Sleep Disturbance in Patients with Rheumatoid Arthritis: Evaluation by Medical Outcomes Study and Visual Analog Sleep Scales
FREDERICK WOLFE, KALEB MICHAUD, and TRACY LI
Objective. Except for some polysomnography studies, there have been no large quantitative studies of sleep disturbance (SD) in rheumatoid arthritis (RA). SD has taken on new importance with the observation that etanercept and infliximab reduce daytime sleepiness, and patient groups indicate that sleep is an important issue.
Methods. We evaluated 8676 patients with RA and a comparison group of 1364 subjects with non-fibromyalgia, noninflammatory disorders (NID) using the Medical Outcome Study (MOS) sleep questionnaire, including 2 MOS sleep problem indexes (SPI-I, SPI-II) and the MOS SD scale. In addition, patients completed a visual analog scale (VAS) sleep disturbance scale (SDS).
Results. The scales had similar mean values: SPI-I 35.4 (19.4), SPI-II 36.0 (19.1), SDS 35.0 (24.7), and VAS sleep 36.1 (29.7), and the values for the MOS scales exceeded population norms by 25% (VAS by 42%). In multivariable analyses SD was primarily determined by pain and mood. Patients receiving anti-tumor necrosis factor (TNF) did not have less abnormal sleep scores. SD was comparable in RA and NID. The VAS scale was more strongly associated with RA clinical variables than the MOS scales; however, the distributional characteristics of the scales differed, with the VAS scales capturing more extreme values. The standard error of the measurement (SEM), which is related to minimal (important) change, was SPI-I 9.0, SPI-II 7.3, SDS 9.6, and VAS sleep 10.4.
Conclusion. SD is increased in RA, and 25% to 42% of SD can be attributed to RA. SD is linked to pain, mood, and disease activity. SD is slightly greater in women and is less with increasing age. All scales appear to be valid in RA, with minimal differences in SEM. (First Release Sept 1 2006; J Rheumatol 2006;33:1942-51)
Key Indexing Terms:
From the National Data Bank for Rheumatic Diseases, University of Kansas School of Medicine, Wichita, Kansas; Stanford University, Stanford, California; and Bristol-Myers Squibb, Princeton, New Jersey, USA.
Supported by a grant from Bristol-Myers-Squibb.
F. Wolfe, MD, National Data Bank for Rheumatic Diseases, University of Kansas School of Medicine; K. Michaud, MS, National Data Bank for Rheumatic Diseases, University of Kansas School of Medicine and Stanford University; T. Li, PhD, Outcomes Research, Bristol-Myers-Squibb.
Address reprint requests to Dr. F. Wolfe, National Data Bank for Rheumatic Diseases, Arthritis Research Center Foundation, 1035 N. Emporia, Suite 230, Wichita, KS 67214. E-mail: email@example.com
Accepted for publication May 5, 2006.