Relationship of physical examination findings and self-reported symptom severity and physical function in patients with degenerative lumbar conditions.

Chronic low back pain (CLBP) affects at least 20% of people older than age 65 years each year. (1) Back pain in older adults is generally thought to arise from degenerative processes, which are ubiquitous in this population and increase with age. (2,3) The aging spine is often described with pathoanatomical terms such as "spondylosis," "vertebral osteophytosis/osteoarthritis," "spinal stenosis," and "disk disease." The degenerative process is muhifactorial, (4) and the source of pain is often unclear because of weak associations between imaging studies and clinical presentation. (2,3,5,6) The apparent discrepancy among pathology, symptoms, and function complicates patient care.

Physical therapists use physical examination (PE) procedures routinely when making clinical decisions. Yet, limited data are available to guide the clinical decision-making process. (7) Clinicians rely on inductive reasoning, intuition, and evidence to formulate clinical decisions. The goal during any examination is to collect and evaluate data in terms of signs and symptoms that tend to fit a pattern. (8) Classification systems were developed based on this rationale with the hope of enhancing patient care and identifying homogeneous groups. (7,9) By identifying subgroups with similar PE findings, clinical researchers can determine which patients are likely to benefit most from a particular intervention. Flynn et al (10) applied this paradigm to identify PE variables that predict which individuals are likely to benefit (reported reduced low back symptoms) from spinal manipulation. For the PE to be more clinically useful and to guide clinical decisions, we must better understand the relationship between PE findings and the patient's symptom severity and function.

Symptom severity and function can be ascertained from self-report questionnaires and medical history. Yet, the use of self-report measures in the clinic is limited. (11) Identifying the best clinical measures that are associated with a patient's condition severity and function will facilitate the clinical decision-making process. For example, if therapists know which PE procedures are associated with worse severity or function, they may appropriately adjust the prognosis and intervention plan. Waddell and Main's statement in 1984 remains a contemporary issue with regard to PE procedures: "at present, there is no satisfactory or accepted method for assessing the severity of low back disorders." (12(p204))

Physical examination and evaluation procedures have been studied extensively in people with low back pain (LBP). (12-19) Impairment-based PE procedures have yielded weak and conflicting associations between condition severity and functional capabilities in people with acute LBP (13,14) and CLBP. (12,15-19) Methodological issues may largely explain these poor associations. For example, the inclusion of people with varying age ranges and clinical symptoms increases the heterogeneity of the sample and may explain the weak associations between PE procedures and self-reported symptom and function. The use of poorly defined clinical measures and reference standards and poor interexaminer reliability also makes interpretation of the literature difficult. Furthermore, biobehavioral factors (ie, cognitive-perceptual, environmental, and psychophysiological) are thought to strongly affect the clinical presentation of people with CLBP. (20) A better understanding of the relationship between PE procedures and low back symptom severity and function may enhance clinical decision making and may improve patient care.

The current investigation examined older adults with chronic lumbar back pain believed to arise from degenerative processes. The purpose of this study was to examine the relationships between select PE procedures and the patients' self-reported symptom severity and function using a condition-specific and a generic health measure. We hypothesized that PE procedures that reduced the opening of the intervertebral foramen and were related to common clinical presentation (eg, muscle weakness of the lower extremity) would be strongly associated with self-reports of low back symptom severity and function.

Method

We used a cross-sectional design and conducted a secondary analysis of data from 2 clinical trials, a prospective randomized controlled trial (RCT) aimed at determining the effects of a 12-week submaximal bicycle program on CLBP (RCT under review) and the pilot trial of endurance training in patients with degenerative lumbar conditions. (21) The rights of human subjects were protected. Participants were recruited from a spine center affiliated with a large tertiary care hospital. We mailed letters to each patient's physician seeking the physician's permission to enroll the patient, to provide medical clearance, and to gather data on the physician's clinical impression of the patient's low back condition. This letter outlined the inclusion and exclusion criteria. To enroll a somewhat homogeneous sample of patients with CLBP due to degenerative changes in the spine, patients were enrolled if they had LBP of [greater than or equal to] 6 months duration, were 55 years of age or older, and reported that their LBP, buttock pain, or lower-extremity pain was made worse with lumbar extension. Patients were eligible only if the spine center physicians felt that their symptoms were degenerative. These inclusion criteria are identical to those used by Stucki et al (22) in the development of the Lumbar Spinal Stenosis (LSS) questionnaire. We confirmed the physicians' clinical impressions on a subset of subjects (70%) for whom radiographs were available. Radiographs were not required for entry into the study but were used to support clinical impressions of the diagnosis. The physicians' clinical impressions were reported using 5 symptom-based diagnostic categories. Table 1 presents the physicians' clinical impressions and the radiologists' radiographic impressions.

Once referred by the spine center physician, eligible patients were contacted by telephone and screened for enrollment in the clinical trial. Patients were asked whether walking increased their pain or bending backward increased their back symptoms and whether they were currently receiving physical therapy for their low back pain or had an epidural injection in the last 6 months. Patients were not enrolled if they had back surgery in the last year; had back pain secondary to vertebral compression fracture, cancer, or infection; had an epidural steroid injection within the last 6 months (identified via medical record review, physician, and self-report); were currently receiving physical therapy (identified via medical record review, physician report, and patient report on telephone screen); were unable to complete questionnaires because of language or cognitive difficulties (identified by physician report and through telephone screen); had medical problems that limited their function more than their back pain (via physician and self-report); or reported increased LBP, buttock pain, or lower-extremity pain with lumbar flexion (via patient report). We did not invite patients who self-reported increased pain with lumbar flexion, because the intervention for the primary clinical trial involved stationary bicycling.

Those patients who met the eligibility criteria attended the clinic visit. A research assistant discussed the study and answered any questions for the participants. The participants signed an informed consent statement, completed the self-report questionnaires, and provided demographic data. On the same day, one author (MDI), who was unaware of the responses on the self-report forms, performed the PE.

Sixty patients from the RCT and 14 patients from the pilot study satisfied the inclusion criteria. Their median age was 64.2 years ([bar.X]=66.8, SD=12.4, range= 55.7-97.8). Most of the participants were female (68%). Sixty-two subjects (84%) were Caucasian, 8 were African American (11%), 3 were Hispanic (4%), and 1 was Arabic (1%). Fifty-nine percent of the participants had an education exceeding the 12th-grade level. Patient demographic data are presented in Table 2.

Self-report Measures

The primary constructs in the study were patients' symptom severity and functional status. We used a condition-specific measure--the LSS questionnaire (22)--and a generic measure--the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36) questionnaire (23)--to assess patients' self-reported clinical symptom severity and function. The LSS questionnaire was developed specifically for use with patients with degenerative lumbar spinal stenosis. The psychometric properties of the scale were tested on patients aged 50 years and older who had LBP, buttock pain, or lowerextremity pain that was typically exacerbated with lumbar extension and evidence of degenerative lesions of the facets, disks, or ligamentum flavum on imaging studies. (22)

The LSS questionnaire has subscales that represent an individual's back-specific symptom severity and physical function. The symptom severity scale ranges from 1 to 5, and the function scale ranges from 1 to 4, with higher scores reflective of increasing severity or decreasing function. The LSS symptom severity scale contains 7 impairment-based questions concerning common clinical symptoms such as the degree of lower-extremity pain, numbness or tingling in the legs or feet, and lower-extremity weakness. (22) The symptom severity scale uses a Likert classification with 5 categories (1="none," 2="mild," 3="moderate," 4="severe," and 5="very severe") and yields data with established construct validity when compared with the overall Sickness Impact Profile (SIP), including the 3 dimensions of the SIP, and a visual analog scale in patients with degenerative spinal conditions (r =.37, P<.01, confidence interval [CI]=0.24-0.49 and r=.52, P<.01, CI=0.41-0.62). (24) Test-retest reliability of data obtained for individual test items, assessed using Pearson correlations, ranged from .60 to .91. (22) The physical function scale has 6 questions that address activities such as walking distance, ability to walk for pleasure, and ability to walk to and from the bedroom and bathroom, and it uses a Likert classification with 4 categories (1="yes, comfortably"; 2="yes, but sometimes with pain"; 3="yes, but always with pain"; and 4="no"). Walking distance was described as ability to walk over 3.2 km (2 mi), over 2 blocks but less than 3.2 km, over 15.24 m (50 ft) but less than 2 blocks, and less than 15.24 m. In a previous study, (22) the internal consistency and the test-retest reliability of the function scale after 14 days were (alpha=.82 and Pearson r=.71-.94, respectively). The developers of the scale established construct validity through comparison with the global and physical function dimension of the SIP (r=.43, P=<.01, CI=0.38-0.59 and r=.49, P<.01, CI=0.38-0.59). (24) The score for each subscale is calculated as an unweighted mean of the items within each subscale. (22)

The SF-36 is a generic measure constructed to survey health status in clinical practice and research. (23,25,26) We used the bodily pain and physical function subscales. The SF-36 raw questionnaire data are scaled from 0 to 100, with lower scores reflective of worse bodily pain and decreased function. The 2-item pain scale is a well-validated tool used to demonstrate an individual's global perception of pain. (23) The physical function scale is a 10-item scale that addresses general activities of daily living. The SF-36 physical function and pain subscales have been used in patients with LBP. (23,25,26)

These self-report measures have proved useful in determining outcomes and are sensitive to clinical change. (22-29) To determine clinically meaningful change in symptoms and function following an intervention, the developers of the LSS questionnaire used a cutpoint of 2.5 on the LSS satisfaction scale (2="somewhat satisfied," 3="somewhat dissatisfied"). (22,24) We used the same rationale to differentiate subjects believed to have clinically meaningful symptom severity and decreased function. A symptom severity score of 2.5 or greater (out of a 5-point scale) or between "none or mild" and "moderate or severe" was the cutpoint used to describe individuals felt to have clinically meaningful symptom severity. Clinically meaningful decreased physical function was operationally defined as 2.0 or greater (out of a 4-point scale).

Physical Examination

One physical therapist (MDI), who was unaware of the self-reported questionnaire data and physician diagnosis, performed the PE. The examiner is an experienced clinician and researcher with 16 years of experience. She used lumbar symptom provocation tests, manual muscle testing (MMT), and sensory testing. All symptom provocation tests were categorized as positive or negative. Specifically, active extension in standing with a 30-second hold, a test used to reproduce back symptoms (30); active lumbar flexion and lateral flexion in standing, the quadrant test (ie, extension, side bending left, and rotation left) (31); the straight-leg-raising test (31); and the Lasegue test, (31) a neural tension test, were completed. A test was considered positive if the patient reported pain or signs of neural compression such as tingling, numbness, or weakness. Pain with active extension in standing was further described by ranking the response using the following scale: 0="no back pain or pain extending no farther than the knee," 1="pain extending beyond the knee." The test descriptions and interpretations of positive results are shown in Table 3.

Because reproducibility is vital for clinical utility, it is important to note that measurements of pain with trunk motion (ie, lumbar flexion, extension, lateral flexion) and pain with straight leg raising (the straight-leg-raising test) have satisfactory reliability between raters (kappa=.40-.80). (32) Strender et al, (33) in a study using 2 physical therapist raters who had worked together for "many years," also concluded that measurements of pain with trunk motion (kappa=.51-.76) and pain with straight leg raising (kappa=.83) have acceptable reliability; agreement between raters for the quadrant test was 94%.

Response to vibration, pinprick, reflexes, and MMT (ie, iliopsoas, quadriceps femoris, hamstring, peroneal, extensor hallucis longus, gastrocnemius-soleus, and tibialis anterior muscles) was assessed to illustrate impairments suggestive of neural compression. A 128-Hz tuning fork was used to determine participants' vibration sense at the medial malleolus (L4), the lateral malleolus (L5), and the lateral head of the fifth metatarsal (S1). Vibration was categorized as normal, diminished, or absent. Participants' response to pinprick was determined as normal, diminished, or absent at the medial malleolus (L4), the lateral malleolus (SI), and the web space between the first and second rays (L5). (34) Patellar tendon and Achilles tendon reflexes were rated as normal, hypertonic, or hypotonic. Manual muscle testing (break test) was performed as outlined by Hislop and Montgomery. (35) Manual muscle test scores for the lower extremity were dichotomized into grades of less than or equal to 4 versus a grade of 5. Individuals with lower-extremity weakness in this study had a MMT grade of less than or equal to 4. The reliability of MMT scores is improved when clearer delineations are used. The agreement (kappa) between 2 orthopedic surgeons in 50 patients with LBP was .65 to 1.00 for MMT, .23 to .39 for reflexes, and .68 for numbness. (32)

We included other PE measures to rule out or determine the possible contribution of adjacent joints. The hip screens included the following: the amount of hip medial (internal) rotation by visual estimate (<10[degrees] or -10[degrees]) and notation of pain with hip medial rotation, (36) the Faber test (31) with notation of whether pain was reproduced in the hip or sacroiliac region, and tenderness to palpation at the greater trochanter. (31) Report of pain with hip medial rotation and decreased hip medial range of motion is associated with osteoarthritis of the hip. (36) The Faber test places stress on both the hip and sacroiliac joints, and thus a report of pain in either area suggests hip or sacroiliac pathology, respectively. (31) Tenderness to palpation in the area of the greater trochanter is suggestive of trochanteric bursitis. (31,37) Although the reliability and validity of data for the Faber test (38) and trochanteric tenderness can be questioned, we included these tests for their potential discriminative value and for their potential relationship to the reference standards.

Individuals who experienced discomfort with lumbar flexion during the PE were not excluded from the study. An individual who reported pain with lumbar flexion also must have reported pain with the quadrant test or pain with sustained lumbar extension. Although individuals with degenerative low back conditions are generally more extension sensitive, lumbar flexion also may reproduce pain, and we feel the relationship is important to investigate.

Other Measures

Several questionnaires were included to account for potential confounding. The Cumulative Illness Rating Scale (CIRS), (39) which is a chart-based measure, and the Charlson comorbidity index, (40) which is a self-report to measure medical comorbidities, were used to measure medical comorbidity. The CIRS rates the degree of pathology and impairments in each of 12 major organ groups, as well as psychiatric behavioral categories, (39) and is considered a valid measure of health status. (41) The primary investigator (MDI) collected these data on subjects from the medical record. A standardized comorbidity value was calculated by taking individual scores on the respective comorbidity scales, subtracting these values from the mean, and dividing this value by the standard deviation of the scores. A musculoskeletal checklist was completed. Patients were asked to indicate whether they had pain in the lower extremity or difficulty walking due to conditions of the lower extremity. The Mental Health Index (MHI-5) subscale of the SF-36 was used to quantify mental health status as CLBP has a psychological dimension. The MHI-5 has been reported to yield reliable and valid data. (23) Patients were allowed to continue their prescribed medications. We collected data on medication used (pain medications, sedatives, muscle relaxants, anticonvulsants, or antidepressants) because these drugs may alter the perception of pain or symptoms during the PE and may affect responses on self-report measures of symptoms and function. Medication usage was abstracted retrospectively from patients' medical records using a standardized form.

Data Analysis

Analyses were performed with the SAS statistical package. *, (42) We used a 3-step approach in the analyses. In step 1, descriptive statistics were used to illustrate the sample's characteristics. In step 2, our goal was to determine the relationship of exploratory variables. We used t tests and Spearman correlation tests, as appropriate, to determine relationships between the PE measures and the condition-specific and generic scales. (43,44) The critical alpha level was .005 to adjust for multiple testing. (43,44)

In step 3, stepwise and hierarchical multivariate linear regression models were used to determine the unique contribution of each explanatory factor to the variance of low back symptom severity and function as measured by the condition-specific and generic self-reported outcome measures. The stepwise modeling approach was confirmed with forward- and backward-selection procedures, with a P value of .10 necessary for entry or removal from the model. Only PE variables that we hypothesized would affect self-report of symptoms and function were used in the modeling (these included quadrant test, pain with lateral flexion, pain with extension, muscle weakness, decreased sensation, trochanteric tenderness, Faber test for hip pain, and pain with lumbar flexion). The PE items were entered into the model as a group. Then only significant PE variables were retained as demographic features such as age, sex, mental health (ie, SF-36 MHI-5), medication use, level of education (ie, high-school or less versus advanced degree as education is a proxy for access to medical care), and comorbidity, defined as the standardized comorbidity value, were considered in the model and entered as a group. (45)

Results

The patients' mean LSS symptom severity and function scores were 2.67 (SD=0.71) and 1.99 (SD=0.61), respectively. The patients' mean SF-36 bodily pain and physical function scores were 43.7 (SD=20.7) and 55.8 (SD=24.6), respectively. These scores indicate that our sample had moderate impairment. Normative data on a comparable sample of 481 adults (mean age=60.4 years) with LBP and a cardiovascular comorbidity were 59.3 and 66.3. (23)

The frequencies of positive findings for the PE measures are displayed in Table 4. The quadrant test was the most common test that reproduced the patients' symptoms. Lower-extremity weakness, abnormal reflexes, and symptoms with sustained lumbar extension followed, respectively, being positive in greater than 61% of the sample.

Power calculations were based on differences of 0.5 point for the symptom severity scale and LSS function scale, indicating a 10% difference. In longitudinal analysis, this value was estimated to be clinically meaningful, whereas in cross-sectional analysis, such as this study, a clinically meaningful value may be as low as 0.3. (22) With alpha set at .005, a stringent level, to adjust for multiple testing in this exploratory analysis and using a mean difference of 0.5 (on the LSS scale) and a standard deviation of 0.6, we had 84% power to detect a difference. Similarly, the clinically relevant change for the SF-36 bodily pain and physical function subscales of 10 points identified by Ware et al (23) was used to detect power for the SF-36 subscales. With the alpha set at .005 and using a mean difference of 10 points and a standard deviation of 19, we had 60% power to detect a 10-point difference between groups.

Table 5 illustrates the mean self-reported symptom severity and function for individuals with and without radiating pain during prolonged lumbar extension (operationally defined as symptoms either above or below the knee). Patients with more distal symptoms had increased LSS symptom severity and function scores than those with localized pain.

Comparison of PE Procedures and the Condition-Specific Questionnaire

We estimated a single PE measure's influence on the self-reported LSS and SF-36 outcome scores using a t test comparing group means (positive and negative PE results) (Tab. 6). Three of the 10 PE measures showed between-group differences for the LSS symptom severity score. Patients who reported symptoms during the quadrant test had higher self-reported LSS symptom severity scores compared with those who did not report symptoms during the quadrant test (t=-3.06, P=.003). This relationship also was observed in those who reported symptoms with lateral flexion (t=-2.96, P=.004) and in those who demonstrated lower-extremity weakness (t=-3.2, P=.002). Examination of the between-group means revealed relative differences in LSS symptom severity scores for these 3 groups of 0.58, 0.47, and 0.53, respectively. The mean LSS symptom severity scores of patients who reported symptoms during the quadrant, test or lateral flexion, and those with lower-extremity weakness were greater than 2.5, and those with negative findings were less than 2.5. Pain with lumbar extension was moderately correlated with the LSS symptom severity score (Spearman r=.31, P=.007).

One PE procedure had a between-group difference for LSS function (Tab. 7). Individuals who had pain with hip medial rotation had decreased LSS functional status compared with those without hip symptoms (t=-2.96, P=-.004). Patients with a positive PE finding, on average, had mean back-specific LSS function scores that exceeded 2.0, whereas the average mean back-specific LSS function scores of patients with negative PE findings were less than 2.0.

Comparison of PE Procedures and the SF-36

One PE procedure was found to have between-group differences for the SF-36 bodily pain subscale and the physical function subscale (Tabs. 8 and 9). Individuals who demonstrated lower-extremity weakness had greater bodily pain compared with those without lower-extremity weakness (t=2.91, P=.005). Patients who reported pain with hip medial rotation had decreased self-reported physical function (t=3.84, P=.0003).

Regression Analysis: Condition-Specific LSS Questionnaire

The quadrant test, trochanteric tenderness, and level of education contributed to the variance of self-reported symptom severity on the LSS questionnaire (Fig. 1). The quadrant test explained the greatest proportion of variance (21%). The [beta] coefficient represents the estimate of the change in the dependent variable (ie, LSS symptom severity scale) given a positive examination finding. The [beta] coefficients for the quadrant test, trochanteric tenderness, and level of education were 0.54 (P=.0009), 0.68 (P=.002), and -0.61 (P=.005), respectively. Thus, individuals who had a positive quadrant test are expected to score a half of a point higher on the LSS symptom severity scale. The change score of -0.61, on the contrary, represents the positive influence of having an education beyond the high school level. Trochanteric tenderness and level of education combined to explain 28% of the variance in the LSS function score, with [beta] coefficients of 0.60 (P=.007) and -0.57 (P=.004), respectively (Fig. 2). The presence of trochanteric tenderness was associated with decreased function, whereas an education beyond the 12th grade was associated with improved function.

Regression Analysis: Generic Health SF-36 Questionnaire No PE measure contributed to the variance of bodily pain as measured by the SF-36 (Fig. 3). Education beyond the high school level was associated with a decreased SF-36 bodily pain score (P=.001), and being female was associated with an increased bodily pain score (P=.005). Pain with extension, trochanteric tenderness, and level of education combined to explain 29% of the variance in the SF-36 physical function score (Fig. 4). Trochanteric tenderness was associated with a large decrease in the physical function score ([beta]=-19.6, P=.01). Education beyond the high school level improved the physical function score ([beta]=18.5, P=.01), and pain reported with extension decreased the physical function score (P=.004).

Discussion

We used an observational design to investigate the relationship between commonly used PE procedures and self-reported symptom severity and function in individuals with degenerative lumbar conditions. The findings illustrate the complex relationship of spinal PE procedures and self-reported condition severity and function. Previous work aimed at clarifying the usefulness of PE procedures in circumstances other than diagnosis has been unenlightening. McGregor et al (14) studied the impact of self-reported and clinical factors on flexion and extension range of motion and ascent and descent speed and concluded that their data were varied and difficult to interpret. Michel et al (13) found that rotation, fingertip-to-floor distance, and lateral flexion range of motion correlated best with self-reported severity in people with nonspecific LBP. We found that the quadrant test distinguished between subjects with clinically important back-specific LSS symptom severity. Pain with lateral flexion and extension, lower-extremity weakness, and trochanteric tenderness were associated with worse LSS symptom severity scores. However, none of the PE measures used in this study were discriminative of clinically important LSS physical function. Pain with lumbar extension was associated with decreased SF-36 physical function scores in regression analyses. Our results suggest that impairment-based symptom provocation procedures are more reflective of the condition severity domain than the functional domain.

Considering variables other than back-specific provocation tests proved important in this study. Hip impairment was present in [is less than or equal to] 22% of the sample. Yet 2 hip PE measures contributed to the variance in function and symptom severity. Pain with hip medial rotation was associated with worse LSS and SF-36 function scores, and pain with trochanteric tenderness was predictive of worse LSS severity scores, worse LSS function scores, and worse SF-36 physical function scores. These findings may be explained by the LSS questionnaire's focus on tasks that require ambulation. In addition, individuals with education beyond high school reported less LSS symptom severity and greater function on the LSS questionnaire, suggesting a potential socioeconomic or psychosocial influence. These findings illustrate the importance of examining a patient globally; an individual's symptom severity and functional status may be influenced equally by comorbid musculoskeletal conditions and psychosocial factors.

The frequency of positive PE findings in this cohort was expected and supports the notion that tests of facet and foraminal compression are more symptom provoking than tests of neural tension and hip or sacroiliac pathology in patients with degenerative LBP. The quadrant test, active extension, and lateral flexion all compromise the foraminal opening and central canal. (31,46,47) In contrast, lumbar flexion enlarges the foraminal space and central canal. (46,47) Because the pathoanatomy of degenerative spinal conditions typically narrows the foramen either centrally or laterally, (48) PE measures that further lessen the foramen (ie, quadrant test) are expected to be more provocative than tests that enlarge the foraminal space (ie, lumbar flexion). In addition to lumbar flexion widening the foraminal space, decreased load is borne by the posterior elements, namely the facet joints, which are highly innervated and are a potential source of symptoms. Symptoms reported with movement also could represent a diskogenic etiology (49-51) and may explain why some individuals reported discomfort with active lumbar flexion. The straight-leg-raising test and the Lasegue test place tension on the sciatic nerve (31) and are not typically positive in degenerative spinal conditions. (34,52) Impairments suggestive of neural compression (ie, abnormal reflexes, pinprick, vibration, and weakness) have been reported in this population, (30,34,48,52) and our data revealed similar findings. Signs of neural compression cannot be attributed solely to pathology of the lumbar spine, however, because sensory and force deficits are generally more prevalent in elderly people. (53)

Self-report measures such as the LSS and SF-36 questionnaires are useful adjuncts to clinical practice and may be more responsive than commonly used physical examination procedures to detect clinical improvement in people with LBP. (54) Because of the currently limited use of self-report measures in clinical practice (secondary to time, coding, and so on), (11) the use of PE procedures to reflect symptom severity and function might be more easily incorporated into daily practice. Physical examination procedures are widely used to discriminate between diagnoses.

Understanding the relationship of PE procedures and self-reported severity and function has the potential to improve clinical decision making. The results of our study illustrate that PE procedures may be useful to discriminate among different levels of severity within a group of people over 55 years of age with CLBP due to degenerative changes. For example, a positive quadrant test was associated with clinically meaningful increased symptom severity. In addition, individuals with more distal symptoms had greater average LSS and SF-36 function scores (Tab. 5). These results are consistent with those of Loisel et al, (55) who found that people with distal radiating pain were more likely to have decreased functional status and increased pain level and were less likely to return to work at 1-year follow-up. We believe that identifying people with clinically meaningful symptom severity and function may indicate a worse prognosis.

Limitations

The study has several limitations. We attempted to assemble a homogeneous cohort of subjects based on physicians' clinical impressions of disease status and symptoms. We used radiographs on a subset of subjects (70%), when available, to confirm the physicians' clinical impressions. However, radiographs were not available on all subjects, so the potential exists for misclassification. Our sample was recruited from a large tertiary care institution and, therefore, may have more severe pain impairments compared with patients from other facilities, potentially limiting the generalizability of the study results. Nonetheless, the SF-36 scores in our sample do not exceed normative data on a comparable sample with similar comorbidities and age. (23) The entry criteria required that patients self-report pain with lumbar extension. Therefore, the prevalence of pain with extension in this cohort is high by definition. The degree to which the sample may be generalized is not clear, because the prevalence of pain with extension in a nonselected cohort of older patients with back pain is unknown. The LSS questionnaire contains general physical function items that attribute functional performance (eg, walking) and symptoms (eg, tingling in legs) to back pain. Given that patients were selected based on report of pain with walking or extension and the disease-specific questionnaire measures symptoms with functional activities performed in standing, it is likely that we may have found stronger relationships between certain PE tests and self-reported back function and symptoms. A fair proportion of patients who reported back pain with walking, however, did not have positive provocation tests. Finally, to be conservative, we used a Bonferroni correction to adjust for multiple testing.

Using these stringent criteria, we had 84% power to detect clinically meaningful differences in LSS-specific scales and 60% power to detect clinically meaningful differences in generic function (SF-36). Because a clinically meaningful difference for the LSS symptom severity scale and LSS function scale is 0.5 point (22) and a clinically meaningful change for the SF-36 bodily pain and physical function subscales is 10 points, (23) some differences in self-reported outcomes exceeded our criteria for clinically meaningful change but did not meet the criteria we established for statistical significance in the study (P=.005), indicating low power or that our alpha level was overly conservative. We suggest interpreting these items with caution. We provide the reader with effect sizes and P values so that the reader can independently interpret the weight of the evidence.

The study is strengthened by the use of well-validated and reliable generic and condition-specific self-report measures (the LSS and SF-36 questionnaires, respectively) and the use of strict inclusion criteria to gather a relatively homogeneous sample with respect to symptom presentation. The examiner performing the PE measures was unaware of the self-report data and diagnosis. Although we do not report on the reliability of the tester, an effort was made to thoroughly define the PE procedures and interpretation of each finding. We feel this is very important and encourage clinicians to standardize their procedures and interpretation of positive findings to facilitate communication and clarity.

Conclusion

To our knowledge, this is the first study comparing condition-specific and generic measures of condition severity and function with the PE findings in a sample of older adults with degenerative spinal conditions. In this study, the associations between the impairment-based movement patterns and self-reported symptom severity are consistent with the pathoanatomy of degenerative spinal conditions. The quadrant test distinguished those patients with clinically meaningful symptom severity, as did lateral flexion and lower-extremity weakness. Pain with hip medial rotation, trochanteric tenderness, and level of education were additional variables that influenced symptom severity and function. We feel that PE procedures may be clinically useful in people with chronic conditions when used in models that are predictive of symptom severity and function. The PE measures used in this study were not discriminative of function, and future studies may benefit from including a clinical measure that may better reflect the function domain. (56-61) Further investigation in this area is warranted to determine if these findings can be applied to guide intervention and prognosis.

References

(1) Lavsky-Shulan M, Wallace RB, Kohout FJ, et al. Prevalence and functional correlates of low back pain in the elderly: the Iowa 65+ Rural Health Study. J Am Geriatr Soc. 1985;33:23-28.

(2) Fujiwara A, Tamai K, Yamato M, et al. The relationship between facet joint osteoarthritis and disc degeneration of the lumbar spine: an MRI study. Eur Spine J. 1999;8:396-401.

(3) Beattie PF, Meyers SP. Magnetic resonance imaging in low back pain: general principles and clinical issues. Phys Ther. 1998;78:738-754.

(4) Buckwalter JA. Aging and degeneration of the human intervertebral disc. Spine. 1995;20:1307-1314.

(5) Magora A, Schwarz A. Relation between the low back pain syndrome and x-ray findings, I: degenerative osteoarthritis. Scand J Rehabil Med. 1976;8:115-125.

(6) Healy JF, Healy BB, Wong WHM, Olson EM. Cervical and lumbar MRI in asymptomatic older male lifelong athletes: frequency of degenerative findings. J Comput Assist Tomogr. 1996;20:107-112.

(7) Delitto A, Snyder-Mackler L. The diagnostic process: examples in orthopedic physical therapy. Phys Ther. 1995;75:203-211.

(8) Guide to Physical Therapist Practice. 2nd ed. Phys Ther. 2001;81:43.

(9) Riddle DL. Classification and low back pain: a review of the literature and critical analysis of selected systems. Phys Ther. 1998;78:708-737.

(10) Flynn T, Fritz J, Whitman J, et al. A clinical prediction rule for classifying patients with low back pain who demonstrate short-term improvement with spinal manipulation. Spine. 2002;27:2835-2843.

(11) Resnik L, Dobrzykowski E. Guide to outcome measurement for patients with low back pain syndromes. J Orthop Sports Phys Ther. 2003;33:307-318.

(12) Waddell G, Main CJ. Assessment of severity in low back disorders. Spine. 1984;9:204-208.

(13) Michel A, Kohlmann T, Raspe H. The association between clinical findings on physical examination and self-reported severity in back pain: results of a population-based study. Spine. 1997;22:296-304.

(14) McGregor AH, Dore CJ, McCarthy ID, Hughes SP. Are subjective clinical findings and objective clinical tests related to the motion characteristics of low back pain subjects? J Orthop Sports Phys Ther. 1998;28:370-377.

(15) Waddell G, Somerville D, Henderson I, Newton M. Objective clinical evaluation of physical impairment in chronic low back pain. Spine. 1992;17:617-628.

(16) Gronblad M, Hurri H, Kouri JP. Relationships between spinal mobility, physical performance tests, pain intensity and disability assessments in chronic low back pain patients. Scand J Rehabil Med. 1997;29:17-24.

(17) Hazard RG, Haugh LD, Green PA, Jones PL. Chronic low back pain: the relationship between patient satisfaction and pain, impairment, and disability outcomes. Spine. 1994;19:881-887.

(18) Cox ME, Asselin S, Gracovetscky SA, et al. Relationship between functional evaluation measures and self-assessment in nonacute low back pain. Spine. 2000;25:1817-1826.

(19) Nattrass CL, Nitschke JE, Disler PB, et al. Lumbar spine range of motion as a measure of physical and functional impairment: an investigation of validity. Clin Rehabil. 1999;13:211-218.

(20) Feuerstein M, Beattie P. Biobehavioral factors affecting pain and disability in low back pain: mechanisms and assessment. Phys Ther. 1995;75:267-281.

(21) Iversen MD, Fossel AH, Katz JN. Enhancing function in older adults with chronic low back pain: a pilot study of an endurance training program. Arch Phys Med Rehabil. 2003;84:1324-1331.

(22) Stucki G, Daltroy L, Liang MH, et al. Measurement properties of a serf-administered outcome measure in lumbar spinal stenosis. Spine. 1996;21:796-803.

(23) Ware JE, Snow KK, Kosinski M, Gandek B. SF-36 Health Survey Manual and Interpretation Guide. Boston, Mass: The Health Institute; 1993.

(24) Stucki G, Liang MH, Fossel AH, Katz JN. Relative responsiveness of condition-specific and generic health status measures in degenerative lumbar spinal stenosis. J Clin Epidemiol. 1995;48:1369-1378.

(25) Patrick DL, Deyo RA. Generic and disease-specific measures in assessing health status and quality of life. Med Care. 1989;27:S217-S232.

(26) Lurie J. A review of generic health status measures in patients with low back pain. Spine. 2000;25:3125-3129.

(27) Taylor SJ, Taylor AE, Foy MA, Fogg AJB. Responsiveness of common outcome measures for patients with low back pain. Spine. 1999;24:1805.

(28) Walsh TL, Hanscom B, Lurie JD, Weinstein JN. Is a condition-specific instrument for patients with low back pain/leg symptoms really necessary? Spine. 2003;28:607-615.

(29) Bombardier C. Outcome assessments in the evaluation of treatment of spinal disorders: summary and general recommendations. Spine. 2000;25:3100-3103.

(30) Katz JN, Dalgas M, Stucki G, et al. Degenerative lumbar spinal stenosis: diagnostic value of the history and physical examination. Arthritis Rheum. 1995;38:1236-1241.

(31) Magee D. Orthopedic Physical Assessment. 3rd ed. Philadelphia, Pa: WB Saunders Co; 1997.

(32) McCombe PF, Fairbank JCT, Cockersole BC, Pynsent PB. Reproducibility of physical signs in low-back pain. Spine. 1989;14:908-917.

(33) Strender L, Sjoblom A, Sundell K, et al. Interexaminer reliability in physical examination of patients with low back pain. Spine. 1997;22: 814-820.

(34) Deyo RA, Rainville F, Kent DL. What can the history and physical examination tell us about low back pain? JAMA. 1992;256:760-765.

(35) Hislop HJ, Montgomery J. Daniel's and Worthingham's Muscle Testing: Techniques of Manual Examination. 6th ed. Philadelphia, Pa: WB Saunders Co; 1995.

(36) Altman R, Alarcon G, Appelrouth D, et al. The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hip. Arthritis Rheum. 1991;34:505-515.

(37) Shbeeb M, Matteson EL. Trochanteric bursitis (greater trochanter pain syndrome). Mayo Clinic Proc. 1996;71:565-569.

(38) Dreyfuss P, Michaelson M, Pauza K, et al. The value of medical history and physical examination in diagnosing sacroiliac joint pain. Spine. 1996;21:2594-2602.

(39) Linn BS, Linn MW, Gurel L. Cumulative illness rating scale. J Am Geriatr Soc. 1968;16:622-626.

(40) Katz JN, Chang LC, Sangha O, et al. Can cormorbidity be measured by questionnaire rather than medical record review? Med Care. 1996; 34:73-84.

(41) Parmelee PA, Thuras PD, Katz IR, Lawton MP. Validation of the cumulative illness rating scale in a geriatric population. J Am Geriatr Soc. 1995;43:130-137.

(42) SAS/STAT User's Guide, Release 6.03 Edition. Cary, NC: SAS Institute Inc; 1988.

(43) Portney LG, Watkins MP. Foundations of Clinical Research: Applications to Practice. 2nd ed. Upper Saddle River, NJ: Prentice-Hall Inc; 2000.

(44) Rosner B. Fundamentals of Biostatistics. 5th ed. Pacific Grove, Calif: Duxbury Thompson; 2000.

(45) Clauw DJ, Williams D, Lauerman W, et al. Pain sensitivity as a correlate of clinical status in individuals with chronic low back pain. Spine. 1999;24:2035-2041.

(46) Dai LY, Yu YK, Zhang WM, Zhou ZH. The effect of flexion-extension motion of the lumbar spine on the capacity of the spinal canal: an experimental study. Spine. 1989;14:523-525.

(47) Fredrickson M, Lee SU, Welsh J, et al. Changes in posterior disc bulging and intervertebral foraminal size associated with flexion-extension movement: a comparison between L4-5 and L5-S1 levels. Spine. 2001;1:10-17.

(48) Spivak JM. Degenerative lumbar spinal stenosis. J Bone Joint Surg Am. 1998;80:1053-1065.

(49) Coppes MH, Marani E, Thomeer RTWM, Groen GJ. Innervation of "painful" lumbar discs. Spine. 1997;22:2342-2349.

(50) Freemont AJ, Peacock TE, Goupille P, et al. Nerve ingrowth into diseased intervertebral disc in chronic back pain. Lancet. 1997; 350(9072):178-181.

(51) Kauppila LI. Ingrowth of blood vessels in disc degeneration: angiographic and histological studies of cadaveric spines. J Bone Joint Surg Am. 1995;77:26-31.

(52) Iversen MD, Katz JN. Examination findings and self-reported walking capacity in patients with lumbar spinal stenosis. Phys Ther. 2001;81: 1296-1306.

(53) Lewis CB. Aging: The Health-Care Challenge. 4th ed. Philadelphia, Pa: FA Davis Co; 2002.

(54) Fritz JM, Piva SR. Physical impairment index: reliability, validity, and responsiveness in patients with acute low back pain. Spine. 2003;28: 1189-1194.

(55) Loisel P, Vachon B, Lemaire J, et al. Discriminative and predictive validity assessment of the Quebec Task Force Classification. Spine. 2002;27:851-857.

(56) 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;82:227-231.

(57) Simmonds MJ, Claveau Y. Measures of pain and physical function in patients with low back pain. Physiotherapy Theory and Practice. 1997;13: 53-65.

(58) Deen HG, Zimmerman RS, Lyonns MK, et al. Measurement of exercise tolerance on the treadmill in patients with symptomatic lumbar spinal stenosis: a useful indicator of functional status and surgical outcome. J Neurosurg. 1995;83:27-30.

(59) Fritz JM, Erhard RE, Vignovic M. A nonsurgical treatment approach for patients with lumbar spinal stenosis. Phys Ther. 1997;77:962-974.

(60) Dong GX, Porter RW. Walking and cycling test in neurogenic and intermittent claudication. Spine. 1989;14:965-969.

(61) Podsiadlo D, Richardson S. The timed "up and go": a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991;39:142-148.

* SAS Institute Inc, PO Box 8000, Cary, NC 27511. MA Lyle, PT, MSPT, S Manes, PT, DPT, M McGuinness, PT, DPT, and S Ziaei, PT, DPT, were students in the Department of Physical Therapy, Simmons College, Boston, Mass, at the time of this study.

MD Iversen, PT, SD, MPH, is Professor, Graduate Programs in Physical Therapy, MGH Institute of Health Professions, and Instructor in Medicine, RBB Arthritis Research Center, Department of Medicine, Division of Rheumatology, Immunology & Allergy, Section of Clinical Sciences, Brigham and Women's Hospital, Harvard Medical School, Boston, Mass. Address all correspondence to Dr Iversen at Graduate Programs in Physical Therapy, MGH Institute of Health Professions, 36 First Ave, Boston, MA 02129 (USA) ([e-mail]).

All authors provided concept/research design and data analysis. Mr Lyle and Dr Iversen provided writing and data analysis. Dr Iversen provided project management, fund procurement, subjects, facilities/equipment, and institutional liaisons.

This study was approved by the institutional review boards of Brigham and Women's Hospital and Simmons College.

Partial funding for this study was provided by a Farnsworth Postdoctoral Fellowship Grant from the Medical Foundation to Dr Iversen, a New Investigator Award from the National Arthritis Foundation to Dr Iversen, and a Research Award from the Foundation for Physical Therapy to Dr Iversen.

The results of this study were presented at the Annual Scientific Meeting of the American College of Rheumatology; October 26, 2002; New Orleans, La.

This article was received August 14, 2003, and was accepted August 8, 2004.

Table 1.

Spine Physicians' Clinical Impressions and Radiologists' Radiographic
Impressions of Patient Condition

                                                     N      %

Spine physicians' clinical impressions (n=74) (a)
  Central low back pain                              40    54
  Central low back pain with radiation               17    23
  Central low back pain with neural deficits          4     5
  Neurogenic claudication due to lumbar spinal       16    22
    stenosis
  Sciatica due to herniated nucleus pulposus          2     2.7
Radiologists' impressions (n=52 radiographs) (a)
  Normal                                              1     2
  Degenerative disk disease                          30    58
  Lumbar spinal stenosis                             26    50
  Herniated nucleus pulposus                         15    29
  Scoliosis                                           9    17
  Facet hyperostosis/osteoarthritis                  14    27

(a) More than one category could be selected.

Table 2.

Demographic Characteristics of Patients With Chronic Low Back Pain
(n=74)

Variable                N (%)     [bar.X]   SD     Median   Range

Age (y)                            66.8     12.4    64.2    55.7-97.8
No. of medications                  1.4      1.2     1      1-5
Sex
  Female                50 (68)
Race
  Caucasian             62 (84)
  African-American       8 (11)
  Hispanic               3 (4)
  Arabic                 1 (1)
Education ([less than
  or equal to] 12 y)    30 (41)

Table 3.

Physical Examination of Measures and Criteria for a Positive Result (a)

Test/Measure           Method                   Positive Result

Quadrant test (left/   Patient actively moves   Patient reports pain
  right)                 into SBL, Ext, and       or numbness or
                         RL and into SBR,         tingling in the area
                         Ext, and RR as far       of the back or lower
                         as tolerated. The        extremities.
                         examiner guides the
                         patient in this
                         motion and provides
                         overpressure if no
                         pain is reported
                         with their active
                         movement. The
                         position is held for
                         3 s.
Lumbar extension       Patient actively         Patient reports pain
                         extends as far as        or numbness or
                         tolerated with the       tingling: location
                         examiner providing       of symptoms were
                         support for balance.     documented as back,
                         The position is held     buttocks, thigh,
                         for 30 s.                calf, and foot and
                                                  ankle.
Lateral flexion        Patient actively side    Patient reports pain
  (right and left)       bends as far as          in the area of the
                         tolerated.               back or lower
                                                  extremities.
Lumbar flexion         Patient actively         Patient reports pain
                         forward bends as far     in the area of the
                         as tolerated.            back or lower
                                                  extremities.
Straight-leg-raising   Patient is positioned    Patient reports pain
  test                   supine. The examiner     distal to the knee.
                         passively flexes the     Pain must be
                         extended lower           reported with [less
                         extremity of the         than or equal to]
                         patient as much as       70[degrees] of hip
                         tolerated.               flexion.
Lasegue test           Patient is positioned    Patient reports pain
                         supine. The examiner     below the knee,
                         passively flexes the     which increases with
                         extended lower           neck flexion and
                         extremity of the         decreases as head
                         patient as               returns to neutral.
                         tolerated. The
                         patient is
                         instructed to flex
                         the neck as
                         tolerated.

(a) SBR=side bending right, SBL=side bending left, Ext=lumbar
extension, RR=rotation right, RL=rotation left.

Table 4.

Frequency of Positive Physical Examination Findings (n=74)

Variable                                  N     %

Quadrant test (a)                         42    70
Lower-extremity weakness (b)              44    64
Abnormal reflexes                         46    62
Pain with lumbar extension                45    61
Abnormal vibration sense                  43    58
Pain with lateral flexion                 40    54
Abnormal pinprick                         37    50
Faber test (positive) for hip pain (c)    21    30
Pain with lumbar flexion                  18    24
Hip pain with medial rotation (d)         15    22
Trochanteric tenderness (e)               13    18
Straight-leg-raising test                 12    16
Lasegue test                              11    15

(a) n=60, quadrant test was performed in randomized controlled trial
only.

(b) n=69.

(c) n=71.

(d) n=68 (not completed on patients with total hip replacement).

(e) n=73.

Table 5.

Mean Self-Reported Lumbar Spinal Stenosis Symptom Severity and Function
Scores in Patients Who Reported Symptoms Proximal or Distal to the Knee
With Sustained Extension (n=74)

                          Mean Score After 30 s of Sustained
                          Extension

                          No Pain Reported     Pain Reported
Self-Report               Distal to the Knee   Distal to the Knee
Measure                   (n=52)               (n=22)               P

LSS symptom severity       2.55                 2.96                .04
LSS physical function      1.91                 2.18                .08
SF-36 (a) bodily pain     44.5                 41.7                 .60
SF-36 physical function   60.0                 46.0                 .02

(a) SF-36=Medical Outcomes Study 36-Item Short-Form Health Survey.

Table 6.

Lumbar Spinal Stenosis Symptom Severity Score Group Means for
Positive and Negative Physical Examination (PE) Measures

                      LSS Symptom Severity Score

                      Positive PE Measure       Negative PE Measure

Variable              [bar.X]            SD     [bar.X]            SD

Quadrant test         2.84               0.67   2.26               0.67
                                (n=42)                    (n=18)
Pain with lateral     2.89               0.73   2.42               0.60
  flexion                       (n=34)                    (n=40)
Pain with lumbar      2.80               0.68   2.63               0.72
  flexion                       (n=18)                    (n=56)
Straight-leg-         3.04               0.66   2.60               0.70
  raising test                  (n=12)                    (n=62)
Lasegue test          2.94               0.72   2.63               0.70
                                (n=11)                    (n=63)
Hip medial rotation   3.06               0.68   2.63               0.68
                                (n=15)                    (n=53)
Faber test for hip    2.94               0.81   2.58               0.64
  pain                          (n=21)                    (n=50)
Trochanteric          3.09               0.62   2.58               0.70
  tenderness                    (n=13)                    (n=60)
Lower-extremity       2.89               0.73   2.36               0.51
  weakness                      (n=44)                    (n=25)

Variable              t       p

Quadrant test         -3.06   .003 (a)

Pain with lateral     -2.96   .004 (a)
  flexion
Pain with lumbar      -0.87   .39
  flexion
Straight-leg-         -2.03   .046
  raising test
Lasegue test          -1.33   .188

Hip medial rotation   -2.16   .03

Faber test for hip    -1.99   .05
  pain
Trochanteric          -2.41   .019
  tenderness
Lower-extremity       -3.20   .001 (a)
  weakness

(a) Significant at P=.001.

Table 7.

Lumbar Spinal Stenosis Physical Function Score Group Means for
Positive and Negative Physical Examination (PE) Measures

                      LSS Physical Function Score

                      Positive PE Measure       Negative PE Measure

Variable              [bar.X]            SD     [bar.X]            SD

Quadrant test         2.07               0.67   1.81               0.41
                                (n=42)                    (n=18)
Pain with lateral     2.11               0.67   1.85               0.50
  flexion                       (n=40)                    (n=34)
Pain with lumbar      2.06               0.59   1.97               0.61
  flexion                       (n=18)                    (n=56)
Straight-leg-         2.13               0.52   1.96               0.62
  raising test                  (n=12)                    (n=62)
Lasegue test          2.11               0.44   1.97               0.44
                                (n=11)                    (n=63)
Hip medial rotation   2.41               0.61   1.91               0.58
                                (n=15)                    (n=52)
Faber test for hip    2.04               0.59   1.99               0.62
  pain                          (n=21)                    (n=50)
Trochanteric          2.25               0.59   1.92               0.59
  tenderness                    (n=13)                    (n=60)
Lower-extremity       2.08               0.65   1.82               0.54
  weakness                      (n=44)                    (n=25)

Variable              t       p

Quadrant test         -1.53   .13

Pain with lateral     -1.89   .06
  flexion
Pain with lumbar      -0.53   .59
  flexion
Straight-leg-         -0.90   .37
  raising test
Lasegue test          -0.71   .48

Hip medial rotation   -2.96   .004 (a)

Faber test for hip    -0.37   .71
  pain
Trochanteric          -1.82   .07
  tenderness
Lower-extremity       -1.76   .08
  weakness

(a) Significant at P.001.

Table 8.

Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36) Bodily
Pain Subscale Score Group Means for Positive and Negative Physical
Examination (PE) Measures

                      SF-36 Bodily Pain Subscale Score

                      Positive PE Measure       Negative PE Measure

Variable              [bar.X]            SD     [bar.X]            SD

Quadrant test         40.4               21     51.6               20
                                (n=42)                    (n=18)
Pain with lateral     39.8               20     48.2               21
  flexion                       (n=40)                    (n=34)
Pain with lumbar      38.1               13     45.5               22
  flexion                       (n=18)                    (n=56)
Straight-leg-         35.5               16.2   45.2               21
  raising test                  (n=12)                    (n=62)
Lasegue test          37.9               18.5   44.7               21
                                (n=11)                    (n=63)
Hip medial rotation   32.1               12.2   45.0               21.5
                                (n=15)                    (n=53)
Faber test for hip    34.2               19.7   46.5               19.5
  pain                          (n=21)                    (n=50)
Trochanteric          37.7               15     45.2               21
  tenderness                    (n=13)                    (n=60)
Lower-extremity       38.9               20.1   53.6               20.5
  weakness                      (n=44)                    (n=25)

Variable              t      p

Quadrant test         1.87   .067

Pain with lateral     1.77   .08
  flexion
Pain with lumbar      1.33   .08
  flexion
Straight-leg-         1.50   .14
  raising test
Lasegue test          1.00   .32

Hip medial rotation   2.99   .004 (a)

Faber test for hip    2.41   .02
  pain
Trochanteric          1.17   .24
  tenderness
Lower-extremity       2.91   .005 (a)
  weakness

(a) Significant at P=.001.

Table 9.

Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36)
Physical Function Subscale Score Group Means for Positive and Negative
Physical Examination (PE) Measures

                          SF-36 Physical Function Subscale Score

                          Positive PE Measure     Negative PE Measure

Variable                  [bar.X]            SD   [bar.X]            SD

Quadrant test             52                 27   63.5               18
                                    (n=42)                  (n=18)
Pain with lateral         50.9               27   61.7               20
  flexion                           (n=40)                  (n=34)
Pain with lumbar          46.5               25   58.8               24
  flexion                           (n=18)                  (n=56)
Straight-leg-raising      45.2               26   57.9               24
  test                              (n=12)                  (n=62)
Lasegue test              50.2               26   56.8               24
                                    (n=11)                  (n=63)
Hip medial rotation       34.8               22   60.2               23
                                    (n=15)                  (n=53)
Faber test for hip pain   49.3               27   58.4               23
                                    (n=21)                  (n=50)
Trochanteric tenderness   45.2               25   58.8               24
                                    (n=13)                  (n=60)
Lower-extremity           51.0               26   64.3               22
  weakness                          (n=44)                  (n=25)

Variable                  t      p

Quadrant test             1.67   .10

Pain with lateral         1.92   .059
  flexion
Pain with lumbar          1.88   .06
  flexion
Straight-leg-raising      1.66   .101
  test
Lasegue test              0.82   .41

Hip medial rotation       3.84   .0003 (a)

Faber test for hip pain   1.43   .16

Trochanteric tenderness   1.85   .068

Lower-extremity           2.19   .03
  weakness

(a) Significant at P=.001.

Figure 1.

Relative contribution of physical examination measures to Lumbar Spinal
Stenosis questionnaire symptom severity score. The pie chart represents
The variance (percentage) of the contributory variables. "Other"
represents the variance that was unaccounted for.

Quadrant Test              21%
Trochanteric Tenderness    13%
Level of Education         13%
Other                      53%

Quadrant Test [beta]=0.54, P=.0009
Trochanteric Tenderness [beta]=0.68, P=.002
Level of Education [beta]=-0.61, P=.0005

Note: Table made from pie chart.

Figure 2.

Relative contribution of physical examination measures to Lumbar Spinal
Stenosis questionnaire physical function score. The pie chart represents
The variance (percentage) of the contributory variables. "Other"
represents the variance that was unaccounted for.

Level of Education         16%
Trochanteric Tenderness    12%
Other                      72%

Level of Education [beta]=-0.57, P=.004
Trochanteric Tenderness [beta]=0.60, P=.007

Note: Table made from pie chart.

Figure 3.

Relative contribution of physical examination measures to Medical
Outcomes Study 36-Item Short-Form Health Survey (SF-36) bodily pain
subscale score. The pie chart represents the variance (percentage)
of the contributory variables. "Other" represents the variance
that was unaccounted for.

Level of Education         20%
Female                     13%
Other                      67%

Level of Education [beta]=16.5, P=.001
Female [beta]=-12.9, P=.005

Note: Table made from pie chart.

Figure 4.

Relative contribution of physical examination measures to Medical
Outcomes Study 36-Item Short-Form Health Survey (SF-36) bodily pain
subscale score. The pie chart represents te variance (percentage)
of the contributory variables. "Other" represents the variance that
was unaccounted for.

Lumber Extension           13
Trochanteric Tenderness     8%
Level of Education          8%
Other                      71%

Lumber Extension [beta]=-6.7,P=.004
Trochanteric Tenderness [beta]=-19.6,P=.01
Level of Education [beta]=18.5,P=.01

Note: Table made from pie chart.