A Rheumatology-Specific Informatics-Based Application With a Disease Activity Calculator

Transcription

1 Arthritis & Rheumatism (Arthritis Care & Research) Vol. 61, No. 4, April 15, 2009, pp DOI /art , American College of Rheumatology ORIGINAL ARTICLE A Rheumatology-Specific Informatics-Based Application With a Disease Activity Calculator DEBORAH S. COLLIER, 1 JONATHAN KAY, 1 GREG ESTEY, 2 DOMINIC SURRAO, 2 HENRY C. CHUEH, 1 AND RICHARD W. GRANT 1 Objective. To design a rheumatology-specific tool with a disease activity calculator integrated into the electronic medical records (EMRs) at our institution and assess physicians attitudes toward the use of this tool. Methods. The Rheumatology OnCall (ROC) application culls rheumatology-pertinent data from our institution s laboratory, microbiology, pathology, radiology, and pharmacy information systems. Attending rheumatologists and rheumatology fellows accessed the ROC and disease activity calculator during outpatient visits at the time of the clinical encounter. Results. During the 12-week study period, 15 physicians used the ROC application and the disease activity calculator during 474 and 429 outpatient clinic visits, respectively. In weekly survey responses, physicians reported that use of the ROC interface improved patient care in 140 (78%) of 179 visits, and that the Disease Activity Score in 28 joints (DAS28) results at the time of the visit would not have changed patient management in 157 (88%) of these, although seeing a trend in DAS28 was useful in 149 (96%) of 156 visits. At the study s conclusion, most physicians reported that the ROC application was useful (11 of 12 physicians) and that seeing a trend in DAS28 improved daily patient care (12 of 13 physicians). Conclusion. The ROC application is useful in daily rheumatologic care, and the disease activity calculator facilitates management of patients with rheumatoid arthritis. However, widespread acceptance and use of such tools depend upon the general acceptance of and access to EMRs in the clinical setting. The utility of the disease activity calculator may be limited by the lack of available acute-phase reactant results at the time of the clinical encounter. INTRODUCTION Dr. Collier s work was supported by an unrestricted educational grant from Centocor. Dr. Kay s work was supported by an unrestricted educational grant from Pfizer. 1 Deborah S. Collier, MD, Jonathan Kay, MD, Henry C. Chueh, MD, Richard W. Grant, MD, MPH: Massachusetts General Hospital and Harvard Medical School, Boston; 2 Greg Estey, EdM, Dominic Surrao, MS: Massachusetts General Hospital, Boston. Address correspondence to Deborah S. Collier, MD, Massachusetts General Hospital, Yawkey Center, Suite 2C-2100, 55 Fruit Street, Boston, MA dscollier@ partners.org. Submitted for publication April 10, 2008; accepted in revised form December 12, The emergence and enhancement of electronic medical records (EMRs) has been driven by private and public sector initiatives with the goals of improving individual patient care, reducing medical errors, reducing health care costs, increasing physician access to information for medical decision making, and facilitating communication between providers. EMRs are being used with increasing frequency in medical practices to make relevant patient data more readily available at the time of the patient encounter. However, data review during a patient visit can be time consuming and cumbersome, because various pieces of data are stored in many different sections of the EMR. Also, EMRs do not routinely have the functionality to calculate rheumatology-specific measures such as the Disease Activity Score in 28 joints (DAS28), which is used to monitor disease activity in patients with rheumatoid arthritis (RA). In the US, physician use of EMRs is not universal and is more prevalent in large group practices and academic settings (1). According to a survey of 2,523 physicians nationwide in 2003, EMR use across all practice environments was 6% in the year 2000, increased to 8% in 2002, and increased to 10% in 2003 (2). In 2004, compiled published survey results indicated that 24% of US physicians used an EMR in the ambulatory setting (3). However, a recent national survey of 2,758 physicians in the ambulatory setting across all practice specialties and practice types reported that although 13% of physicians had a basic EMR, only 4% had an extensive, fully functional EMR (1). The adoption of an EMR yields potential benefits but also may impede practice in ways that are both general to all of medicine and specific to rheumatology. Benefits include improved access to patient information and gen- 488

2 Design and Assessment of a Rheumatic Disease Activity Calculator 489 Figure 1. The Rheumatology OnCall (ROC) Web page is accessed via an icon on the electronic medical records summary page. Displayed data is aggregated from multiple data sources. eral medical knowledge (given the complex nature of many rheumatologic diseases), ease of prescribing, and ease of complying with national quality performance indicators that are increasingly driving physician practice and reimbursement (4). Barriers to the universal use of EMRs include cost of software purchase and lack of inhouse technical support, the need to integrate multiple sites, fewer potential savings for solo practitioners or smaller practice groups, and decreased provider efficiency due to the time-consuming nature of data entry and review (2). Additional barriers may include a lack of clear return on investment, provider resistance to the adoption and adaptation of an EMR to rheumatology practice, and the use of macros or templates. Decreased provider efficiency is a concern not only because it could lead to a reduction in provider reimbursement, but also because it might otherwise occupy the time of rheumatologists when these subspecialists are in short supply. It is predicted that demand for the services of adult and pediatric rheumatologists will increase by 40% and 37%, respectively, and that the demand for rheumatologists will exceed supply by 2,576 rheumatologists in the year 2025 (5). EMR products designed for general medical practice can be cumbersome to use in the routine care of patients with rheumatic diseases because rheumatologists need to rapidly review a panel of information that is often stored in many different parts of the EMR, including notes and data from laboratory, radiology, pharmacy, and pathology repositories. In addition, rheumatologists may seek to collect information that is not typically found in an EMR, such as pain scores, joints counts, functional measures, and other data that are used to calculate disease activity indices such as a DAS28. These data are critical because tight disease control through quantitatively driven management improves outcomes in rheumatic diseases (6 8). We developed Rheumatology OnCall (ROC) in order to facilitate physician access to multiple data items relevant to the patient encounter but located in various separate parts of the medical record. ROC also facilitates calculation of the DAS28 for patients with RA and promotes efficient, quantitatively driven, and quality rheumatologic clinical practice. ROC is a rheumatology-specific, informatics-based application that is integrated into the Longitudinal Medical Record (LMR), which is a Certification Commission for Healthcare Information Technology certified EMR internally developed by our institution. The ROC provides a potential platform for quantitatively driven management tools for numerous rheumatic diseases. Whereas this initial version incorporates a DAS28 calculator, future iterations will be enriched by the addition of quantitatively driven management tools for RA other than the DAS28 and for rheumatic diseases other than RA. MATERIALS AND METHODS Application development and features. We collaborated with rheumatologists at the Massachusetts General Hospital and Brigham and Women s Hospital and with software engineers at Massachusetts General Hospital to develop the ROC application with disease activity calculator to access all relevant data from our hospital s information systems (laboratory, microbiology, pathology, radiology, and pharmacy data) to document detailed joint examinations and to facilitate DAS28 calculation. ROC was built as an extension of the LMR, employing the OnCall EMR application framework developed by the Laboratory of Computer Science at Massachusetts General Hospital using Microsoft Active Server Pages (Microsoft,

3 490 Collier et al Trended clinical and laboratory data dashboard. Twelve-month results of erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) level, plasma creatinine level, aspartate aminotransferase level, hemoglobin level, and DAS28 scores are each graphed separately (Figure 2). The compact grouping of graphs allows the clinician to quickly assess trends and facilitates the tracking of disease activity, organ involvement, and monitoring for potential medication related adverse effects. Disease activity calculator. The DAS28 is a composite score based on measures of swollen and tender joints, an acute-phase reactant (ESR or CRP), and patient-reported assessment of global health: DAS28(ESR) 0.56 sqrt(t28) 0.28 sqrt(sw28) 0.70 ln(esr) GH DAS28(CRP) 0.56 sqrt(t28) 0.28 sqrt(sw28) 0.36 ln(crp 1) GH 0.96 Figure 2. The Rheumatology On Call Web page consolidates and tracks data from multiple data sources in the electronic medical record. Vashon, WA), a Web applications language, and SQL Server databases. The OnCall framework enables custom configuration of discipline-specific modules to support tailored visual data summaries and data collection capabilities that are unique to a particular specialty. Other specialty-tailored OnCall installations at Massachusetts General Hospital include those for diabetes, cardiology, and the human immunodeficiency virus/acquired immunodeficiency syndrome. The ROC application is accessed through an icon on the patient summary page in the EMR (Figure 1). The single Web page format includes a dashboard view of clinical and laboratory data relevant to rheumatic disease, medicationrelated adverse effects graphed over time, a DAS28 calculator and trend reporter, tables of rheumatology-specific and general laboratory results, a list of all hyperlinked radiographic studies, a current medication list, and a medication history bar graph (Figure 2). where sqrt square root, T tender, SW swollen, ln natural logarithm, and GH global health. This quantitative measure of RA disease activity has been validated for use in clinical trials (9) and, when used as a measure to achieve tight disease control in patients with RA, has resulted in improved physical function and decreased radiographic progression (6 8). In clinical practice, use of the DAS28 is limited by the relative difficulty of its calculation and the typical unavailability of recent acute-phase reactant test results at the time of the clinical encounter (10). We set out to develop a disease activity calculator that would facilitate automated DAS28 calculation within 24 hours of the clinical encounter, and that would also be a generic platform for calculation of other disease activity indices such as the Clinical Disease Activity Index (CDAI). Our disease activity calculator facilitates documentation of the tender and swollen joint examinations, documentation of the patient s global health assessment using a visual analog scale (VAS), and calculation and reporting of the DAS28 score. The onscreen (10-cm) VAS is invisibly sectioned into 1-mm increments, and the patient or physician uses the computer mouse to mark the location on the line that corresponds to the patient s assessment of his or her general health in the last week. Then a joint homunculus with radio buttons is used by the clinician to label each of 66 joints as normal, swollen, tender, swollen and tender, replaced, or repaired (Figure 3). When the joint count is saved with the global health assessment data entered, a text version of the joint examination is generated in sentence format and appears in a text box. This text can then be copied and inserted into a clinical progress note in the EMR. The ESR and CRP level are ordered at the time of the visit. Each night, DAS28 scores are calculated and recorded in ROC for all patients for whom the disease activity calculator was used on the previous day, and for whom an ESR or CRP level was ordered within 7 days. Cumulative DAS28 ESR and DAS28 CRP scores are listed in a table

4 Design and Assessment of a Rheumatic Disease Activity Calculator 491 Figure 3. Radio button-labeled joints and an on-screen visual analog scale allow clinician and patient data input for Disease Activity Score in 28 joints score calculation. next to the disease activity calculator with hyperlinks to each previous joint count, and are plotted as separate lines on the dashboard graph of DAS28 scores. An automated correspondence is sent to each clinician every week, containing a table of the previous week s calculated DAS28 scores for his or her patients with RA. Rheumatology-relevant and general laboratory results. The most recent rheumatology-relevant laboratory results are culled from hematology, chemistry, immunology, and pathology clinical data repositories and displayed in tabular format in ROC (Figure 2). The 2 most recent results of relevant general laboratory tests are displayed in a separate table. Radiographic studies. A list of all radiographic studies performed within our institution for a given patient is shown in tabular format. The name of each study is hyperlinked to the full-text radiology report. Medication list and history. A list of all medications with the dose and date of the most recent refill is retrieved from the pharmacy database within the EMR. The accuracy of these data depends upon clinicians entering and regularly updating the medication list. A bar graph depicts all medications prescribed for rheumatologic care, and the time period during which each was prescribed. When applicable, medications included in the bar graph include nonsteroidal antiinflammatory drugs, glucocorticoids, disease-modifying antirheumatic drugs (DMARDs), and biologic agents. ROC use and utility study design. We conducted a 12- week prospective cohort study of attending rheumatologists and rheumatology fellows at our institution who used the ROC and disease activity calculator as part of routine rheumatologic care and responded to weekly and end-ofstudy surveys on the use and utility of the ROC. Study subject eligibility. All attending rheumatologists and rheumatology fellows providing at least 1 half-day session of outpatient clinical care per week at Massachusetts General Hospital or Brigham and Women s Hospital were eligible for enrollment in the study. Data collection. Over the 12-week study period, participating physicians accessed the ROC through the EMR during the course of routine care of patients with varying rheumatic conditions. Each week, each study physician who had accessed the disease activity calculator at least once during the previous week received an automated correspondence that contained a summary table listing the proportion of DAS28 predictions that were in or out of range during the previous week and cumulatively up to that point in the study; and survey questions regarding the utility of the ROC and disease activity calculator in the care of 3 randomly selected patients who had been seen during the previous week. At the end of the study, physicians completed a Web-based closeout survey pertaining to the utility of the ROC and disease activity calculator. RESULTS Study population. Of 47 attending rheumatologists and rheumatology fellows at Massachusetts General Hospital and Brigham and Women s Hospital who saw patients during at least 1 half-day outpatient clinic session, 25 agreed to participate in the study. Of the 25 rheumatologists who agreed to participate, 15 (9 attendings, 6 fellows) accessed the ROC application and used the disease activity calculator at least once over the course of the 12-week study. Of those 15 rheumatologists, 11 submitted at least 1 completed weekly survey, and 13 responded to the Webbased closeout survey. Data capture. During the 12-week study period, the ROC was accessed 656 times for 541 unique patients. The number of times that the ROC was accessed by each physician during the 12-week study period ranged from 1 to 256 (Figure 4A). The disease activity calculator was accessed during 474 patient visits, and an ESR or CRP was available to calculate a DAS28 score after 398 of those visits. For 45 (9.5%) of the 474 visits during which a joint examination and general health VAS were documented, the patient was reported as not having RA. In 32 (6.8%) of the 474 visits, an ESR or CRP level had not been obtained within 7 days of the visit, and therefore a DAS28 could not be calculated.

5 492 Collier et al Figure 4. A, physician use and B, physician-perceived utility of the Rheumatology OnCall (ROC) application. Physicians 1 11 in panel A correspond to physicians 1 11 in panel B. Weekly survey responses. Eleven of the 15 study physicians completed at least 1 weekly survey (Table 1). In the cumulative responses to the weekly surveys, physicians reported that use of the ROC interface improved patient care in 140 (78%) of 179 patient visits. In the cumulative responses of individual physicians, use of the ROC interface was reported to improve patient care in % of patient visits (Figure 4B). Physicians indicated that DAS28 results provided after the clinic visit would not have changed patient management at the time of the visit in 157 (88%) of 179 visits. However, the use of the disease activity calculator affected patient management in 134 (75%) of 179 visits, and the graphs of trended DAS28 were useful during 147 (96%) of 156 visits for which a previous DAS28 had been reported. In these weekly surveys, responding physicians reported that discordance between calculated and physician-predicted DAS28 ranges was most often due to the patient s global health score or acute-phase reactants being elevated for reasons other than RA. Closeout survey responses. At study conclusion, 13 of 15 study physicians responded to a followup survey. Twelve of the 13 who responded reported that use of the ROC improved patient care, that they would continue to use the ROC application in daily patient care, and that Table 1. Weekly physician survey results* Weekly survey question Yes No Did use of the ROC application improve patient care? Would having the recent DAS28 score at the time of the visit have changed your management of this patient? Did the process of performing joint count and patient global assessment change your management of this patient? Is seeing a trend in DAS28 score useful? 140 (78) 39 (22) 22 (12) 157 (88) 134 (75) 45 (25) 147 (96) 9 (4) * Values are the number (percentage). ROC Rheumatology On- Call; DAS28 Disease Activity Score in 28 joints. seeing a trend in DAS28 was useful in the daily care of patients with RA (Table 2). Eight of the 13 physicians reported that use of the disease activity calculator improved care of their RA patients definitely or somewhat, and 9 of the 13 physicians reported that they would probably or definitely continue to use the disease activity calculator in daily patient care. DISCUSSION The ROC is a tool that facilitates patient care and addresses barriers to the regular use of an EMR in rheumatologic practice; the disease activity calculator facilitates rheumatology-specific quantitatively driven management. Use of the ROC expedites data review by presenting data culled from many data repositories within the EMR on a single Web page. Therefore, the rheumatologist is not required to navigate through many different computer screens or windows to review all data that are relevant to decision making in the course of a time-limited patient encounter. The trended laboratory data graphs, DAS28 history table, and medication history chart facilitate review of a patient s prior treatment, encourage quantitatively driven disease management, and can be shared with the patient during the clinical encounter. It is noted that among our study physicians, the frequency of use of the ROC and disease activity calculator varied widely. This may be due to lack of comfort with or perceived inefficiency or value of computer-based documentation and data review. However, the study physicians who used these tools most often found them to be more useful than did physicians who used the tools infrequently (Figure 4). These observations in our study cohort echo those barriers to EMR adoption that have been seen in larger cohorts of EMR users (1,2,5). Advantages of the ROC application for rheumatologists at our institution include its full integration into the EMR, the availability of an integrated disease activity calculator, and the presentation of a concise visual summary of rheumatology-specific data. Data review is expedited by presenting data culled from many data repositories within the EMR on a single Web page. Also, because the medication list and medication history bar graph on the ROC page are

6 Design and Assessment of a Rheumatic Disease Activity Calculator 493 Table 2. End-of-study survey results* Response, n (%) Response, % Question Yes No Definitely Some Not sure Not very much Definitely not Did use of the ROC application improve 12 (92) 1 (8) patient care? Will you continue to use the ROC 12 (92) 1 (8) application in daily patient care? Is seeing a trend in DAS28 score useful? 12 (92) 1 (8) Did use of the DAS28 calculator improve your care of patients with RA? Will you continue to use the DAS28 calculator in daily patient care? * ROC Rheumatology OnCall; DAS28 Disease Activity Score in 28 joints; RA rheumatoid arthritis. generated from the pharmacy database within the EMR, these data are identical to medication data found in other sections of the EMR. This ensures consistency between various parts of the EMR and prevents confusion due to otherwise inevitable discrepancies. However, if medication data, including start and stop dates for each prescription, are entered inaccurately into the EMR, this will result in the transfer of inaccurate information to the ROC medication list and medication history bar graph. If this were to occur, a physician looking at the medication data and the disease activity data presented on the same ROC page might draw incorrect conclusions about the efficacy, lack of efficacy, or toxicity of the inaccurately entered medications. This underscores the importance of carefully and consistently maintaining an accurate medication list. The radio button labeled homunculus facilitates the documentation of swollen and tender joint counts and produces not only tabular documentation of joint counts but also text documenting the specific swollen and/or tender joints. This text version of the specific joints that are swollen and/or tender is saved and can be viewed at later visits by hyperlink to each previous DAS28 score in the table of cumulative DAS28 scores. Although having the ability to click a single button labeled metacarpophalangeal (MCP) joints 2 5 or proximal interphalangeal (PIP) joints 2 5 may ease documentation of the joint examination, it may allow a physician to click this button when not all of the MCP or PIP joints are swollen and/or tender, thereby falsely elevating the joint count. The ability of a DAS28 calculator to quantitate RA disease activity at the time of an outpatient visit is limited when recent acute-phase reactant test results are not available at the time of the patient encounter. To address this issue, we chose to use ESR or CRP values obtained within 7 days of the visit in the DAS28 calculation, automated daily DAS28 calculation, and weekly reporting to each physician. However, whereas ESR values change over the course of weeks following an inflammatory event or medication adjustment, CRP levels change more rapidly over 2 14 days. Therefore, because use of a CRP value obtained within 7 days of the visit may occasionally cause the calculated DAS28 to be either higher or lower than it might have been had a CRP value drawn on the day of the visit been used (11). To facilitate calculation of RA disease activity at the time of each visit without the need for measurement of an acute-phase reactant, the disease activity calculator might be augmented with a clinicianassessed global health VAS to allow calculation of the CDAI (12). However, this index was not yet published at the time that the ROC was developed. In addition to facilitating DAS28 calculation, the ROC application provides a platform for tracking patient information, monitoring disease activity, and following cohorts of patients with various rheumatic diseases. The display of DAS28 trends supports patient-based clinical decision making and complies with national quality performance indicators, and the ROC application architecture also allows for the future addition of disease-specific tracking systems and disease activity calculators for RA (other than the DAS28) and many different rheumatic disease processes. Adding a clinician-assessed global health VAS would allow calculation of the CDAI, a validated measure of RA disease activity that can be calculated at the time of the patient encounter because it does not include the measurement of an acute-phase reactant (12). Other planned enhancements to the application include disease activity calculators for systemic lupus erythematosus (e.g., the British Isles Lupus Assessment Group Index, the Systemic Lupus Erythematosus Disease Activity Index) (13), ankylosing spondylitis (e.g., the Bath Ankylosing Spondylitis Disease Activity Index) (14), and osteoarthritis (e.g., the Western Ontario and McMaster Universities Osteoarthritis Index) (15). Clinical decision support can be implemented by creating automated prompts for DAS28- or CDAI-driven dose adjustment of DMARDs in patients with RA, purified protein derivative (PPD) screening in patients starting DMARD or biologic therapy, medication-appropriate toxicity monitoring, dual x-ray absorptiometry scanning and initiation of calcium, vitamin D and bisphosphonates in patients receiving glucocorticoid therapy, and lipid screening in all patients with systemic inflammatory diseases.

7 494 Collier et al The ROC is an application integrated into the EMR that facilitates data retrieval and documentation for disease management and has the potential to support quality assessment activities, research, and practice improvement initiatives. The ROC with disease activity calculator is a model of a rheumatology-specific tool that facilitates ease and thoroughness of patient care in this technologic era wherein the use of EMRs in routine clinical practice is increasingly accepted and expected. Tracking cohorts of patients with each of the rheumatic diseases, including documentation of disease-specific activity indices, will facilitate clinical and translational research and allow for both prospective and retrospective study of the natural history and treatment of these conditions. AUTHOR CONTRIBUTIONS Dr. Collier had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. Collier, Kay, Estey, Surrao, Chueh, Grant. Acquisition of data. Collier, Kay. Analysis and interpretation of data. Collier, Kay, Estey, Surrao, Grant. Manuscript preparation. Collier, Kay, Grant. Statistical analysis. Collier, Grant. REFERENCES 1. DesRoches CM, Campbell EG, Rao SR, Donelan K, Ferris TG, Jha A, et al. Electronic health records in ambulatory care: a national survey of physicians. N Engl J Med 2008;359: Anderson MR, and the American College of Rheumatology Committee on Rheumatologic Care. Electronic medical records for the physician s office. URL: org/products/coding/03emr_toc.asp. 3. Jha AK, Ferris TG, Donelan K, DesRoches C, Shields A, Rosenbaum S, et al. How common are electronic health records in the United States? A summary of the evidence. Health Aff (Millwood) 2006;25: American College of Rheumatology. Quality measures. URL: 5. Deal CL, Hooker R, Harrington T, Birnbaum N, Hogan P, Bouchery E, et al. The United States rheumatology workforce: supply and demand, Arthritis Rheum 2007;56: Goekoop-Ruiterman YP, de Vries-Bouwstra JK, Allaart CF, van Zeben D, Kerstens PJ, Hazes JM, et al. Clinical and radiographic outcomes of four different treatment strategies in patients with early rheumatoid arthritis (the BeSt study): a randomized, controlled trial. Arthritis Rheum 2005;52: Grigor C, Capell H, Stirling A, McMahon AD, Lock P, Vallance R, et al. Effect of a treatment strategy of tight control for rheumatoid arthritis (the TICORA study): a single-blind randomised controlled trial. Lancet 2004;364: Verstappen SM, Jacobs JW, van der Veen MJ, Heurkens AH, Schenk Y, ter Borg EJ, et al. Intensive treatment with methotrexate in early rheumatoid arthritis: aiming for remission. Computer Assisted Management in Early Rheumatoid Arthritis (CAMERA, an open-label strategy trial). Ann Rheum Dis 2007;66: Prevoo ML, Kuper IH, van t Hof MA, van Leeuwen MA, van de Putte LB, van Riel PL. Validity and reproducibility of self-administered joint counts: a prospective longitudinal followup study in patients with rheumatoid arthritis. J Rheumatol 1996;23: Gardiner PV, Bell AL, Taggart AJ, Wright G, Kee F, Smyth A, et al. A potential pitfall in the use of the Disease Activity Score (DAS28) as the main response criterion in treatment guidelines for patients with rheumatoid arthritis [letter]. Ann Rheum Dis 2005;64: Gabay C, Kushner I. Acute-phase proteins and other systemic responses to inflammation. N Engl J Med 1999;340: Aletaha D, Nell VP, Stamm T, Uffmann M, Pflugbeil S, Machold K, et al. Acute-phase reactants add little to composite disease activity indices for rheumatoid arthritis: validation of a clinical activity score. Arthritis Res Ther 2005;7:R Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang CH, and the Committee on Prognosis Studies in SLE. Derivation of the SLEDAI: a disease activity index for lupus patients. Arthritis Rheum 1992;35: Garrett S, Jenkinson T, Kennedy LG, Whitelock H, Gaisford P, Calin A. A new approach to defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease Activity Index. J Rheumatol 1994;21: Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15: