The costs of chronic low back

Transcription

1 [ research report ] DARREN Q. CALLEY, PT, DScPT, OCS1 Steven Jackson, PT, MSPT2 Heather CollinS, PT, DPT3 Steven Z. George, PT, PhD 4 Identifying Patient Fear-Avoidance Beliefs by Physical Therapists Managing Patients With Low Back Pain Journal of Orthopaedic & Sports Physical Therapy The costs of chronic low back pain to industrialized nations in terms of disability and financial costs are substantial. 43 Identifying which patients are at risk for becoming disabled with chronic low back pain is, therefore, of great importance. 18 Psychosocial factors have shown some ability to predict chronic disability from low back pain, whereas patient characteristics, clinical examination findings, and imaging studies appear to be less predictive. 6,8,18 Cross-sectional and prospective studies have shown a strong relationship with elevated fear-avoidance beliefs and disability in patients with acute and chronic low back pain. 5,8,13,19,21,2 3,32,39,42,59,61 Identification of elevated fearavoidance beliefs is, therefore, a potentially important aspect to consider when evaluating patients with low back pain. Elevated fear-avoidance beliefs are a maladaptive emotional response toward an excessive fear of pain that can eventually lead to avoidance behavior. 41 Fear-avoidance beliefs are used in clinical prediction rules and can be used to guide decision making about biopsycho- social physical therapy approaches. 9,11,30,21,37,38 Two self-report questionnaires that have been validated for quantifying fear- t Study Design: Cross-sectional. t Objectives: To evaluate the accuracy with which physical therapists identify fear-avoidance beliefs in patients with low back pain by comparing therapist ratings of perceived patient fear-avoidance to the Fear-Avoidance Beliefs Questionnaire (FABQ), Tampa Scale of Kinesiophobia 11-item (TSK-11), and Pain Catastrophizing Scale (PCS). To compare the concurrent validity of therapist ratings of perceived patient fear-avoidance and a 2-item questionnaire on fear of physical activity and harm, with clinical measures of fear-avoidance (FABQ, TSK-11, PCS), pain intensity as assessed with a numeric pain rating scale (NPRS), and disability as assessed with the Oswestry Disability Questionnaire (ODQ). t Background: The need to consider psychosocial factors for identifying patients at risk for disability and chronic low back pain has been well documented. Yet the ability of physical therapists to identify fear-avoidance beliefs using direct observation has not been studied. t Methods: Eight physical therapists and 80 patients with low back pain from 3 physical therapy clinics participated in the study. Patients completed the FABQ, TSK-11, PCS, ODQ, NPRS, and a dichotomous 2-item fear-avoidance screening questionnaire. Following the initial evaluation, physical therapists rated perceived patient fear-avoidance on a 0-to-10 scale and recorded 2 influences on their ratings. Spearman correlation and independent t tests determined the level avoidance beliefs are the Fear- Avoidance Beliefs Questionnaire (FABQ) and the Tampa Scale for Kinesiophobia (TSK). 54,58,61 of association of therapist 0-to-10 ratings and 2-item screening with fear-avoidance and clinical measures. t RESULTS: Therapist ratings of perceived patient fear-avoidance had fair to moderate interrater reliability (ICC 2,1 = 0.663). Therapist ratings did not strongly correlate with FABQ or TSK-11 scores. Instead, they unexpectedly had stronger associations with ODQ and PCS scores. Both 2-item screening questions were associated with FABQ-physical activity scores, while the fear of physical activity question was also associated with FABQ-work, TSK-11, PCS, and ODQ scores. t Conclusion: Therapists ratings of perceived patient fear-avoidance were not associated with self-reported fear-avoidance scores, showing a potential disconnect between therapist judgments and commonly used fear-avoidance measures. Instead, therapist ratings had small but statistically significant correlations with pain catastrophizing and disability, findings that may support therapists inability to discriminate fear-avoidance from these other factors. The 2-item screening questions based on fear of physical activity and harm showed potential to identify elevated FABQ physical activity scores. t LEVEL OF EVIDENCE: Differential diagnosis, level 2b. J Orthop Sports Phys Ther 2010;40(12): doi: /jospt t Key Words: FABQ, low back pain, screening 1 Outpatient Clinical Education Coordinator, Instructor in Physical Therapy, College of Medicine, Mayo Clinic, Rochester, MN. 2 Clinic Director, Hammond Clinic, St John, IN. 3 Staff Physical Therapist, Mayo Clinic Arizona, Scottsdale, AZ. 4 Associate Professor, Department of Physical Therapy, Center for Pain Research and Behavioral Health, University of Florida, Gainesville, FL. This study was approved by The Institutional Review Board at Mayo Clinic and by The Institutional Review Board at the University of Maryland. At the time of the study, the primary author was completing a final project for a Doctor of Science in Physical Therapy degree through the University of Maryland. Address correspondence to Dr Darren Calley, 200 1st St SW, Rochester, MN dcalley@mayo.edu 774 december 2010 volume 40 number 12 journal of orthopaedic & sports physical therapy

2 Journal of Orthopaedic & Sports Physical Therapy The TSK and FABQ address similar avoidance-related constructs and correlation between the 2 scales is moderately strong. 54 Pain catastrophizing, the tendency to fear the worst, is a psychological construct related to fear-avoidance and is also associated with disability, physical activity, and quality of life in patients with chronic low back pain. 15,56 Decreasing patient catastrophizing has been advocated to mediate improved functioning in patients with chronic low back pain. 49 While it is clear that identifying psychosocial factors in patients with low back pain is important, each questionnaire utilized in practice increases patient and clinician burden, and the number of available questionnaires continues to grow. 29,31,54 Streamlined identification of patients with elevated fear-avoidance beliefs would decrease patient and clinician burden and likely improve therapist utilization of a fear-avoidance screening tool. Additionally, physical therapists may believe that identifying fear-avoidance beliefs during the clinical exam is intuitive, as several factors observed during the clinical exam have been found to correlate with fear-avoidance scores. 2,3,12,13,20,23,48,55,58 Experienced physical therapists may feel that they are able to discern which patients have elevated fear-avoidance beliefs without administering validated self-report questionnaires. Two possible methods to streamline identification of elevated fear-avoidance beliefs are item and clinician driven. Identifying statistically meaningful items from existing fear-avoidance questionnaires in a large cohort has already been reported. Hart et al 29 identified single items from the FABQ that classified elevated (higher than median scores) physical activity or work fear-avoidance more than 90% of the time. A few studies have examined clinician-driven methods to identify patient psychosocial attributes. Haggman et al 27 found that physical therapists were not able to accurately identify depression based on the clinical exam, a finding that was consistent with physician studies on the same topic. 26,45,51 Jellema and colleagues 33 reported sensitivities of 0.20 to 0.33 from general practitioners judgments of pain catastrophizing, fear-avoidance, and distress after a clinical exam in a group of 143 individuals with acute low back pain. Waddell and Richardson 62 reported moderate clinician interrater reliability for rating overt pain behaviors following multiple pilot studies in a group of 120 individuals with chronic low back pain. While several authors have reported on identifying impairments and pain behaviors during the clinical exam, 34-36,47,60,62,63 none have examined a physical therapist s ability to identify elevated fear-avoidance using direct observation of the patient s clinical presentation. The primary purpose of this study was to evaluate the accuracy with which physical therapists identify patient selfreported fear-avoidance beliefs using direct observation during a clinical exam. The study design permitted comparison of physical therapists judgments of perceived patient fear-avoidance beliefs with commonly used measures, including the FABQ physical activity scale (FABQ-PA), FABQ work scale (FABQ-W), Tampa Scale of Kinesiophobia 11-item (TSK- 11), Pain Catastrophizing Scale (PCS), and clinical measures of disability using the modified Oswestry Disability Questionnaire (ODQ), and pain intensity assessed using a numeric pain rating scale (NPRS). We also investigated physical therapist clinical reasoning by recording 2 therapist-selected influences on their perceived patient fear-avoidance ratings. A secondary purpose was to investigate the concurrent validity of an item-driven approach to identification. A 2-item questionnaire on fear of physical activity and harm was compared to established measures of fear-avoidance, pain catastrophizing, disability, and pain intensity. Responses to the 2-item questionnaire were also compared with predetermined cut-off values of elevated FABQ-PA, FABQ-W, and ODQ scores, to assess the potential value of each question as a screening tool. Methods A ll patients were referred to physical therapy departments at 3 different outpatient clinics in Florida, Arizona, and Minnesota. Eligible patients were individuals between 18 and 65 years of age with low back pain. Patients were excluded if they had a history of lumbar surgery, spinal malignancy, spinal infection, cauda equina syndrome, ongoing pregnancy, or an inability to read and understand English. Patients referred to physical therapy were contacted by phone or in person by the primary investigator at each site to review inclusion and exclusion criteria. This study was approved by The Institutional Review Board at Mayo Clinic, MN and the University of Maryland, MD, and all participating patients completed an approved informed consent form. Physical therapists with at least 2 years experience working with patients with low back pain were recruited for participation in the study. Initial therapist interest in participating was sought by each site primary investigator. All participating therapists fulfilled Institutional Review Board competency requirements for working with patients. Each participating therapist completed personal demographic information, including years of practice, years of practice treating patients with low back pain, specialty certification, familiarity with the FABQ, duration of prior use of the FABQ, personal history of low back pain, and confidence rating (0-10) with the ability to identify patient fear-avoidance. Therapists also completed a selfrated FABQ, which is a modified version of the FABQ, to assess the treating therapist s own fear-avoidance beliefs about low back pain. 46 Therapists attended a 30-minute education session to review the purposes of the study, the concept of fear-avoidance, and the requirements for enrolling patients. Consecutive patients meeting inclusion criteria and agreeing to participate in the study were scheduled on therapist s lists until 10 patients had been evaluated by each therapist. journal of orthopaedic & sports physical therapy volume 40 number 12 december

3 [ research report ] Journal of Orthopaedic & Sports Physical Therapy Patients were asked to complete a standard physical therapy intake form, the FABQ, the TSK-11, the PCS, the ODQ, and 2 dichotomous questions on fear of physical activity and harm, including (1) Are you afraid physical activity will cause an increase in your low back pain? and (2) Are you afraid that moving your back will be harmful to you? The intent of asking these 2 questions was to determine if there was a simple question that could accurately predict elevated fearavoidance beliefs, which could be used as a screening tool clinically to decrease respondent and examiner burden. Wording for question 1 was developed from the FABQ-PA, which asks questions related to fear of physical activity. Wording for question 2 was developed from the TSK- 11, which asks questions related to fear of injury and harm. Feedback on question wording was sought from 2 therapists with experience using the FABQ, and final selection of the 2 questions was made by the primary author. Patients completed all forms prior to the therapist evaluation. The order of completion of the questionnaires was prerandomized for each patient to reduce the potential for order bias. Typically, patients required 15 to 25 minutes to complete the study questions. Once completed, all questionnaires were placed in a coded envelope by the primary site investigator and the treating therapist was blinded to patient responses. Treating therapists were allowed to use the patient-completed standard intake form during the evaluation but were blinded from the remainder of the questionnaires. Therapists completed a standard physical therapy evaluation using therapist-selected physical exam tests and measures. Therapists were specifically instructed not to ask patients how fearavoidant they were or any question similar to it. Following the evaluation, the therapist completed a 0-to-10 perceived rating of the patient s level of fear-avoidance (0, no fear-avoidance; 10, very high fear-avoidance), and selected 2 specific factors from a list of 14 choices of why they rated the patient with this level of fear-avoidance. Therapists were allowed to write in up to 2 factors, if not included on the list, that guided their clinical reasoning. A cut-off score of greater than 5 was selected to designate elevated 0-to-10 therapist-perceived patient fearavoidance ratings in our study, based on a median split of the 0-to-10 scale. To achieve greater than 80% power to reject the null hypothesis that therapist ratings of actual fear-avoidance would have a better than chance association, it was determined a priori that 70 patients would be needed if 40% of the patients studied had elevated FABQ scores. 28 Eighty patients were included in the study to provide additional power. At the end of the therapist evaluation, the intake form, questionnaires, and the therapist 0-to-10 ratings of patient fear-avoidance were placed in a coded envelope and all forms were returned to the primary investigator. The treating therapists remained blinded to gathered patient responses and were not given feedback on their rating accuracy during the study. At the beginning and end of the study, therapists were asked a 0-to-10 confidence rating on their perceived ability to identify patient fear-avoidance beliefs (0, not confident; 10, very confident). Interrater reliability of the physical therapist rating of patient fear-avoidance beliefs was assessed concurrently during data collection on 10 patients. This was accomplished by having 1 therapist perform a videotaped evaluation of 10 consecutive patients and rate their perceived patient level of fear-avoidance. Two other physical therapists then separately viewed the video tape and completed their own ratings of perceived patient fear-avoidance. Videotape has been used in other studies to assess reliability. Fritz et al 17 used videotape to determine the interrater reliability of therapist assessment of the centralization phenomenon in patients with low back pain. Each patient included in the reliability portion of the study gave informed consent to allow the therapy session to be videotaped. Measures Fear-Avoidance Beliefs The FABQ, used to measure fear-avoidance, is a 16-item spine-specific questionnaire consisting of physical activity and work subscales. Responders rate their beliefs about work and activity from 0 to 6. Not all items are used in scoring. The FABQ-PA is scored using questions 2, 3, 4, and 5, with a possible total of 24. The FABQ-W is scored using questions 6, 7, 9, 10, 11, 12, and 15, with a possible total of Higher scores indicate higher levels of fearavoidance. In individuals with chronic low back pain, internal consistency has been reported to have an alpha of.52 to.77 for the FABQ-PA and.84 to.92 for the FABQ-W. 13,61 Scores on the FABQ have been associated with pain, catastrophizing, physical impairment, and disability in patients with low back pain. 24 No universally accepted cut-off scores for the FABQ exist. 7,18,22,23 For the FABQ- W, George et al 22 reported a 2.67 positive likelihood ratio for predicting 6-month improvement in disability, with a FABQ- W score greater than 25 in a cohort of individuals with nonoccupational acute and chronic low back pain. We selected cut-off scores of greater than 25 FABQ- W in our study to classify elevated fearavoidance beliefs for work, based on a cut-off score that has been used previously and is associated with disability from low back pain. 22 Burton et al 7 used a FABQ-PA cut-off score of greater than 14 to designate elevated fear-avoidance for physical activity, based on a median split. We selected a cut-off score of greater than 15 for the FABQ-PA to classify elevated fear-avoidance beliefs for physical activity in our study, a slightly greater than median split value that has been used in other randomized controlled trials. 7,21,25 While use of a median split is a less than ideal method to determine a FABQ-PA cut-off, it is likely the best available method until other studies present empirically derived cut-off scores. Fear of Movement and Injury The TSK is a common scale used to measure fear of movement and injury. 54 A modified december 2010 volume 40 number 12 journal of orthopaedic & sports physical therapy

4 Journal of Orthopaedic & Sports Physical Therapy item TSK has shown similar psychometric properties to the original version. 64 Items are scored from 1 to 4, where 1 is strongly disagree and 4 is strongly agree. A total of 44 points is possible, with higher scores indicating higher levels of kinesiophobia. The TSK-11 assesses fear of harm or injury with movement and activity avoidance that can be used with patients who have multiple areas of pain, whereas the FABQ asks disease-specific questions related to fear of physical activity and work in individuals, specifically, with back pain. Internal consistency of the TSK was an alpha of.68 to.80, when used with individuals with chronic low back pain. 13,58 Spearman correlations between the TSK and the FABQ scales were reported to vary from r = 0.55 to 0.76 (P.001) in individuals with chronic low back pain and r = 0.33 to 0.59 (P =.01) in individuals with acute low back pain. 13,54 Pain Catastrophizing Sullivan et al 35 developed the PCS, a 13-item self-report questionnaire used to quantify patient pain catastrophizing. Items on the PCS are rated on a 5-point scale, and the questionnaire can be subdivided into 3 components of rumination, magnification, and helplessness. The PCS has demonstrated good concurrent validity with anxiety symptoms, and acceptable reliability and internal consistency. 44,52,57 The PCS has been shown to correlate with multiple psychosocial measures, including fear-avoidance and depression. 24 Disability The modified ODQ was used to assess patient disability. The ODQ includes constructs of pain and functional limitations performing common activities of daily living. The ODQ is a 10-item questionnaire, scored from 0 to 5, that has been shown to have good test-retest reliability and face and content validity for disability in patients with low back pain. 16 Scores are doubled to achieve a disability rating out of 100 points possible and higher scores indicate more disability. A cut-off score of greater or equal to 30 was selected in our study to indicate elevated disability, based on values used in other studies, and a score that represents a moderate level of disability. 10 Pain Intensity Patients were asked to rate their worst, best, and current level of pain in the previous 24 hours on a 0-to-10 (0, no pain; 10, maximal pain) numeric pain rating scale. These 3 pain ratings were averaged to create an average numeric pain rating (NPRS). Average 24-hour pain ratings have been shown to be responsive to changes in individuals with acute low back pain and correlate with changes in global ratings of change. 10 Statistical Analysis SPSS Version 16.0 (SPSS Inc, Chicago, IL) and Microsoft Excel 2003 were used for the majority of data analysis. Intraclass correlation coefficient (ICC 2,1 ) values for interrater reliability of therapist ratings were calculated using SAS Version 9.1 (SAS, Cary, NC). Therapist Perception of Fear-Avoidance Compared to Established Measures Spearman correlation was used to determine correlations among therapist perception of fear-avoidance and measures of pain-related fear (FABQ-PA, FABQ-W, and TSK-11), pain catastrophizing (PCS), disability (ODQ), and pain intensity (NPRS). Frequency counts for therapist reasons for making fear-avoidance perception ratings were reported descriptively to supplement these analyses. Likelihood ratios with 95% confidence intervals were calculated using 2-by-2 contingency tables for the most frequent therapist-selected responses to determine relationships with therapistselected reasoning and elevated FABQ- PA and ODQ scores. Validity of 2-Item Screening Questions Independent t tests were used to determine differences in measures of pain-related fear (FABQ-PA, FABQ- W, and TSK-11), pain catastrophizing (PCS), and therapist 0-to-10 perception of patient fear-avoidance, based on the response to the 2-item fear of activity and fear of harm screening questions. Independent t tests were also used to determine differences in clinical measures for disability (ODQ) and pain intensity (NPRS). Then the potential of the screening items was further assessed for ability to identify those with elevated FABQ-PA, FABQ-W, and ODQ scores with 2-by-2 contingency tables and chi-square analyses. Two-by-two contingency tables were used to generate sensitivities, specificities, and resultant likelihood ratios with 95% confidence intervals for the 2 questions on fear of physical activity and harm to predict elevated FABQ-PA and FABQ- W scores. Shifts in pretest to posttest probability of elevated FABQ-PA scores based on responses to the 2-item screening questions were generated specific to the study sample using Bayes theorem. 4 Results One hundred two patients with low back pain, between the ages of 18 and 65 years, referred to the 3 physical therapy clinics between April 2008 and October 2008, were contacted, and 80 patients were enrolled in the study. Reasons for not participating included patients not interested (n = 8), rescheduled/no show (n = 6), symptoms resolved (n = 1), sought care elsewhere (n = 1), and not meeting inclusion criteria once contacted due to prior lumbar surgery (n = 3), thoracic spine pain not low back pain (n = 2), and lumbar infection (n = 1). Baseline scores for enrolled patient demographic and self-report data are found in Table 1. Baseline therapist demographic and self-report scores are found in Table 2. Interrater reliability calculations of 2 therapist s blinded 0-to-10 ratings of perceived patient fear-avoidance from 10 videotaped patients compared with the initial therapist 0-to-10 ratings yielded ICC 2,1 = (95% CI: 0.424, 0.880). Therapist Perception of Fear-Avoidance Compared to Established Measures Average 0-to-10 self-reported therapist confidence in their ratings of patient-perceived fear-avoidance was 5.4 prestudy and 6.3 poststudy (Table 2). Spearman journal of orthopaedic & sports physical therapy volume 40 number 12 december

5 [ research report ] Journal of Orthopaedic & Sports Physical Therapy TABLE 1 correlations of therapist 0-to-10 ratings of perceived patient fear-avoidance with self-reported questionnaires are found in Table 3. Therapist 0-to-10 ratings showed Patient Demographic and Self-Report Measures Patient Variables values* Age (y) Gender (n female patients) 46/80 (57.5%) Duration of pain (wk) Low back pain 3 mo (n) 53/80 (66.3%) Employment (n full-time employed) 56/78 (71.8%) Workman's compensation 7/80 (8.8%) Average 24-h numeric pain rating scale (0-10 rating) Elevated pain, numeric pain rating scale 5 21/80 (26.3%) Self-reported liking to exercise 56/78 (71.8%) Fear-Avoidance Beliefs Questionnaire physical activity scale (score range, 0-24) (0-24) Elevated Fear-Avoidance Beliefs Questionnaire physical activity scale ( 15) 30/80 (37.5%) Fear-Avoidance Beliefs Questionnaire work scale (0-42) (0-39) Elevated Fear-Avoidance Beliefs Questionnaire work scale ( 25) 8/80 (10.0%) Yes to "Are you afraid physical activity will cause an increase in your low back pain?" 44/80 (53.0%) Yes to "Are you afraid that moving your back will be harmful to you?" 21/80 (25.3%) Tampa Scale of Kinesiophobia 11-item (11-44) (16-42) Pain Catastrophizing Scale (0-52) (0-50) Modified Oswestry Disability Questionnaire (0-100) (8-70) Elevated disability (ODQ 30) 37/80 (46.3%) Abbreviation: ODQ, Modified Oswestry Disability Questionnaire. * Values are reported as mean SD (range), where applicable, or n/total sample (percentage). TABLE 2 Therapist Demographic and Self-Report Measures Measures values* Years as a physical therapist Years working with patients with low back pain Familiar with the Fear-Avoidance Beliefs Questionnaire 6/8 (75.0%) Years using the Fear-Avoidance Beliefs Questionnaire An entry-level doctoral degree or higher 2/8 (25.0%) Board specialty certification 3/8 (37.5%) Have acute low back pain 4/8 (50.0%) Have chronic low back pain 1/8 (12.5%) Have recurrent low back pain 2/8 (25.0%) Confidence rating in self-ability to identify fear-avoidance beliefs prestudy (0-10 rating) Confidence rating in self-ability to identify fear-avoidance beliefs poststudy (0-10 rating) Self-rated Fear-Avoidance Beliefs Questionnaire physical activity scale Self-rated Fear-Avoidance Beliefs Questionnaire work scale Elevated self-rated Fear-Avoidance Beliefs Questionnaire scores 0/8 (0.0%) * Values are reported as mean SD, or n/total sample (percentage). statistically significant association with the ODQ and PCS, and approached significance with the NPRS (P =.056). No statistically significant correlation was found between therapist 0-to-10 perceived patient fear-avoidance ratings and the FABQ-PA, FABQ-W, or TSK-11. Therapist-reported reasoning for rating perceived patient fear-avoidance is shown in Table 4. Therapists selected factors that favored lower fear-avoidance in 41 patients (51.3%), higher fear-avoidance in 28 patients (35%), and both low and high fear-avoidance in 11 patients (13.8%). Therapists rated patients to have elevated fear-avoidance 12.5% of the time (greater than 5 on 0-to-10 scale), while actual elevated (greater than 15) FABQ- PA scores were present in 37.5% of the patients. The most common reported influence on therapist ratings was a patient responding liking to exercise (31.3%), followed by patient reported lower pain rating (20.0%). Self-reported liking to exercise was not correlated with elevated FABQ-PA scores (positive likelihood ratio [LR], 1.41; 95% CI: 0.70, 2.84; negative LR, 0.87; 95% CI: 0.64, 1.18) or elevated ODQ scores (positive LR, 1.17; 95% CI: 0.58, 2.37; negative LR, 0.94; 95% CI: 0.71, 1.25). Validity of 2-Item Screening Questions Independent t tests of the afraid of physical activity question showed significant mean differences for all measures of interest, except for the NPRS (Table 5). In all instances, higher fear, catastrophizing, and disability scores were associated with a positive response to the physical activity screening question. Independent t tests of the afraid of harm question showed significant differences for the FABQ-PA only (Table 5). Higher FABQ-PA scores were associated with a positive response to the harm screening question. Further investigation of the screening potential of these questions by chi-square analysis indicated that both items identified elevated FABQ-PA scores better than chance (Table 6), but not elevated FABQ- W or ODQ scores. Follow-up analysis of the fear of physical activity question generated a negative LR of 0.28 (95% CI: 0.12, 0.63) and a positive LR of 2.11 (95% CI: 1.43, 3.09) 778 december 2010 volume 40 number 12 journal of orthopaedic & sports physical therapy

6 Journal of Orthopaedic & Sports Physical Therapy TABLE 3 for identifying elevated (greater than 15) FABQ-PA scores. The fear of harm question generated a negative LR of 0.62 (95% CI: 0.44, 0.89) and a positive LR of 3.2 (95% CI: 1.5, 7.0) for identifying elevated (greater than 15) FABQ-PA scores. Using a 37.5% pretest probability prevalence of elevated FABQ-PA (generated from our cohort and not necessarily representative of other populations), a Association Between Therapist 0-to-10 Ratings and Self-Reported Patient Questionnaires* Variable FABQ-W nprs odq PCS tsk-11 therapist 0-10 FABQ-PA FABQ-W NPRS ODQ PCS TSK Abbreviations: FABQ-PA, Fear-Avoidance Beliefs Questionnaire physical activity scale; FABQ-W, Fear-Avoidance Beliefs Questionnaire work scale; NPRS, numeric pain rating scale; ODQ, Modified Oswestry Disability Questionnaire; PCS, Pain Catastrophizing Scale; TSK-11, Tampa Scale of Kinesiophobia 11-item. * Spearman rank correlation coefficients. n = 80, except for PCS, where n = 79. Correlation is significant at the 0.01 level (2-tailed). Correlation is significant at the 0.05 level (2-tailed). TABLE 4 Frequency of Therapist-Selected Reasoning for Rating Perceived Patient Fear-Avoidance Levels* Reason Frequency Patient likes to exercise 25 (31.3%) Low pain rating 16 (20.0%) No pain behaviors 14 (17.5%) Limited lumbar range of motion 11 (13.8%) Pain behaviors 10 (12.5%) Muscle guarding 10 (12.5%) Good patient effort 10 (12.5%) Slow speed of movement 9 (11.3%) Fast speed of movement 7 (8.8%) High pain rating 6 (7.5%) Positive attitude 6 (7.5%) Doesn't like to exercise 5 (6.3%) Yellow flags present 5 (6.3%) Facial grimacing 4 (5.0%) No muscle guarding 4 (5.0%) Normal lumbar range of motion 4 (5.0%) Other reasons 14 (17.6%) * Values represent frequency (percent) out of 80 possible patients (n = 160). Reasons likely influencing higher therapist-perceived patient fear-avoidance ratings. no response to the fear of physical activity question (negative LR, 0.28) shifts a posttest fear-avoidance probability to 14.4%. 4 A yes response to the fear of harm question (positive LR, 3.2) shifts a posttest fear-avoidance probability to 65.8%. 4 Likelihood ratios of the 2-item questions for elevated (greater than 25) FABQ-W scores generated small shifts in posttest elevated FABQ-W probabilities (negative LR of 0.84 to 2.27 and a positive LR of 0.20 to 1.12). Discussion The primary aim of the present study was to determine if physical therapists could identify elevated patient fear-avoidance beliefs measured by the FABQ, TSK-11, or PCS. Therapist 0-to-10 ratings of perceived patient fear-avoidance were not associated with patient fear-avoidance as measured by FABQ-PA, FABQ-W, and TSK-11, but had statistically significant association with the PCS. Therapists were moderately confident in their ability to identify elevated fear-avoidance beliefs. The difficulty with identifying elevated self-reported fear-avoidance beliefs seen in our study is consistent with the reported difficulty therapists and physicians have had identifying other psychosocial factors. 26,27,45,51 One explanation of our study findings could be that therapists judgments consistently incorrectly interpreted patient fear-avoidance beliefs. Therapist-perceived patient fearavoidance ratings were fair to moderately reliable (ICC 2,1 = 0.663; 95% CI: 0.424, 0.880) but not associated with self-report fear-avoidance questionnaires. This would suggest that the behavioral signs of fear-avoidance may differ from what is assessed on self-report questionnaires like the FABQ and TSK-11. The disconnect reported in our study between observable patient fear-avoidant behaviors and self-reported fear-avoidance measures could bring into question the concurrent validity of the FABQ and TSK-11 scales with behavioral aspects of fear-avoidance. It may be that clinician judgments of perceived patient fear-avoidance, in our study, considered different facets of the complex construct of fear-avoidance that were not captured by the FABQ and TSK-11, such as patient affect and behavior during clinical examination. 14 Another interpretation of these results is that therapist 0-to-10 ratings of fear-avoidance were influenced by differ- journal of orthopaedic & sports physical therapy volume 40 number 12 december

7 [ research report ] Journal of Orthopaedic & Sports Physical Therapy TABLE 5 ent but related constructs, such as pain catastrophizing or disability. In our study, therapist 0-to-10 ratings had small but statistically significant correlation with pain catastrophizing, another major component of the fear-avoidance model. Pain catastrophizing is comprised of rumination, magnification, and helplessness cognitions and has been associated with heightened pain responses in both clinical and experimental studies. 53 It is plausible that heightened pain responses associated with pain catastrophizing might have been more readily apparent to a clinician during examination in Comparison of Group Mean Responses to 2-Item Screening Questions With Self-Reported Patient Questionnaires* yes to Afraid of no to Afraid of Physical Activity Physical Activity yes to Afraid of no to Afraid of Variable (n = 44) (n = 36) P Value Harm (n = 21) Harm (n = 59) P Value FABQ-PA FABQ-W NPRS ODQ PCS TSK Abbreviations: FABQ-PA, Fear-Avoidance Beliefs Questionnaire physical activity scale; FABQ-W, Fear-Avoidance Beliefs Questionnaire work scale; NPRS, numeric pain rating scale; ODQ, Modified Oswestry Disability Questionnaire; PCS, Pain Catastrophizing Scale; TSK-11, Tampa Scale of Kinesiophobia 11-item. * Values represent mean SD, except where indicated otherwise. 2-tailed independent t test. TABLE 6 Chi-Square Analysis of 2-Item Screening Questions With Elevated FABQ-PA, FABQ-W, and ODQ Scores yes to Afraid of no to Afraid of yes to Afraid no to Afraid Variable Physical Activity Physical Activity P Value of Harm of Harm P Value FABQ-PA score Yes 25 (56.8%) 5 (13.9%) 14 (66.7%) 16 (27.1%) No 19 (43.2%) 31 (86.1%) 7 (33.3%) 43 (72.9%) FABQ-W score Yes 5 (11.4%) 3 (8.3%) 1 (4.8%) 7 (11.9%) No 39 (88.6%) 33 (91.7%) 20 (95.2%) 52 (88.1%) ODQ score Yes 23 (52.3%) 14 (38.9%) 10 (47.6%) 27 (45.8%) No 21 (47.7%) 22 (61.1%) 11 (52.4%) 32 (54.2%) Abbreviations: FABQ-PA, Fear-Avoidance Beliefs Questionnaire physical activity scale; FABQ-W, Fear- Avoidance Beliefs Questionnaire work scale; ODQ, Modified Oswestry Disability Questionnaire. comparison to fear-avoidance behaviors, though therapists did not specifically rate pain catastrophizing in our study. Our study found that therapist 0-to-10 ratings had the strongest association with disability scores among all self-reported questionnaires. One reason therapist ratings were associated with elevated ODQ scores could be that the ODQ addresses functional limitations that can be observed during the physical exam, including walking, sitting, and standing. 16 As part of the physical examination process, therapists are trained to use clinical exam findings to form a differential diagnosis and prognosis. 1 Perhaps therapists were better able to identify elevated disability in this study because experienced physical therapists are skilled in identifying functional limitations and pain quantified by the ODQ. These skills may have confounded their ability to make isolated determinations about fear-avoidance. This finding may have implications for therapists ability to make perceived judgments about psychosocial prognoses in patients with low back pain. We attempted to understand therapist-driven identification of patient fear-avoidance by recording 2 therapistselected influences on their ratings of perceived patient fear-avoidance. The most common influence on therapist ratings was patient self-reporting liking to exercise (31.3%), and patient self-reported liking to exercise was high in the study (71.8%). However, self-reported liking to exercise was not associated with FABQ scores in our study and might have been a confounder to accurate fear-avoidance ratings. Therapists listed low pain ratings as an influence 20% of the time, but therapists listing low pain rating as an influence correctly identified only 50% of the patients with low FABQ-PA scores. Therapist underrating perceived patient fear-avoidance in comparison to actual FABQ-PA scores was common in our study. Examiner underrating psychosocial variables has been reported in other studies, particularly among patients with elevated pain. 50,65 Underrating actual scores in our study may be due to the overly positive view physical therapists have about their patients or a variable relationship between the expressional components of fear and self-reported fear-avoidance beliefs. 40 Another aim of the present study was the investigation of dichotomous 2-item screening question association with elevated fear-avoidance, pain catastrophizing, and disability. The fear of physical activity question was broadly associated with FABQ-PA, FABQ-W, TSK-11, PCS, and ODQ scores, while the fear of harm question was associated only with 780 december 2010 volume 40 number 12 journal of orthopaedic & sports physical therapy

8 Journal of Orthopaedic & Sports Physical Therapy FABQ-PA scores. Because the fear of harm question showed differences only with the FABQ-PA, and was able to identify elevated FABQ-PA scores better than chance, it may have greater value as a specific screening tool for elevated FABQ- PA scores. Follow-up chi-square analysis showed that both 2-item questions were associated with elevated (greater than 15) FABQ-PA scores but not elevated FABQ- W or ODQ scores. Use of item-driven identification of fear-avoidance beliefs has been utilized in the literature and may have direct clinical implications. The identification of a single question to identify patients at risk for elevated fear-avoidance scores may have value for increasing exam efficiency and decreasing patient and examiner burden. Hart et al 29 reported that the use of a single FABQ-PA item, I should not do physical activities which (might) make my pain worse, with a cut-off score of 4, showed excellent specificity (0.98) and good sensitivity (0.82), highlighting another possible 1-item screen of fearavoidance beliefs that was developed with item response theory. Limitations and Future Studies There are several potential limitations to our study. Therapists in this study had considerable experience working with patients with low back pain but had limited experience using the FABQ. Perhaps more experience using the FABQ would have improved therapist accuracy in rating fear-avoidance, although it should be noted that clinician experience wasn t a significant factor in other studies examining the screening accuracy of psychosocial factors. 27,33 Several physical exam factors have been shown to correlate with elevated FABQ scores, and our study did not assess therapists awareness of the physical examination factors. 2,3,12,13,20,23,48,55,58 No specific training was given in this study for identifying factors that correlate with elevated fear-avoidance, so that unbiased therapist judgments were assessed, and small improvements have been seen following clinician training in other studies. 33,50 Perhaps training therapists in how to identify factors that correlate with elevated fear-avoidance on exam and allowing therapists to ask specific follow-up questions about fear-avoidance would improve therapist accuracy. The sample size of 80 was chosen a priori on the basis of the number needed to have greater than 80% power to reject the null hypothesis, based on previously reported frequencies of elevated fearavoidance. A larger sample size would have helped to decrease the potential of a type II error. Although the ranges of self-reported questionnaire scores in our study were broad, our modest sample size might have captured a group of individuals with narrow psychosocial profiles, which could have altered therapist rating accuracy. Our study sample was comprised of a small number of patients receiving workers compensation (8.8%), and only 10% of the patients had elevated FABQ-W scores. Because therapist 0-to-10 ratings weakly correlated with the FABQ-W, it may be that therapist judgments of perceived patient fear-avoidance would be different in a population with a higher prevalence of patients receiving workers compensation. The shift in probabilities calculated in our study, based on responses to the 2-item screening questions, was highly sample dependent and may not be representative of the prevalence of elevated fear-avoidance in a larger cohort of individuals with low back pain. It also cannot be assumed that therapist judgments of perceived patient fear-avoidance generalize to other professions or healthcare populations. The primary purpose of our study was to assess therapist rating accuracy of patient fear-avoidance after an initial evaluation; therefore, a patient followup was not included. Because therapist fear-avoidance ratings correlated with elevated baseline disability and pain catastrophizing, it may be of interest to further evaluate if perceived therapist 0-to-10 ratings are able to predict longterm disability outcomes or changes in pain catastrophizing. Further evaluation of the psychometric properties of the 2 screening questions in a larger population is needed to validate the utility of a dichotomous fear-avoidance screening question. Finally, it may be of value to develop a standard clinical tool that captures the behavioral aspects of fearavoidance behaviors and beliefs during routine clinical examination, which may be different from the current construct of fear-avoidance captured by the FABQ and TSK-11. Conclusion Identifying psychosocial factors is potentially valuable for clinicians working with patients with low back pain. Therapist 0-to-10 ratings of perceived patient fear-avoidance had moderate reliability but did not correlate with FABQ or TSK-11 scores. Therapist ratings had small but statistically significant correlations with pain catastrophizing and disability, findings that may support therapists inability to discriminate fear-avoidance from these related factors. The fear of physical activity question had significant associations with all fear-avoidance, pain catastrophizing, and disability measures and showed potential to identify lower FABQ-PA scores. The fear of harm question had significant association specific only to the FABQ-PA and showed potential to identify elevated FABQ-PA scores. t Key Points FINDINGS: Therapist ratings of perceived patient fear-avoidance were not associated with fear-avoidance measures, but were associated with pain catastrophizing and disability scores. Two dichotomous questions related to physical activity and harm showed potential for identifying elevated fear-avoidance beliefs. Implications: Therapists should not rely solely on their own judgments to identify self-reported fear-avoidance beliefs. journal of orthopaedic & sports physical therapy volume 40 number 12 december

9 [ research report ] Journal of Orthopaedic & Sports Physical Therapy Caution: Clinician accuracy for identifying self-reported patient elevated fear-avoidance beliefs might show differences when studied in a patient population with different demographic factors, with training provided ahead of time, or by including therapists that have more experience using the FABQ. Acknowledgments: Special thanks to Roy Bechtel, PT, PhD for guidance with this research project, and to the physical therapy and support staff at Mayo Clinic Rochester, Arizona and Florida for assistance with this project. references 1. Guide to Physical Therapist Practice. Second Edition. American Physical Therapy Association. Phys Ther. 2001;81: Al-Obaidi SM, Al-Zoabi B, Al-Shuwaie N, Al- Zaabie N, Nelson RM. The influence of pain and pain-related fear and disability beliefs on walking velocity in chronic low back pain. Int J Rehabil Res. 2003;26: Al-Obaidi SM, Nelson RM, Al-Awadhi S, Al- Shuwaie N. The role of anticipation and fear of pain in the persistence of avoidance behavior in patients with chronic low back pain. Spine (Phila Pa 1976). 2000;25: Bernstein J. Decision analysis. J Bone Joint Surg Am. 1997;79: Boersma K, Linton SJ. Screening to identify patients at risk: profiles of psychological risk factors for early intervention. Clin J Pain. 2005;21:38-43; discussion Burton AK, Tillotson KM, Main CJ, Hollis S. Psychosocial predictors of outcome in acute and subchronic low back trouble. Spine (Phila Pa 1976). 1995;20: Burton AK, Waddell G, Tillotson KM, Summerton N. Information and advice to patients with back pain can have a positive effect. A randomized controlled trial of a novel educational booklet in primary care. Spine (Phila Pa 1976). 1999;24: Carragee EJ, Alamin TF, Miller JL, Carragee JM. Discographic, MRI and psychosocial determinants of low back pain disability and remission: a prospective study in subjects with benign persistent back pain. Spine J. 2005;5: dx.doi.org/ /j.spinee Childs JD, Fritz JM, Flynn TW, et al. A clinical prediction rule to identify patients with low back pain most likely to benefit from spinal manipulation: a validation study. Ann Intern Med. 2004;141: Childs JD, Piva SR, Fritz JM. Responsiveness of the numeric pain rating scale in patients with low back pain. Spine (Phila Pa 1976). 2005;30: Cleland JA, Childs JD, Fritz JM, Whitman JM, Eberhart SL. Development of a clinical prediction rule for guiding treatment of a subgroup of patients with neck pain: use of thoracic spine manipulation, exercise, and patient education. Phys Ther. 2007;87: ptj Crombez G, Vervaet L, Lysens R, Baeyens F, Eelen P. Avoidance and confrontation of painful, back-straining movements in chronic back pain patients. Behav Modif. 1998;22: Crombez G, Vlaeyen JW, Heuts PH, Lysens R. Pain-related fear is more disabling than pain itself: evidence on the role of painrelated fear in chronic back pain disability. Pain. 1999;80: Davenport TE. How should we interpret measures of patients fear of movement, injury, or reinjury in physical therapist practice? J Orthop Sports Phys Ther. 2008;38: dx.doi.org/ /jospt Elfving B, Andersson T, Grooten WJ. Low levels of physical activity in back pain patients are associated with high levels of fear-avoidance beliefs and pain catastrophizing. Physiother Res Int. 2007;12: Fairbank JC. Use and abuse of Oswestry Disability Index. Spine (Phila Pa 1976). 2007;32: BRS.0b013e31815b9e9d 17. Fritz JM, Delitto A, Vignovic M, Busse RG. Interrater reliability of judgments of the centralization phenomenon and status change during movement testing in patients with low back pain. Arch Phys Med Rehabil. 2000;81: Fritz JM, George SZ. Identifying psychosocial variables in patients with acute work-related low back pain: the importance of fear-avoidance beliefs. Phys Ther. 2002;82: Fritz JM, George SZ, Delitto A. The role of fearavoidance beliefs in acute low back pain: relationships with current and future disability and work status. Pain. 2001;94: Geisser ME, Haig AJ, Wallbom AS, Wiggert EA. Pain-related fear, lumbar flexion, and dynamic EMG among persons with chronic musculoskeletal low back pain. Clin J Pain. 2004;20: George SZ, Fritz JM, Bialosky JE, Donald DA. The effect of a fear-avoidance-based physical therapy intervention for patients with acute low back pain: results of a randomized clinical trial. Spine (Phila Pa 1976). 2003;28: dx.doi.org/ /01.brs A2 22. George SZ, Fritz JM, Childs JD. Investigation of elevated fear-avoidance beliefs for patients with low back pain: a secondary analysis involving patients enrolled in physical therapy clinical trials. J Orthop Sports Phys Ther. 2008;38: George SZ, Fritz JM, McNeil DW. Fear-avoidance beliefs as measured by the fear-avoidance beliefs questionnaire: change in fear-avoidance beliefs questionnaire is predictive of change in self-report of disability and pain intensity for patients with acute low back pain. Clin J Pain. 2006;22: George SZ, Valencia C, Beneciuk JM. A psychometric investigation of fear-avoidance model measures in patients with chronic low back pain. J Orthop Sports Phys Ther. 40: George SZ, Zeppieri G, Jr., Cere AL, et al. A randomized trial of behavioral physical therapy interventions for acute and sub-acute low back pain (NCT ). Pain. 2008;140: Grevitt M, Pande K, O Dowd J, Webb J. Do first impressions count? A comparison of subjective and psychologic assessment of spinal patients. Eur Spine J. 1998;7: Haggman S, Maher CG, Refshauge KM. Screening for symptoms of depression by physical therapists managing low back pain. Phys Ther. 2004;84: Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982;143: Hart DL, Werneke MW, George SZ, et al. Screening for elevated levels of fear-avoidance beliefs regarding work or physical activities in people receiving outpatient therapy. Phys Ther. 2009;89: ptj Hicks GE, Fritz JM, Delitto A, McGill SM. Preliminary development of a clinical prediction rule for determining which patients with low back pain will respond to a stabilization exercise program. Arch Phys Med Rehabil. 2005;86: Hill JC, Dunn KM, Main CJ, Hay EM. Subgrouping low back pain: a comparison of the STarT Back Tool with the Orebro Musculoskeletal Pain Screening Questionnaire. Eur J Pain. 14: ejpain Jellema P, van der Horst HE, Vlaeyen JW, Stalman WA, Bouter LM, van der Windt DA. Predictors of outcome in patients with (sub)acute low back pain differ across treatment groups. Spine (Phila Pa 1976). 2006;31: dx.doi.org/ /01.brs aa 33. Jellema P, van der Windt DA, van der Horst HE, Blankenstein AH, Bouter LM, Stalman WA. Why is a treatment aimed at psychosocial factors not effective in patients with (sub)acute low back pain? Pain. 2005;118: org/ /j.pain Keefe FJ, Block A. Development of an observational method for assessing pain behavior in chronic low back pain patients. Behav Ther. 1982;13: Keefe FJ, Hill RW. An objective approach to quantifying pain behavior and gait patterns in low back pain patients. Pain. 1985;21: december 2010 volume 40 number 12 journal of orthopaedic & sports physical therapy