD||10.6 ± 6.0||11.0 ± 5.9||8.4 ± 4.0||0.0010|
|SLAM (TD), mean ± SD||13.0 ± 6.1||13.5 ± 7.0||9.6 ± 5.8||<0.0001|
|SDI (T0), mean ± SD||0.69 ± 1.2||0.82 ± 1.4||0.52 ± 1.0|
|MCS-SF-36, mean ± SD||44.4 ± 12.8||44.5 ± 11.1||48.5 ± 12.1||<0.0001|
|PCS-SF-36||37.7 ± 12.1||34.5 ± 11.0||34.2 ± 13.2||0.022|
|Learned Helplessness, mean ± SD||40.0 ± 7.6||39.7 ± 6.6||39.7 ± 8.1|
|ISEL, mean ± SD||7.5 ± 1.9||8.0 ± 1.6||8.5 ± 1.7||0.0008|
|IBQ, total score||19.2 ± 7.1||17.9 ± 7.7||14.5 ± 6.7||<0.0001|
The frequency distribution of the discrepancy scores for all patients and by ethnic group is depicted in Figure 1. If a discrepancy greater than 1 cm is considered clinically relevant, over half the patients (58%) exhibited a discrepancy. Discrepancy occurred more frequently in the positive (40%) than in the negative (18%) direction.
Table 2 shows the mean discrepancy scores as a function of ethnic group; the discrepancy score was higher in the Caucasian group than in non-Caucasians; in turn, the score was higher in the Hispanic group than in the African American group. These differences were highly significant (P < 0.0001).
|Ethnic group||Mean ± SD|
|Hispanic||0.93 ± 0.31|
|African American||0.84 ± 0.26|
|Caucasian||1.61 ± 0.28|
The relationship between specific SLAM domain scores, helplessness, social support, abnormal illness-related behaviors, self-perceived physical and mental functioning, and the discrepancy scores are shown in Tables 3 and 4. In general, there was a positive association between subjective SLAM manifestations and discrepancy; on the other hand, a negative association was observed between discrepancy and objective SLAM findings, particularly laboratory manifestations. There was no association between the discrepancy score and the number of visits.
|Specific SLAM domains||r||P†|
|Abnormal laboratory findings||−0.17||0.0050|
|ISEL total score||−0.17||0.0048|
|Abnormal Illness-related Behaviors||0.20||0.0008|
The results of the multivariable regression analyses are shown in Table 5. Variables retained in the model included abnormal laboratory findings, poor self-perceived functioning, physician SLAM global VAS score, and joint involvement as per SLAM; the direction of the predictor variables can also be appreciated in Table 5. In short, articular involvement and poor self-perceived functioning were positively associated with discrepancy and abnormal laboratory findings were associated negatively. Ethnicity was not retained in the model. These variables explained 31% of the variance.
|Variable*||F||P||R2 (partial)||R2 (total)|
|Physician SLAM global VAS score||−28.6||<0.0001||8.63|
|SLAM, articular involvement||−26.7||<0.0001||8.84|
|SLAM abnormal laboratory findings||6.3||0.0128||1.62|
VAS scores are widely used in both clinical and research settings to evaluate self-perceived clinical manifestations such as pain, fatigue, or overall disease activity. As opposed to a categoric rating, a VAS offers the advantage of having a continuous scale, which can detect very small changes for the feature being evaluated (17). Of all available instruments to measure disease activity in SLE, only the SLAM includes a VAS. It has been suggested that patient and physician VAS are so strongly correlated that one may be used as a proxy for the other (18). However, studies conducted in a Canadian SLE cohort have shown that a sizable proportion of patients either under- or over-estimate their degree of disease activity when compared with their physicians' assessments (5, 6).
Our data support the fact that patients and physicians may rate SLE disease activity differently. Rather than suggest that either the patient or the physician SLAM global VAS score is invalid, we believe that patients and physicians rate them differently for a variety of reasons, as discussed below. The fact that the physician scored the VAS only after the laboratory data were available does not invalidate these scores, given that only a short time elapsed between the study visit and the completion of the SLAM.
Patients appear to weigh disease activity subjective manifestations or poor self-perceived functioning (physical and mental) as more relavant when compared with physician's assessments. Conversely, physicians weigh more objective manifestations, particularly abnormal laboratory findings. It is of interest that negative discrepancy of patients relative to physicians occurs when there are abnormal laboratory findings, independent of ethnicity. Because the disease expresses itself differently among ethnic groups, with non-Caucasians having more severe disease, our initial impression was that this was an ethnic-specific finding. Preliminary analyses performed with ∼50% of the study population indicated that at the intake or baseline visit, Caucasian patients were more likely than non-Caucasian patients to judge their disease as being more active than it was judged by the study physicians (19); subsequent analyses, however, failed to support this finding (20). Only when examining the multivariable models in detail and with more subjects, were we able to sort out the role of ethnicity versus the role of disease manifestations (20). Physicians may need to work harder to help patients become aware of their laboratory results and what they mean. Patients may lack the knowledge necessary to interpret them adequately in relation to disease activity. Our data, taken together with the data from the Canadian studies, further emphasize the fact that laboratory abnormalities, whose significance patients may not appreciate, and self-perceived functioning, whose significance physicians may not appreciate, rather than ethnicity, explain our findings.
In conclusion, it cannot be assumed that patients and physicians perceive disease activity similarly at any given time. Physicians may judge disease activity based on their overall experience with SLE whereas, unless they know of many other patients with SLE, it is rare for patients with SLE to be able to make a similar comparison. Although the impact of this discordant perception of disease activity cannot be determined from the above analyses, it can be conjectured that patients who do not realize the significance of the presence of different clinical manifestations, laboratory findings included, may be less likely to follow a given therapeutic regimen.
Current LUMINA investigators and staff:
At the University of Alabama at Birmingham: Graciela S. Alarcón, MD, MPH, Alfred A. Bartolucci, PhD, Holly M. Bastian, MD, MSPH, Barri J. Fessler, MD, Gerald McGwin Jr., MS, PhD, Jeffrey M. Roseman, MD, PhD, MPH, Martha L. Sanchez, MD, MPH, and Ellen Sowell, AA.
At the University of Texas-Houston Health Science Center: John D. Reveille, MD, Alan W. Friedman, MD, Chul Ahn, PhD, Jo McLain, BSN, RN, Kim Jordan, BS, Rudyard Lanete, BA, and Robert Sandoval, BA.
At the University of Texas Medical Branch at Galveston: Bruce A. Baethge, MD, Shrilekha Sairam, MD, and Sonya E. Hunnicutt, MS.
The authors express their gratitude to previous LUMINA staff at the University of Texas Houston-Health Science Center and the University of Alabama at Birmingham (Gastón Benavides, BA, and Allan Chiunda, MPH, respectively) for their assistance in the conduct of study visits, to Ella Henderson for her assistance in the preparation of this manuscript, and to our patients without whom this study would not have been possible.
- 4the LUMINA Study Group. Systemic lupus erythematosus in three ethnic groups. I. The effect of HLA class II, C4, and CR1 alleles, socioeconomic factors, and ethnicity at disease onset. Arthritis Rheum 1998; 41: 1161–72., , , , , , et al. for
- 7Guidelines for culturally appropriate health care. In: HarwoodA, editor. Ethnicity and medical care. Cambridge: Harvard University Press; 1981. p. 482–508..
- 10the LUMINA Study Group. Systemic lupus erythematosus in three ethnic groups. V. Acculturation, health-related attitudes and behaviors and disease activity in Hispanic patients from the LUMINA cohort. Arthritis Care Res 1999; 12: 267–74., , , , , , for
- 12Measuring the functional components of social support. In: SarasonI, SarasonBR, editors. Social support: theory, research, applications, Boston: Martinus Nijhoff; 1985 p. 73–94., , , .
- 13Illness Behavior Questionnaire (IBQ). In: CorcoranK, FischerJ, editors. Measures for clinical practice: a source book. New York: Free Press; 1996. p. 182–5., .
- 16Comparison of methods for the scoring and statistical analysis of SF-36 health profile and summary measures: summary of results from the Medical Outcomes Study. Med Care 1995; 33 Suppl 14: AS264–79., , , , , .
- 19Discrepant perception of disease activity among patients with SLE from three different ethnic groups [abstract]. Arthritis Rheum 1995; 38: S386., , , , , , et al.
- 20Absence of renal disease but not ethnicity accounts for discrepant perception of disease activity among lupus patients from different ethnic groups. Arthritis Rheum 1997; 40: S160., , , , , .