Do medical factors predict disability in older adults with persistent low back pain?

Do medical factors predict disability in older adults with persistent low back pain?

Pain 112 (2004) 214–220 www.elsevier.com/locate/pain Do medical factors predict disability in older adults with persistent low back pain? Debra K. We...

105KB Sizes 0 Downloads 70 Views

Pain 112 (2004) 214–220 www.elsevier.com/locate/pain

Do medical factors predict disability in older adults with persistent low back pain? Debra K. Weinera,b,c,e,*, Thomas E. Rudyb,c,e,f, Young-Sin Kimg,1, Sara Gollad a

Department of Medicine, Division of Geriatric Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA b Department of Anesthesiology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA c Department of Psychiatry, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA d Department of Radiology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA e Pain Evaluation and Treatment Institute, University of Pittsburgh, Pittsburgh, PA, USA f Department of Biostatistics, University of Pittsburgh, Pittsburgh, PA, USA g University of Pittsburgh School of Medicine, Pittsburgh, PA, USA Received 27 April 2004; received in revised form 18 August 2004; accepted 30 August 2004

Abstract Persistent low back pain (LBP) is one of the most common and challenging persistent pain conditions in older adults. Medical comorbidity also is common in these individuals, but its impact on disability has not been examined. The purpose of this study was, using a cross-sectional design, to examine the functional impact of pain-related and general medical comorbidity on 100 community dwelling older adults (mean age 74.3) with persistent mechanical LBP. Subjects received a structured history and physical examination, lumbosacral spine X-rays, and standardized tests of physical function. Pain-related variables included intensity, duration, extent, and lumbar motion-induced pain. General medical variables included age, comorbidity, number of medications, depressive symptoms, back range of motion, body mass index, and severity of radiographic pathology. Function/disability measures included self-reported disability, gait speed, and mean number of static lifts and amount of work performed during a dynamic lifting task. Structural equation modeling was used to evaluate the influence of pain and medical variables on function/disability. The overall regression model indicated pain and medical variables were significantly associated with function/disability measures (R2Z0.45, P!0.01). Individual regression coefficients, however, indicated that only pain duration (rZK0.36, P!0.05) and pain severity (rZ0.37, P!0.001) were significantly associated with function/disability. Despite the prevalence of medical co-morbidities in older PLBP subjects, they appear to be of limited utility in understanding level of disability. These findings also underscore the need to optimize pain treatment in independent older adults to optimize physical function and delay the onset of dependent living status. q 2004 Published by Elsevier B.V. on behalf of International Association for the Study of Pain. Keywords: Low back pain

1. Introduction Pain is one of the most common symptoms with which older adults present to their primary care providers

* Corresponding author. Address: Research Department, UPMC Pain Medical Program at Centre Commons, 5750 Centre Avenue, Suite 400, Pittsburgh, PA 15206, USA. Tel.: C1 412 665 8051; fax: C1 412 665 8067. E-mail address: [email protected] (D.K. Weiner). 1 Young-Sin Kim is a second year medical student at the University of Pittsburgh School of Medicine.

worldwide (Elliott et al., 1999; Gureje et al., 2001; Hasselstrom et al., 2002; Mantyselka et al., 2001), and low back pain (LBP) is among the most disabling and therapeutically challenging pain conditions (Leveille et al., 1999; Mantero-Atienza et al., 1992; Weiner et al., 2003). Multiple studies attest to the widespread under-treatment of pain in older adults, and the diverse settings in which pain under-management or mismanagement errors occurs (Bernabei et al., 1998; Feldt et al., 1998). Because LBP is often recurrent or persistent (Lawrence et al., 1998), primary health care providers constantly grapple with how best to evaluate and manage this clinical syndrome, and

0304-3959/$20.00 q 2004 Published by Elsevier B.V. on behalf of International Association for the Study of Pain. doi:10.1016/j.pain.2004.08.027

D.K. Weiner et al. / Pain 112 (2004) 214–220

often operate under the assumption that gathering objective medical information (e.g. X-rays, MRIs, physical examination findings) will help to guide treatment decisions (Bertakis et al., 2003). Although persistent pain may impact a wide range of parameters pertinent to older adults (e.g. mood, physical function, sleep, appetite), loss of independence related to impaired physical function is of central importance (Leveille et al., 1999). Prior studies in younger LBP patients have failed to identify medical factors (e.g. spinal flexibility, severity of radiographic disease, physical examination findings) that predict physical function or its response to pain treatment (Michel et al., 1997; Parks et al., 2003; Poitras et al., 2000; Sullivan et al., 2000; Waddell and Burton, 2001). A variety of medical factors have been shown to impact physical function in older adults regardless of pain status, such as comorbidity (Rozzini et al., 1997), body mass index (BMI; (Landi et al., 1999)), and polypharmacy (Rozzini et al., 1997). Severity of radiographic pathology has been shown to correlate with spinal impairment in pain-free older adults (Weiner et al., 1994), but this has not been examined in older adults with LBP. The role of medical factors in the older adult with persistent pain has not been thoroughly evaluated. Farrell and colleagues demonstrated that medical comorbidity predicts decreased activity in older adult pain clinic patients (Farrell et al., 1995). Others have demonstrated a strong association between LBP and self-reported functional difficulties in disabled (Leveille et al., 1999) and nondisabled older adults (Weiner et al., 2003). Reid and associated recently demonstrated a relationship between depressive symptoms and activity restriction in older adults with LBP (Reid et al., 2003). These studies, however, did not comprehensively evaluate the influence of medical factors (e.g. medications, spinal range of motion, and X-ray findings) on physical function. In addition, while these studies included self-reported as well as performance-based measures of physical function, the performance-based measures were not axial-specific, and no association between pain severity and performance was demonstrated. Given the prevalence of non-pain-related comorbidity and polypharmacy, the ubiquitous nature of degenerative radiographic pathology, and the generalized nature of osteoarthritis in older adults as compared with younger individuals, the relative impact of these factors on physical function in the older adult with LBP is important for defining evaluative strategies and prioritizing treatment efforts. The purpose of this investigation was to determine the impact of medical factors routinely evaluated in the context of primary care on physical function, assessed using robust measures targeted to the axial skeleton, in community dwelling older adults with persistent LBP. We hypothesized that: (1) comorbidity, number of medications, depressive symptoms, and BMI would be positively associated with higher levels of low back-related physical disability, but that

215

severity of radiographic pathology and spinal flexibility would not; and (2) pain severity would be a stronger predictor of physical function and disability than any medical factor.

2. Methods 2.1. Subjects Subjects were recruited using newspaper advertisements and mass mailings. They consisted of 100 English-speaking community dwelling older adults (age 65–84) with persistent LBP, defined as pain of at least moderate intensity, every day or almost every day, for at least the past 3 months (mean pain durationZ15.5 years). All subjects were cognitively intact and signed informed consent prior to their participation. They were screened in two phases, first over the telephone, then on site by one of the investigators (DW) with a structured history and physical examination. Exclusion criteria included cognitive impairment (Folstein mini-mental state examination !21 adjusted for age and education), severe visual or hearing impairment, acute illness or pain, medical conditions that could make the lifting task (see below) potentially unsafe (e.g. postural instability, severe cardiac or respiratory disease), or X-ray evidence of O 20 degree scoliosis or vertebral compression fractures. Subject demographics are shown in Table 1. 2.2. Procedures All subjects underwent the following assessments.

Table 1 Demographic characteristics of study sample Age (mean in years) Duration of pain (mean in years) Gender (%) Female Male Ethnic group (%) African American Caucasian Asian Education (%) High school graduate Some college (or trade school) College graduate Marital status (%) Single Divorced/separated Married Widowed G are SD (standard deviations).

74.3G4.7 15.5G16.5 42.9 57.1 9.2 88.8 2.0 21.4 18.4 52.1 3.1 12.2 68.4 16.3

216

D.K. Weiner et al. / Pain 112 (2004) 214–220

2.2.1. Pain severity 1. Pain intensity was measured with the McGill Pain Questionnaire Short Form (MPQ-SF) that has been validated in community dwelling older adults with persistent LBP (Gangliese and Melzack, 1997; Weiner et al., 1996). 2. Pain extent was the number of painful non-low back musculoskeletal (e.g. neck, shoulders, hips) and nonmusculoskeletal (e.g. abdomen, bladder) body sites detected by a structured history and physical examination. 3. Lumbar motion-induced pain was assessed by having subjects maximally flex (forward, left, right), extend and rotate (left and right) their lower back, and report whether or not they experience pain with these maneuvers. A summary score (0–6) was then created. Inter-rater reliability was 0.85 (intraclass R) for this summary score, which was calculated for the first 18 subjects, who were examined by two physicians on the same day. 4. The number of pain medications was the total of all currently active prescriptions for pain as well as overthe-counter analgesics. This method was used instead of calculating opioid equivalents because of the low use of opioid medications in this sample (7%).

2.2.2. Pain duration Pain duration was expressed as a continuous variable in years, estimated by subjects’ response to the following question: ‘Tell me the month and year when your back pain first began?’ 2.2.3. Depressive symptoms Depressed mood was assessed with the Geriatric Depression Scale (GDS; (Yesavage et al., 1983)), a 30item self-report scale of depression developed and normed for older adults. 2.2.4. Body mass index (BMI) BMI was included as a potential predictor because it has been shown to be predictive of functional dependence in other populations (Landi et al., 1999), and was calculated using standard procedures. 2.2.5. General medical comorbidity 1. Medical comorbidity was assessed using the cumulative illness rating scale (Linn et al., 1968), collected during the structured history and physical examination. This well-validated instrument is comprised of 14 body system items (e.g. cardiac, hypertension, respiratory, psychiatric/behavioral) each of which is scored from 0 (no impairment to that organ/system) to 4 (extremely severe impairment, that is, impairment is life threatening; treatment is urgent or of no avail; prognosis is grave,

for example myocardial infarction, cerebrovascular accident, gastrointestinal bleeding, or embolus). A summary score is then created. 2. Number of medications was counted and expressed as a summed score.

2.2.6. Lumbar Impairment Back range of motion was measured because of its common inclusion in disability determination. This was assessed using a gravity goniometer (Burdett et al., 1986). Mean lumbar flexion over two trials, performed on the same day, was calculated. Test–retest reliability (intraclass R) between the two trials was 0.75. 2.2.7. Radiographic pathology Severity of degenerative lumbosacral pathology was scored using an established system (Weiner et al., 1994), with disc and facet involvement scored for each level (T12 through S1; 0, no disease, 3, severe disease). Disc and facet summed scores were then created. 2.2.8. Physical function/disability measures 2.2.8.1. Observed measures. 1. A lifting task that has been validated in young and older patients with LBP (Boston et al., 1995; Rudy et al., 2003; Slaboda et al., 2002) was performed on a Work Simulator (Baltimore Therapeutic Equipment (BTE) Company, Baltimore, MD, USA), with the subject standing on a force platform (AMI OR-6, MA, USA). Subjects lifted a 12-inch handle (using 40% of the static lift weight; see below) from knee to waist level, then returned the handle to the holder. The BTE Work Simulator applied resistance during the up-phase of the lift. Subjects performed the task for a maximum of 15 min, with a 15-s rest interval between lifts. Lifting strength was measured by calculating the mean of three trials of a static lift at knee level, using a Chatillon Muscle Strength Dynamometer (Sammons Preston, Bolingbrook, IL, USA). A work index was calculated as the weight lifted times the number of lifts performed during the lifting task. Two subjects were eliminated from further analysis because of equipment failure during the lifting task. 2. Gait speed was included as a general measure of physical function because of its predictive validity for disability in older adults (Guralnik et al., 2000). Twenty-five foot gait speed was assessed based on the methods of Bohannon (Bohannon, 1997).

2.2.8.2. Self-reported measures. 1. Back-specific disability was assessed with the Roland and Morris questionnaire, a scale developed to measure

D.K. Weiner et al. / Pain 112 (2004) 214–220

disability specifically in patients with LBP, and validated in older adults (Roland and Morris, 1983; Weiner et al., 1996). 2. Functional status was measured with the pain subscale of the Functional Status Index, that defines amount of pain experienced in performing specific activities of daily living (Jette, 1980). 2.2.8.3. Data analysis. The LISREL V8.54 computer program (Jo¨reskog and So¨rbom, 1996a) was used to compute multivariate regression with latent variables (often referred to as structural equation modeling (SEM) with latent variables). Full-information maximum-likelihood (FIML) estimation procedures were used. The model tested contains two primary components: (1) a measurement model and (2) a regression model. The measurement model is simply a confirmatory factor analysis (CFA) model that specifies which measured or observed variables are hypothesized to load or correlate with which latent dimensions or constructs. For example, four measures were used to operationalize the pain severity construct, pain intensity, pain extent, motioninduced pain, and pain medications. When multiple measures of a construct are available, CFA allows the extraction of that component (covariance) of each measure related to a common dimension. This approach reduces the effects of errors in measurement by using only common covariations among the observed measures in the structural equation component, thus correcting for attenuation among the latent constructs. The regression component specifies the hypothesized association among the latent variables or constructs, that is, which constructs are predictors, which are criterion, and which simply covary or correlate. In this

217

application of SEM, a direct extension of multiple regression was used, with the physical disability construct serving as the criterion variable and the remaining measures and constructs as the predictor variables. A preliminary data analytic step in SEM applications is the examination of the distributions of the measures and the need for appropriate transformations before correlations or covariance are computed. The PRELIS V2.54 computer program (Jo¨reskog and So¨rbom, 1996b) was used to conduct preliminary screening of the 18 observed measures used in the SEM. PRELIS also was used to standardize the measure and to compute the correlation matrix that was used by the LISREL program to conduct the CFA and SEM regression analyses.

3. Results 3.1. Descriptive statistics Descriptive statistical and distributional information for the 18 variables used in the SEM is provided in Table 2. All measures displayed adequate ranges, and inspection of the skewness for these 18 scales indicated reasonably normally distributed measures (skewness for all measures !j2.0j). 3.2. SEM results 3.2.1. Measurement model The measurement or CFA component of our model indicated a good fit of the measures to their intended constructs. Maximum-likelihood factor loading estimates,

Table 2 Descriptive and distributional statistics for observed measures used in SEM (nZ98) Construct/measure Pain severity MPQ-SF No. of other painful body sites Lumbar motion-induced pain No. of pain medications Age (in years) Pain duration (in years) Depressive symptoms (GDS) General medical comorbidity CIRS No. of medications Back ROM Radiographic pathology Disc severity Facet severity Body mass index (BMI) Physical function/disability Back disability questionnaire Functional status index Gait speed (s) Static lift (lbs) Work index

Mean

SD

Skewness

Range

11.67 1.59 2.66 1.15 74.28 15.54 4.22

8.38 0.85 1.92 0.80 4.73 16.46 3.70

0.89 1.01 0.19 1.05 0.14 1.03 1.42

1–33 1–4 0–6 0–4 65–84 0.33–60.0 0–18

9.16 5.20 15.01

3.21 2.35 7.45

0.19 0.38 0.08

2–17 1–13 0–35

6.79 11.53 29.08

3.22 4.43 4.27

0.07 0.42 0.74

0–16 0–25 20.5–45.2

8.26 0.43 14.11 108.28 1698.09

5.12 0.32 3.54 24.78 864.75

0.14 0.80 1.49 0.47 0.46

0–19 0–1.5 9.3–36.5 33.3–221.7 26.6–4078.7

SEM, structural equation modeling; MPQ-SF, McGill pain questionnaire short form; CIRS, cumulative illness rating scale; ROM, range of motion.

218

D.K. Weiner et al. / Pain 112 (2004) 214–220

Table 3 Factor loadings for measurement model Construct/measure

Construct Pain severity

Pain severity MPQ-SF No. of other painful body sites Lumbar motion-induced pain No. of pain medications General medical comorbidity CIRS No. of medications Radiographic pathology Disc severity Facet severity Physical function/disability Back disability questionnaire Functional status index Gait speed (s) Static lift (lbs) Work index

Medical comorbidity

X-rays

Physical disability

0.61 0.25a 0.42 0.08a 0.81 0.80 0.45 0.75 0.80 0.75 0.54 K0.50 K0.60

MPQ-SF, McGill pain questionnaire short form; CIRS, cumulative illness rating scale. a These loadings are not significant at P!0.05, all other loadings significant at P!0.001.

which can range from K1.0 to 1.0, are provided in Table 3. All the measures selected to represent the constructs of General Medical Comorbidity, Radiographic Pathology, and Physical Function/Disability showed large and statistically significant factor loadings. This suggests that these individual measures or indicators adequately represented their intended constructs. However, the number of other painful body sites and the number of pain medications did not fit adequately the pain severity construct and were eliminated from further analyses. 3.2.2. Regression model The primary aim of this study was to evaluate the effects of the constructs of pain severity, general medical comorbidity, depressive symptoms, and radiographic pathology on physical disability in older adults with chronic LBP. In addition, age, gender, pain duration, back range of motion, and body mass index were included in the SEM regression model as separate predictors of physical disability. The regression component of our model indicated that, collectively, these nine predictors accounted for a significant proportion of the variance in physical disability (R2Z0.45, P!0.01). The results of this regression analysis for each of the hypothesized predictors is presented in Table 4. Inspection of the individual regression coefficients, however, indicated that only pain duration (rZK0.36, P!0.05) and pain severity (rZ0.37, P!0.001) were significantly associated with function/disability.

4. Discussion This study of independent community dwelling older adults with persistent LBP indicates that medical data

commonly collected in primary care settings, including X-rays, body mass index, spinal flexibility, and general medical comorbidity, predict neither physical disability nor physical performance. Pain severity, on the other hand, did predict low back-related disability and surprisingly, shorter duration of back pain predicted greater disability. It is worth noting that all self-report and performance-based factors loaded significantly on the physical function/disability construct, indicating that this battery of measures is generally reflective of function and disability in older adults with persistent LBP. The lack of association between spinal flexibility and function/disability is consistent with existing literature in younger patients with persistent LBP (Parks et al., 2003; Poitras et al., 2000). While radiographic pathology has been shown to modestly correlate with spinal flexibility but not with pain (Weiner et al., 1994), the relationship between radiographic pathology and disability in older adults with Table 4 Predictors of physical disability, SEM regression results (nZ98) Model parameter

Standardized beta

Standard error

t-value

P-value

Age Gender Pain duration Back range of motion Body mass index Pain severity Depressive symptoms General medical comorbidity Radiographic pathology

0.22 0.07 K0.36 K0.09 0.03 0.37 0.06

0.20 0.17 0.17 0.12 0.21 0.11 0.11

1.11 0.38 K2.16 K0.77 0.16 3.55 0.53

0.270 0.705 0.033 0.444 0.873 0.001 0.597

K0.17

0.15

K1.12

0.265

0.24

0.31

0.77

0.443

D.K. Weiner et al. / Pain 112 (2004) 214–220

LBP has not been investigated. We have done so in this study, and have failed to identify a significant association. Despite the frequency with which X-rays are ordered in patients with mechanical LBP, our findings in combination with that of others (Jarvik and Deyo, 2002; van Tulder et al., 1997) indicate that this practice is not helpful either in guiding treatment or in predicting disability. While other studies have indicated that medical comorbidity impacts physical function, our findings do not support this relationship. This discrepancy may relate to the fact that Farrell and colleagues studied pain clinic patients (Farrell et al., 1996), while our subjects were community dwelling older adults who happened to have LBP and were generally high functioning. Thus, while medical comorbidity was certainly prevalent in our subjects, the impact of this comorbidity on their back-related physical performance and self-reported disability was undetectable. Our results also failed to demonstrate a relationship between depressive symptoms and physical disability, contrary to that of others (Reid et al., 2003). As shown in Table 2, the prevalence of depressive symptomatology was low in our subjects, attesting to the fact that this was a healthy, community dwelling sample. Our results, therefore, are applicable to these types of individuals, and probably cannot be generalized to older adults attending pain clinics, who may have a greater burden of depressive symptoms than observed in our sample. Our findings underscore the importance of carefully assessing pain severity when trying to determine its impact on physical function and disability. Both pain intensity and lumbar motion-induced pain significantly loaded on the pain severity factor in our analytic model. The short form McGill pain questionnaire is likely too time consuming a measure to be used in the primary care setting, although a more feasible pain self-report instrument such as the pain thermometer, as suggested by recently published clinical guidelines, should be strongly considered (AGS Panel on Persistent Pain in Older Persons, 2002). Assessment of lumbar motioninduced pain is also highly feasible, and our results suggest that incorporation of such an assessment into routine clinical care may provide useful information. The lack of significant loading of number of pain medications on the pain severity factor probably relates to the infrequency with which pain medications were used, despite the fact that all subjects reported at least moderate pain. This observation suggests that while polypharmacy is common in older adults, pain medications do not appear to be over-utilized. The lack of loading of ‘other painful sites’ on this factor likely relates to the fact that prominent LBP, in excess of pain at other sites, was a criterion for inclusion in our study. The inclusion of subjects with prominent pain at other sites would have likely increased our chances of detecting an additive effect of ‘other pain’ and LBP on physical performance and selfreported disability. The inverse relationship between pain duration and physical disability (i.e. the shorter the duration of pain,

219

the greater the disability) was an unexpected finding, not previously demonstrated in studies of younger patients with persistent LBP. These findings clearly contradict the commonly held belief that chronicity automatically implies disability. One possible explanation is that greater duration of pain may allow more time for the development of more effective coping skills such as acceptance (McCracken et al., 2004) and, therefore, less impact of pain on physical dysfunction. Further exploration of this phenomenon warrants additional research efforts. While this study had many strengths, its limitations must also be pointed out. First, while pain of at least moderate intensity was required for study participation, the mean score on the McGill Pain Questionnaire was modest. Thus, our findings may not be applicable to older adults with more severe low back pain. The low rate of opioid use underscores the relatively modest levels of pain experienced by our research subjects, and prevented us from examining the relationship between opioid equivalents (as a proxy for pain severity) and disability. Second, examination of several psychosocial factors that contribute to disability (e.g. fear, low self-efficacy, and poor coping skills) was deliberately excluded from this investigation to allow specific focus on medical factors routinely examined by primary care providers. Third, the cross-sectional nature of the data allows for an examination of an association between medical factors and disability, but not necessarily causation. Longitudinal studies are needed to definitively determine the relationship between pain, disability, and other important mediating factors. Our findings emphasize that much of the medical data collected in busy practice settings are of limited use in understanding pain-related disability in older adults with persistent mechanical LBP, but that careful evaluation of pain severity and duration is key. While practitioners may, for a variety of reasons, feel compelled to gather such data (pressure from the patient and/or family, the desire to protect oneself against litigation), they must not allow this pressure to cause inappropriate treatment decisions. Our findings also underscore the importance of optimizing pain reduction for independent, community dwelling older adults. While all of our subjects were independent, many had low back pain-related functional impairment. Recognition of this fact should encourage the development of research efforts to maintain maximum physical function in these individuals, reduce future disability, and forestall dependence.

Acknowledgements This research was supported by USPHS Research Grants R01AG18299 and R01AT000985 from the National Institute on Aging and National Center for Complementary and Alternative Medicine, National Institutes of Health, Bethesda, MD 20892.

220

D.K. Weiner et al. / Pain 112 (2004) 214–220

References AGS Panel on Persistent pain in Older Persons. The management of persistent pain in older persons. J Am Geriatr Soc 2002;50:S205–S224. Bernabei R, Gambassi G, Lapane K, Landi F, Gatsonis C, Dunlop R, Lipsitz L, Steel K, Mor V. Management of pain in elderly patients with cancer. J Am Med Assoc 1998;279:1877–82. Bertakis KD, Azari R, Callahan FJ. Patient pain: its influence on primary care physician-patient interaction. Fam Med 2003;35:119–23. Bohannon RW. Comfortable and maximum walking speed of adults aged 20–79 years: reference values as determinants. Age Aging 1997;26: 15–19. Boston JR, Rudy TE, Lieber SJ, Stacey BR. Measuring treatment effects on repetitive lifting for patients with chronic low back pain: speed, style, and coordination. J Spinal Disord 1995;8:342–51. Burdett RG, Brown KE, Fall MP. Reliability and validity of four instruments for measuring lumbar spine and pelvic positions. Phys Ther 1986;66:677–84. Elliott AM, Smith BH, Penny KI, Smith WC, Chambers WA. The epidemiology of chronic pain in the community. Lancet 1999;354: 1248–52. Farrell MJ, Gerontol M, Gibson SJ, Helme RD. The effect of medical status on the activity level of older pain clinic patients. J Am Geriatr Soc 1995; 43:102–7. Farrell MJ, Gibson SJ, Helme RD. Measuring the activity of older people with chronic pain. Clin J Pain 1996;12:6–12. Feldt KS, Ryden MB, Miles S. Treatment of pain in cognitively impaired compared with cognitively intact older patients with hip-fracture. J Am Geriatr Soc 1998;46:1079–85. Gangliese L, Melzack R. The assessment of pain in the elderly. Handbook of pain and aging. London: Plenum; 1997 pp. 69–96. Guralnik JM, Ferrucci L, Pieper CF, Leveille SG, Markides KS, Ostir GV, Studenski S, Berkman LF, Wallace RB. Lower extremity function and subsequent disability: consistency across studies, predictive models, and value of gait speed alone compared with the short physical performance battery. J Gerontol: Med Sci 2000;55A:M221–M231. Gureje O, Simon GE, Von Korff M. A cross-national study of the course of persistent pain in primary care. Pain 2001;92:195–200. Hasselstrom J, Liu-Palgren J, Rasjo-Wraak G. Prevalence of pain in general practice. Eur J Pain 2002;6:375–85. Jarvik JG, Deyo RA. Diagnostic evaluation of low back pain with emphasis on imaging. Ann Intern Med 2002;137:586–97. Jette AM. Functional status index: reliability of a chronic disease evaluation instrument. Arch Phys Med Rehabil 1980;61:395–401. Jo¨reskog KG, So¨rbom D. LISREL 8: user’s reference guide. Chicago, IL: Scientific Software International; 1996. Jo¨reskog KG, So¨rbom D. PRELIS 2 user’s reference guide: a program for multivariate data screening and summarization. Chicago, IL: Scientific Software International; 1996. Landi F, Zuccala G, Gambassi G, Incalzi RA, Manigrasso L, Pagano F, Carboni P, Bernabei R. Body mass index and mortality among older people living in the community. J Am Geriatr Soc 1999;47:1072–6. Lawrence RC, Helmick CG, Arnett FC, Deyo RA, Felson DT, Giannini EH, Heyse SP, Hirsh R, Hochberg MC, Hunder GG, Liang MH, Pillemer SR, Steen VD, Wolfe F. Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States. Arthritis Rheum 1998;41:778–99. Leveille SG, Guralnik JM, Hochberg M, Hirsch R, Ferrucci L, Langlois J, Rantanen T, Ling S. Low back pain and disability in older women:

independent association with difficulty but not inability to perform daily activities. J Gerontol Series A—Biol Sci Med Sci 1999;54: M487–M493. Linn BS, Linn MW, Gurel L. Cumulative illness rating scale. J Am Geriatr Soc 1968;16:622–6. Mantero-Atienza E, Beach RS, Sotomayor MG, Christakis G, Baum MK. Nutritional status of institutionalized elderly in South Florida. Arch Latinoamer Nutr 1992;42:242–9. Mantyselka P, Kumpusalo E, Ahonen R, Kumpusalo A, Kauhanen J, Viinamaki H, Halonen P, Takala J. Pain as a reason to visit the doctor: a study in Finnish primary health care. Pain 2001;89:175–80. McCracken LM, Carson JW, Eccleston C, Keefe FJ. Acceptance and change in the context of chronic pain. Pain 2004;109:4–7. 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. Parks KA, Crichton KS, Goldford RJ, McGill SM. A comparison of lumbar range of motion and functional ability scores in patients with low back pain. Spine 2003;28:380–4. Poitras S, Loisel P, Prince F, Lemaire J. Disability measurement in persons with back pain: a validity study of spinal range of motion and velocity. Arch Phys Med Rehabil 2000;81:1394–400. Reid MC, Williams CS, Concato J, Tinetti M, Gill TM. Depressive symptoms as a risk factor for disabling back pain in communitydwelling older persons. J Am Geriatr Soc 2003;51:1710–7. Roland M, Morris R. A study of the natural history of back pain-part I: development of a reliable and sensitive measure of disability in lowback pain. Spine 1983;8:141–4. Rozzini R, Frisoni GB, Ferrucci L, Barbisoni P, Bertozzi B, Trabucchi M. The effect of chronic diseases on physical function. Comparison between activities of daily living scales and the physical performance test. Age Ageing 1997;26:281–7. Rudy TE, Boston JR, Lieber SJ, Kubinski JA, Stacey BR. Body motion during repetitive isodynamic lifting: a comparative study of normal subjects and low-back pain patients. Pain 2003;105:319–26. Slaboda JC, Rudy TE, Boston JR, Lieber SJ. Lifting characteristics of the elderly with chronic lower back pain. Proc. IEEE-EMBS/BMES 2002; 45:2465–6. Sullivan MS, Shoaf LD, Riddle DL. The relationship of lumbar flexion to disability in patients with low back pain. Phys Ther 2000; 80:240–50. van Tulder MW, Assendelft WJ, Koes BW, Bouter LM. Spinal radiographic findings and nonspecific low back pain. A systematic review of observational studies. Spine 1997;22:427–34. Waddell G, Burton AK. Occupational health guidelines for the management of low back pain at work: evidence review. Occup Med 2001;51: 124–35. Weiner DK, Distell B, Studenski S, Martinez S, Lomasney L, Bongiorni D. Does radiographic oseoarthritis correlate with flexibility of the lumbar spine? J Am Geriatr Soc 1994;42:257–63. Weiner D, Pieper C, McConnell E, Martinez S, Keefe F. Pain measurement in elders with chronic low back pain: traditional and alternative approaches. Pain 1996;67:461–7. Weiner DK, Haggerty CL, Kritchevsky SB, Harris T, Simonsick EM, Nevitt M, Newman A, for the Health, Aging and Body Composition Research Group. How does low back pain impact physical function in independent, well-functioning older adults? Evidence from the Health ABC cohort and implications for the future Pain Med 2003;4:311–20. Yesavage JA, Brink TL, Rose TL, Lum O. Developement and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res 1983;17:37–49.