Self-Reports and Clinician-Measured Physical Function Among Patients With Low Back Pain: A Comparison C. Ellen Lee, MS, PT, Maureen J. Simmonds, PhD, PT, Diane M. Novy, PhD, Stanley Jones, MD ABSTRACT. Lee CE, Simmonds MJ, Novy DM, Jones S. Self-reports and clinician-measured physical function among patients with low back pain: a comparison. Arch Phys Med Rehabil 2001;82:227-31. Objective: To determine the relationships among self-re- ported activity limitation and clinician-measured functional performance tests. Design: Case series survey. Setting: A referral-based orthopedic spine clinic in Houston, TX. Patients: Eighty-three patients (48 women, 35 men) with low back pain (LBP). Interventions: The Roland-Morris Disability Questionnaire (RMDQ) and a physical performance test (PPT) battery. Main Outcome Measures: Self-reported activity limitation (eg, walking, bending, getting out of chair, putting on sock, doing heavy jobs) was assessed by the RMDQ. Clinician- measured functional performance was assessed with the PPT, a battery comprised 6 tests: lumbar flexion range of motion, a 50-foot walk at fastest speed, a 5-minute walk, 5 repetitions of sit-to-stand, 10 repetitions of trunk flexion, and loaded reach task (patients reached forward while holding a weight weighing 5% of their body weight). Results: Pearson’s product-moment correlations between total RMDQ score and each of the performance tests ranged from .29 to .41. Point biserial correlations between individual RMDQ items and their corresponding performance tests were slightly lower, ranging from .20 to .33. Conclusion: There were moderate correlations between self- reported activity limitation and corresponding clinician-mea- sured performance tests. The unique perspective each method provides appears to be useful for a comprehensive understand- ing of physical function in patients with LBP. Key Words: Activities of daily living; Low back pain; Rehabilitation. © 2001 by the American Congress of Rehabilitation Medi- cine and the American Academy of Physical Medicine and Rehabilitation P HYSICAL FUNCTION IS DEFINED as the sensorimotor performance of an individual that includes fundamental and complex activities of daily living (ADLs). 1 Traditional methods of clinical assessment have given little attention to this functional aspect of individuals with low back pain (LBP). However, with recognition of the limitations of traditional impairment measures (eg, range of motion [ROM], strength) in predicting functional limitations, 2-11 there has been a shift recently toward using assessment of physical function as com- plementary to traditional impairment measures. 12 One method of physical function assessment involves the use of patient self-reporting of activity limitation by using standardized ques- tionnaires. These self-reports probe patients’ perceptions of their abilities to perform certain ADLs. Numerous physical function questionnaires have been developed for individuals with LBP since the 1980s. 13-24 One of the advantages of using these questionnaires is their ease of administration. They also sample a wide variety of activities. The Roland-Morris Dis- ability Questionnaire (RMDQ) is among the most widely used and evaluated questionnaires. 25,26 Studies have shown the va- lidity, adequacy of interrater agreement and test-retest reliabil- ity, and high internal consistency of the RMDQ. 16,26-29 RMDQ scores have high correlations with clinical findings or symp- toms 29 as well as with other established questionnaires and patient self-ratings of physical function. The RMDQ also is responsive to change in physical function when the initial score is taken into account. 30-32 Nevertheless, a questionnaire method can be influenced by patients’ perception of their abilities to perform activities based on their attitudes, 33 expectancies of pain or reinjury, 34,35 and psychologic distress levels. 35,36 Thus, there may be a discrepancy between patients’ self-reports of activity limitation and actual physical function. Another method of measuring physical function is clinician- measured performance tests. This approach uses measured tasks such as load lifting 37-40 and pulling and pushing weights. 39 However, few of these measures have documented psychometric properties, and they are often too expensive and time consuming to perform. In addition, the tasks are usually tested in a constrained manner that cannot be generalized to functional daily tasks. In recent years, batteries that assess general ADLs have been developed for this population. 41,42 The battery developed by Simmonds et al 42 includes a series of tasks in which the time taken or distance reached or walked is measured. The timed tasks included a 50-foot walk at fastest speed, repeated trunk flexion, and repeated sit-to-stand tasks. The distance tests included a 5-minute walk and a loaded reach task (ie, patients reach forward while holding a weight weigh- ing 5% of their body weight). These tests showed excel- lent interrater reliability ([intraclass correlation coefficients] ICC 1,1  .95) and good test-retest (2wk) reliability (ICC 1,1 = .76 –.91). They also had good discriminant validity between groups of healthy subjects and subjects with LBP (F 10,63 = 3.25, p = .002). 42 Similar to patient self-reports of activity limitation, clinician-measured performance tests are selected samples of physical function. They could also be influenced by psychosocial factors. However, researchers have found that the impact of certain psychosocial factors on clinician-measured From the School of Physical Therapy, Texas Woman’s University (Lee, Sim- monds); Department of Anesthesiology, University of Texas/Houston Medical School (Novy); and Spine Care Southwest (Jones); Houston, TX. Accepted February 17, 2000. Supported in part by Texas Woman’s University Research Enhancement Award (grant no. 10-0131382), the National Institute of Health EARDA Research Pilot Project Grant (grant no. 1997-8), and the National Institute of Disability and Reha- bilitation Research Training Grant (grant no. 19-1180064). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the author(s) is/are associated. Reprint requests to Maureen J. Simmonds, PhD, PT, School of Physical Therapy, Texas Woman’s University, 1130 M.D. Anderson Blvd, Houston, TX 77030; e-mail: hf_simmonds@twu.edu. 0003-9993/01/8202-6064$35.00/0 doi:10.1053/apmr.2001.18214 227 Arch Phys Med Rehabil Vol 82, Februry 2001