SPINE Volume 39, Number 17, pp 1393-1400 ©2014, Lippincott Williams & Wilkins
CLINICAL CASE SERIES
Do Comorbid Fibromyalgia Diagnoses Change After a Functional Restoration Program in Patients With Chronic Disabling Occupational Musculoskeletal Disorders? Meredith M. Hartzell, MS,* Randy Neblett, MA, LPC, BCB,* Yoheli Perez, PT, DPT,* Emily Brede, PhD, RN,* Tom G. Mayer, MD,† and Robert J. Gatchel, PhD, ABPP‡
Study Design. A retrospective study of prospectively collected data. Objective. To determine whether comorbid fibromyalgia, identified in patients with chronic disabling occupational musculoskeletal disorders (CDOMDs), resolves with a functional restoration program (FRP). Summary of Background Data. Fibromyalgia involves widespread bodily pain and tenderness to palpation. In recent studies, 23% to 41% of patients with CDOMDs entering an FRP had comorbid fibromyalgia, compared with population averages of 2% to 5%. Few studies have examined whether fibromyalgia diagnoses resolve with any treatment, and none have investigated diagnosis responsiveness to an FRP. Methods. A consecutive cohort of patients with CDOMDs (82% with spinal disorders and all reporting chronic spinal pain) and comorbid fibromyalgia (N = 117) completed an FRP, which included quantitatively directed exercise progression and multimodal disability management. Diagnosis responsiveness, evaluated at discharge, created 2 groups: those who retained fibromyalgia and those who did not. These groups were compared with chronic regional lumbar pain only patients (LO group, n = 87), lacking widespread pain and fibromyalgia. Results. Of the patients with comorbid fibromyalgia, 59% (n = 69) retained the fibromyalgia diagnosis (RFM group) and 41% (n = 48) From the *PRIDE Research Foundation, Dallas, TX; †Department of Orthopedic Surgery, University of Texas Southwestern Medical Center, Dallas, TX; and ‡Department of Psychology, College of Science, University of Texas at Arlington, Arlington, TX. Acknowledgment date: October 15, 2013. First revision date: February 27, 2014. Second revision date: April 13, 2014. Acceptance date: April 22, 2014. The manuscript submitted does not contain information about medical device(s)/drug(s). No funds were received in support of this work. Relevant financial activities outside the submitted work: consultancy, grants, employment, royalties, stocks. Address correspondence and reprint requests to Tom G. Mayer, MD, PRIDE Research Foundation, 5701 Maple Ave, Ste 100, Dallas, TX 75235; E-mail: [email protected] DOI: 10.1097/BRS.0000000000000406 Spine
lost the fibromyalgia diagnosis (LFM group) at discharge. Although all 3 groups reported decreased pain intensity, disability, and depressive symptoms from admission to discharge, RFM patients reported higher symptom levels than the LFM and LO groups at discharge. The LFM and LO groups were statistically similar. At 1-year follow-up, LO patients demonstrated higher work retention than both fibromyalgia groups (P < 0.03). Conclusion. Despite a significant comorbid fibromyalgia prevalence in a cohort of patients with CDOMDs entering an FRP, 41% of patients with an initial fibromyalgia diagnosis no longer met diagnostic criteria for fibromyalgia at discharge and were indistinguishable from LO patients on pain, disability, and depression symptoms. However, both fibromyalgia groups (LFM and RFM) had lower work retention than LO patients 1 year later, suggesting that an FRP may suppress symptoms of fibromyalgia in a subset of patients, but prolonged fibromyalgia-related disability may be more difficult to overcome. Key words: fibromyalgia, central sensitization, functional restoration, tender points, chronic disabling occupational musculoskeletal disorder, work return, outcomes, depression, pain intensity, disability, interdisciplinary treatment. Level of Evidence: 2 Spine 2014;39:1393–1400
F
ibromyalgia affects approximately 2% to 5% of the general population in the United States,1 although the prevalence is much higher in chronic pain populations. Recent studies found comorbid fibromyalgia prevalence rates of 23% to 41% in patients with chronic disabling occupational musculoskeletal disorders (CDOMDs) entering a functional restoration program (FRP).2,3 The 1990 American College of Rheumatology (ACR) criteria for diagnosing fibromyalgia have 2 major components: (a) the presence of chronic widespread pain for at least 3 months; and (b) the presence of at least 11 of 18 tender points with a manual tender point assessment.4 Chronic widespread pain is defined as pain above and below the waist, on the left and right side of the body, with at least 1 point along the axial skeleton. www.spinejournal.com
1393
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINEIP1315_LR 1393
7/3/14 9:28 PM
CLINICAL CASE SERIES Few studies have examined fibromyalgia as a transient diagnosis. Most research assumes that fibromyalgia is stable and chronic.5,6 Therefore, diagnosis changes or tender point responsiveness related to specific intervention is infrequently studied.7–10 The small number of studies that have examined fibromyalgia diagnosis changes have all examined fibromyalgia as the primary diagnosis and found that 30% to 70% of patients lost the diagnosis after discharge.11,12,13 Treatments resulting in loss of the fibromyalgia diagnosis involved relaxation training or exercise in a rheumatology clinic,11 cognitive behavioral therapy,12 and an interdisciplinary treatment at a community pain clinic.13 None of the studies specifically identified fibromyalgia as a comorbid condition in a workers’ compensation (WC) population of patients with CDOMDs or assessed responsiveness to an FRP, which is an intensive interdisciplinary treatment program that includes extensive biopsychosocial treatment, with the goal of returning patients to socioeconomic productivity. FRP effectiveness is well documented for chronic lumbar, lower extremity, upper extremity, and cervical disorders,14–22 with average full-duty work return rates between 85% and 93% and work retention rates 1 year post-treatment between 78% and 85%.23–25 Given these FRP success rates, it was reasonable to hypothesize that an FRP would provide similar improvements in fibromyalgia diagnoses. The major goal of the present study was to evaluate FRP treatment responsiveness in a group of patients with a primary CDOMD diagnosis, who also met diagnostic criteria for comorbid fibromyalgia. Those patients who met diagnostic criteria for fibromyalgia at treatment admission were divided into 2 groups: those who lost and those who retained the fibromyalgia diagnosis at discharge. These 2 groups were compared with a group of patients with chronic regional lumbar pain only (no fibromyalgia or widespread pain) on psychosocial self-report measures and relevant socioeconomic outcomes (including work return and work retention) 1 year after treatment discharge.
MATERIALS AND METHODS Participants From consecutive patients with CDOMD (82% with chronic spinal disorders and all patients reporting chronic spinal pain) with extended pain/disability, referred for functional restoration rehabilitation, 147 patients were diagnosed, upon admission, with comorbid fibromyalgia, under the ACR 1990 criteria.4 Most patients (98%) were referred for treatment of compensable injuries under state or federal WC. Thirty patients with comorbid fibromyalgia (18%) did not complete the FRP, leaving 117 patients with comorbid fibromyalgia available for treatment responsiveness analysis. No significant differences existed between completers and noncompleters in age, sex, tender point count, or injured body parts. At discharge, patients with fibromyalgia were broken into subgroups: those who lost the comorbid fibromyalgia diagnosis (LFM group, n = 48) and those who retained the comorbid fibromyalgia diagnosis (RFM group, n = 69). 1394
www.spinejournal.com
Diagnosis Change in Patients With Comorbid Fibromyalgia • Hartzell et al
Noncompleters were unavailable for fibromyalgia evaluation at discharge; therefore, fibromyalgia diagnosis determination was not possible. In addition, a consecutive sample of patients with chronic disabling occupational regional lumbar disorders only (n = 104), without other involved body parts, chronic widespread pain, or fibromyalgia, were selected as a comparison group. These lumbar only (LO) patients were admitted to the FRP within the same time period as the other patients with CDOMDs with comorbid fibromyalgia. Seventeen LO patients did not complete treatment (16%), leaving 87 LO patients for analysis at discharge. No significant differences in age, sex, or injured body parts were found between LO patients who did and did not complete treatment. As seen in Table 1, no differences were found among the LO and fibromyalgia groups in age, ethnicity, length of disability (time from injury to program entrance), total temporary disability (the number of months not working between injury and program entrance), attorney retention, disability income, or number of presenting injuries. However, patients with fibromyalgia were more likely to be female, which is an established risk factor for fibromyalgia.3,26
Procedure Patients were referred to a regional interdisciplinary FRP and consented to the collection of information for treatment management and clinical research purposes. All patients met the following criteria: (1) 4 or more months had passed since the work-related injury; (2) nonsurgical care failed to improve symptoms sufficiently to allow full return-to-work; (3) surgery did not resolve the injury or was not an option; (4) severe pain and functional limitations remained; and (5) patients communicated in English or Spanish. Although data were collected prospectively as part of the standard medical record, data were analyzed retrospectively, and the study was granted exemption from institutional review board review. Treatment consisted of a medically supervised, quantitatively directed exercise progression program, combined with a multimodal disability management approach, which included cognitivebehavioral therapy, stress management/biofeedback training, education, and vocational integration.27 Treatment duration for program completers was approximately 160 hours. FRP effectiveness has been well documented, with patients significantly increasing strength and function,20,21,28–30 and randomized clinical trials have demonstrated the effectiveness of both this specific FRP program15,31 and several similar international programs.28,32–35 At admission, demographic data were collected, and patients with CDOMD were assessed for chronic widespread pain using a quantified pain drawing.27,36 Patients who were positive for chronic widespread pain were then assessed for comorbid fibromyalgia using a standardized Manual Tender Point Survey.37 All testers were licensed physical or occupational therapists and consistently reproduced 4 kg of thumb pressure (as required) by practicing with a dynamometer. Inter-rater reliability with this methodology was determined to be very good in a previously published study by our group August 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINEIP1315_LR 1394
7/3/14 9:28 PM
CLINICAL CASE SERIES
Diagnosis Change in Patients With Comorbid Fibromyalgia • Hartzell et al
TABLE 1. Pretreatment Demographics Among Patients With Fibromyalgia Who Lost the Diagnosis at
Discharge (LFM), Retained the Diagnosis at Discharge (RFM), and Lumbar Pain Only (LO) Comparison Subjects (Total N = 204) RFM Group (N = 69)
Variable
LFM Group (N = 48)
Area of injury, n (%)
LO Group (N = 87)
F/χ2
Significance
N/A
7.67
0.22
Lumbar only
2 (3)
2 (4)
Cervical only
6 (9)
11 (23)
Extremity only
10 (15)
6 (13)
Multiple spinal
13 (19)
10 (21)
Spinal plus additional musculoskeletal areas
35 (52)
16 (34)
1 (2)
2 (4)
Sex, n (% male)
30 (45)
24 (51)
62 (65)
7.16
0.01
Age, mean (SD)
49.7 (8.8)
47.6 (9.8)
46.56
2.05
0.13
9.16
0.32
0.89
Other
Ethnicity, n (%) Caucasian
26 (39)
21 (48)
55 (60)
African American
23 (35)
11 (25)
15 (17)
Hispanic
16 (24)
11 (25)
20 (22)
1 (2)
1 (2)
1 (1)
Length of disability in months, mean (SD)
25.9 (27.7)
25.5 (33.1)
28.0 (35.3)
0.12
Total temporary disability in months, mean (SD)
15.9 (15.9)
27.3 (35.6)
20.3 (27.6)
0.20
2 (3)
4 (11)
8 (7)
0.34
N/A
0.93
22 (20)
0.20
Asian
Receiving disability income, n (% yes) Compensable injuries, n (%) 1 injury
15 (22)
14 (30)
More than 1 injury
52 (78)
32 (70)
Attorney retention, n (% yes)
13 (31)
10 (32)
Effect Size*
0.18
0.29
*Effect Size (partial η ): small = 0.1, medium = 0.3, large = 0.5. RFM indicates retained fibromyalgia diagnosis; LFM, lost fibromyalgia diagnosis; LO, lumbar pain only; N/A, not applicable. 2
(r = 0.9, P = 0.001).3 Using the 1990 ACR criteria, patients were diagnosed with comorbid fibromyalgia if they reported chronic widespread pain for at least 3 months and reported 11 of 18 tender points when digitally palpitated at 4 kg of pressure.4 This procedure was repeated at discharge. In addition, psychosocial variables assessing pain, function, health-related quality of life, and depressive symptoms were collected at admission and discharge. These included the Pain Disability Questionnaire (PDQ)38,39; the Oswestry Disability Index (ODI)40,41; the Beck Depression Inventory42; the 36-Item Short Form Health Survey (SF-36)43–45; and the pain visual analogue scale.46 Finally, socioeconomically relevant outcomes were assessed 1 year after discharge in a structured face-to-face or telephone interview. Work outcomes included work return, which assessed whether the patient reentered the work force at any point in the year after discharge, and work Spine
retention, which assessed whether the patient was still working at 1-year follow-up. Health utilization outcomes incorporated the number of surgical procedures to the original site(s) of claimed injury, the percentage of patients seeking health care from a new provider, and the number of visits to a new provider. A third outcome domain involved new WC injury claims to the original body part(s), with lost work time.46
Statistical Analyses Analyses of variance were used to compare the LFM, RFM, and LO groups for continuous variables. Independent χ2 tests were used for categorical variables. Admission-to-discharge changes were calculated using repeated measures analyses of variance for all psychosocial variables. Effect sizes were reported for all significant variables: partial η2 was used for continuous variables and Cohen W47 for categorical variables. www.spinejournal.com
1395
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINEIP1315_LR 1395
7/3/14 9:28 PM
CLINICAL CASE SERIES
Diagnosis Change in Patients With Comorbid Fibromyalgia • Hartzell et al
In addition, to correct for the number of multiple comparisons and to control for type I error, all pairwise comparison tests were corrected for using the Holm-Bonferroni step-down procedure48; however, this procedure did not change the overall significance levels of any tests, and thus unadjusted P values for omnibus tests were reported. A power analysis, conducted with G*Power 3.1,49,50 specified a minimum of 35 patients per group to detect a medium effect size (partial η2 = 0.09); these requirements were fulfilled.
RESULTS Of the 117 patients with comorbid fibromyalgia at treatment admission, and who successfully completed the treatment program, 68 (59%) retained the fibromyalgia diagnosis at discharge (i.e., still had >11 tender points and the presence of chronic widespread pain), whereas 48 (41%) no longer met diagnostic criteria (Figure 1). Of those 48 patients who lost the diagnostic criteria, 24 (50%) had less than 11 tender points but still reported chronic widespread pain. Of the remaining 24, 17 (35%) had less than 11 tender points and no chronic widespread pain, and 7 (15%) had no chronic widespread pain but more than 11 tender points. As expected, a repeated measures analysis of variance revealed a greater decrease in mean number of tender points for the LFM group [M (change) = −7.0, SE (change) = 0.49], relative to the RFM group [M (change) = −0.9, SE (change) = 0.4], P < 0.001, partial η2 = 0.45. As shown in Table 2, few psychosocial differences were found among the LFM, RFM, and LO groups at admission. The RFM group had significantly higher pain intensity (P = 0.04) and more depressive symptoms (P = 0.05). All other psychosocial self-report variables were nonsignificant. However, many psychosocial differences between the LFM, RFM, and LO groups were identified at discharge. Compared with both the LO and LFM groups, the RFM group had significantly higher scores on the Beck Depression Inventory (P < 0.001), pain intensity VAS (P < 0.05), and both disability measures (PDQ [P = 0.05] and ODI [P = 0.04]). The RFM group scored significantly lower than the LO group on the SF-36 Mental Health (P = 0.05) and Physical Health (P < 0.01) Summary scales. The LFM group, in contrast to the RFM group, was statistically similar to the LO group on all discharge measures and did not differ significantly in self-reported pain intensity (P = 1.0), depressive symptoms (P = 0.09), the SF-36 Mental Health (P = 0.5) and Physical Health (P = 0.2) Summary Scales, or perceived disability, as measured by the ODI (P = 0.1) and PDQ (P = 1.0). All 3 groups significantly improved from admission to discharge on both measures of perceived disability (PDQ and ODI), although the RFM group consistently and significantly showed the least amount of improvement. In addition, all groups showed significant improvement on the Beck Depression Inventory and pain intensity VAS, with no significant differences in the amount of improvement among the 3 groups. There was, however, significantly more change on the SF-36 Physical Health Summary from admission to discharge in the 1396
www.spinejournal.com
Figure 1. Number and percentage of patients with chronic disabling occupational musculoskeletal disorder with comorbid FM at treatment admission who retained and who lost the FM diagnosis at treatment discharge. FM indicates fibromyalgia; LFM, lost fibromyalgia diagnosis; RFM, retained fibromyalgia diagnosis.
LO group than in the RFM group but no differences in the SF-36 Mental Health Summary component (P = 0.08). One year postdischarge outcomes are presented in Table 3. Outcomes were significantly different among the 3 groups in work retention (P = 0.03, W = 0.02), with the LO group having the highest work retention rate (82%) and the LFM and RFM groups having similarly low work retention rates (59% and 60%, respectively). Work return was not significantly different among the 3 groups (P = 0.07). There were no significant differences in health care utilization, including the number of new surgical procedures or percent seeking health care from new providers. There was also no difference in number of new claimed lost-time injuries within the year after discharge.
DISCUSSION Previous studies have found a high comorbidity between CDOMDs and the linked diagnoses of chronic widespread pain and fibromyalgia.3,36 The purpose of this present study was to evaluate treatment responsiveness of comorbid fibromyalgia diagnoses in a sample of patients with CDOMDs with spinal pain and active WC claims. These patients initially met criteria for fibromyalgia and subsequently completed an FRP. Nearly half (41%) of the patients with comorbid fibromyalgia at treatment admission no longer met the diagnostic August 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINEIP1315_LR 1396
7/3/14 9:28 PM
CLINICAL CASE SERIES
Diagnosis Change in Patients With Comorbid Fibromyalgia • Hartzell et al
TABLE 2. Psychosocial Self-Report Measures Among Patients With Fibromyalgia Who Lost the
Diagnosis at Discharge (LFM), Retained the Diagnosis at Discharge (RFM), and Lumbar Pain Only (LO) Comparison Subjects (Total N = 204) RFM Group (N = 69)
Variable
LFM Group (N = 48)
LO Group (N = 87)
F
Significance
Effect Size*
Pain Intensity, mean (SD) Pre
7.5 (1.9)
7.2 (1.9)
6.8 (1.7)
3.26
0.04†
0.03
Post
5.9 (2.7)
4.7 (2.2)
4.6 (2.3)
5.29
0.01†,‡
0.05
−1.6 (0.3)§
−2.4 (0.4)§
−2.2 (0.3)§
1.38
0.26
Change
Beck Depression Inventory, mean (SD) Pre
23.5 (11.6)
22.4 (12.8)
19.3 (9.9)
3.01
0.05†
0.03
Post
18.6 (13.1)
13.2 (9.4)
9.6 (7.7)
14.02
CLINICAL CASE SERIES
Do Comorbid Fibromyalgia Diagnoses Change After a Functional Restoration Program in Patients With Chronic Disabling Occupational Musculoskeletal Disorders? Meredith M. Hartzell, MS,* Randy Neblett, MA, LPC, BCB,* Yoheli Perez, PT, DPT,* Emily Brede, PhD, RN,* Tom G. Mayer, MD,† and Robert J. Gatchel, PhD, ABPP‡
Study Design. A retrospective study of prospectively collected data. Objective. To determine whether comorbid fibromyalgia, identified in patients with chronic disabling occupational musculoskeletal disorders (CDOMDs), resolves with a functional restoration program (FRP). Summary of Background Data. Fibromyalgia involves widespread bodily pain and tenderness to palpation. In recent studies, 23% to 41% of patients with CDOMDs entering an FRP had comorbid fibromyalgia, compared with population averages of 2% to 5%. Few studies have examined whether fibromyalgia diagnoses resolve with any treatment, and none have investigated diagnosis responsiveness to an FRP. Methods. A consecutive cohort of patients with CDOMDs (82% with spinal disorders and all reporting chronic spinal pain) and comorbid fibromyalgia (N = 117) completed an FRP, which included quantitatively directed exercise progression and multimodal disability management. Diagnosis responsiveness, evaluated at discharge, created 2 groups: those who retained fibromyalgia and those who did not. These groups were compared with chronic regional lumbar pain only patients (LO group, n = 87), lacking widespread pain and fibromyalgia. Results. Of the patients with comorbid fibromyalgia, 59% (n = 69) retained the fibromyalgia diagnosis (RFM group) and 41% (n = 48) From the *PRIDE Research Foundation, Dallas, TX; †Department of Orthopedic Surgery, University of Texas Southwestern Medical Center, Dallas, TX; and ‡Department of Psychology, College of Science, University of Texas at Arlington, Arlington, TX. Acknowledgment date: October 15, 2013. First revision date: February 27, 2014. Second revision date: April 13, 2014. Acceptance date: April 22, 2014. The manuscript submitted does not contain information about medical device(s)/drug(s). No funds were received in support of this work. Relevant financial activities outside the submitted work: consultancy, grants, employment, royalties, stocks. Address correspondence and reprint requests to Tom G. Mayer, MD, PRIDE Research Foundation, 5701 Maple Ave, Ste 100, Dallas, TX 75235; E-mail: [email protected] DOI: 10.1097/BRS.0000000000000406 Spine
lost the fibromyalgia diagnosis (LFM group) at discharge. Although all 3 groups reported decreased pain intensity, disability, and depressive symptoms from admission to discharge, RFM patients reported higher symptom levels than the LFM and LO groups at discharge. The LFM and LO groups were statistically similar. At 1-year follow-up, LO patients demonstrated higher work retention than both fibromyalgia groups (P < 0.03). Conclusion. Despite a significant comorbid fibromyalgia prevalence in a cohort of patients with CDOMDs entering an FRP, 41% of patients with an initial fibromyalgia diagnosis no longer met diagnostic criteria for fibromyalgia at discharge and were indistinguishable from LO patients on pain, disability, and depression symptoms. However, both fibromyalgia groups (LFM and RFM) had lower work retention than LO patients 1 year later, suggesting that an FRP may suppress symptoms of fibromyalgia in a subset of patients, but prolonged fibromyalgia-related disability may be more difficult to overcome. Key words: fibromyalgia, central sensitization, functional restoration, tender points, chronic disabling occupational musculoskeletal disorder, work return, outcomes, depression, pain intensity, disability, interdisciplinary treatment. Level of Evidence: 2 Spine 2014;39:1393–1400
F
ibromyalgia affects approximately 2% to 5% of the general population in the United States,1 although the prevalence is much higher in chronic pain populations. Recent studies found comorbid fibromyalgia prevalence rates of 23% to 41% in patients with chronic disabling occupational musculoskeletal disorders (CDOMDs) entering a functional restoration program (FRP).2,3 The 1990 American College of Rheumatology (ACR) criteria for diagnosing fibromyalgia have 2 major components: (a) the presence of chronic widespread pain for at least 3 months; and (b) the presence of at least 11 of 18 tender points with a manual tender point assessment.4 Chronic widespread pain is defined as pain above and below the waist, on the left and right side of the body, with at least 1 point along the axial skeleton. www.spinejournal.com
1393
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINEIP1315_LR 1393
7/3/14 9:28 PM
CLINICAL CASE SERIES Few studies have examined fibromyalgia as a transient diagnosis. Most research assumes that fibromyalgia is stable and chronic.5,6 Therefore, diagnosis changes or tender point responsiveness related to specific intervention is infrequently studied.7–10 The small number of studies that have examined fibromyalgia diagnosis changes have all examined fibromyalgia as the primary diagnosis and found that 30% to 70% of patients lost the diagnosis after discharge.11,12,13 Treatments resulting in loss of the fibromyalgia diagnosis involved relaxation training or exercise in a rheumatology clinic,11 cognitive behavioral therapy,12 and an interdisciplinary treatment at a community pain clinic.13 None of the studies specifically identified fibromyalgia as a comorbid condition in a workers’ compensation (WC) population of patients with CDOMDs or assessed responsiveness to an FRP, which is an intensive interdisciplinary treatment program that includes extensive biopsychosocial treatment, with the goal of returning patients to socioeconomic productivity. FRP effectiveness is well documented for chronic lumbar, lower extremity, upper extremity, and cervical disorders,14–22 with average full-duty work return rates between 85% and 93% and work retention rates 1 year post-treatment between 78% and 85%.23–25 Given these FRP success rates, it was reasonable to hypothesize that an FRP would provide similar improvements in fibromyalgia diagnoses. The major goal of the present study was to evaluate FRP treatment responsiveness in a group of patients with a primary CDOMD diagnosis, who also met diagnostic criteria for comorbid fibromyalgia. Those patients who met diagnostic criteria for fibromyalgia at treatment admission were divided into 2 groups: those who lost and those who retained the fibromyalgia diagnosis at discharge. These 2 groups were compared with a group of patients with chronic regional lumbar pain only (no fibromyalgia or widespread pain) on psychosocial self-report measures and relevant socioeconomic outcomes (including work return and work retention) 1 year after treatment discharge.
MATERIALS AND METHODS Participants From consecutive patients with CDOMD (82% with chronic spinal disorders and all patients reporting chronic spinal pain) with extended pain/disability, referred for functional restoration rehabilitation, 147 patients were diagnosed, upon admission, with comorbid fibromyalgia, under the ACR 1990 criteria.4 Most patients (98%) were referred for treatment of compensable injuries under state or federal WC. Thirty patients with comorbid fibromyalgia (18%) did not complete the FRP, leaving 117 patients with comorbid fibromyalgia available for treatment responsiveness analysis. No significant differences existed between completers and noncompleters in age, sex, tender point count, or injured body parts. At discharge, patients with fibromyalgia were broken into subgroups: those who lost the comorbid fibromyalgia diagnosis (LFM group, n = 48) and those who retained the comorbid fibromyalgia diagnosis (RFM group, n = 69). 1394
www.spinejournal.com
Diagnosis Change in Patients With Comorbid Fibromyalgia • Hartzell et al
Noncompleters were unavailable for fibromyalgia evaluation at discharge; therefore, fibromyalgia diagnosis determination was not possible. In addition, a consecutive sample of patients with chronic disabling occupational regional lumbar disorders only (n = 104), without other involved body parts, chronic widespread pain, or fibromyalgia, were selected as a comparison group. These lumbar only (LO) patients were admitted to the FRP within the same time period as the other patients with CDOMDs with comorbid fibromyalgia. Seventeen LO patients did not complete treatment (16%), leaving 87 LO patients for analysis at discharge. No significant differences in age, sex, or injured body parts were found between LO patients who did and did not complete treatment. As seen in Table 1, no differences were found among the LO and fibromyalgia groups in age, ethnicity, length of disability (time from injury to program entrance), total temporary disability (the number of months not working between injury and program entrance), attorney retention, disability income, or number of presenting injuries. However, patients with fibromyalgia were more likely to be female, which is an established risk factor for fibromyalgia.3,26
Procedure Patients were referred to a regional interdisciplinary FRP and consented to the collection of information for treatment management and clinical research purposes. All patients met the following criteria: (1) 4 or more months had passed since the work-related injury; (2) nonsurgical care failed to improve symptoms sufficiently to allow full return-to-work; (3) surgery did not resolve the injury or was not an option; (4) severe pain and functional limitations remained; and (5) patients communicated in English or Spanish. Although data were collected prospectively as part of the standard medical record, data were analyzed retrospectively, and the study was granted exemption from institutional review board review. Treatment consisted of a medically supervised, quantitatively directed exercise progression program, combined with a multimodal disability management approach, which included cognitivebehavioral therapy, stress management/biofeedback training, education, and vocational integration.27 Treatment duration for program completers was approximately 160 hours. FRP effectiveness has been well documented, with patients significantly increasing strength and function,20,21,28–30 and randomized clinical trials have demonstrated the effectiveness of both this specific FRP program15,31 and several similar international programs.28,32–35 At admission, demographic data were collected, and patients with CDOMD were assessed for chronic widespread pain using a quantified pain drawing.27,36 Patients who were positive for chronic widespread pain were then assessed for comorbid fibromyalgia using a standardized Manual Tender Point Survey.37 All testers were licensed physical or occupational therapists and consistently reproduced 4 kg of thumb pressure (as required) by practicing with a dynamometer. Inter-rater reliability with this methodology was determined to be very good in a previously published study by our group August 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINEIP1315_LR 1394
7/3/14 9:28 PM
CLINICAL CASE SERIES
Diagnosis Change in Patients With Comorbid Fibromyalgia • Hartzell et al
TABLE 1. Pretreatment Demographics Among Patients With Fibromyalgia Who Lost the Diagnosis at
Discharge (LFM), Retained the Diagnosis at Discharge (RFM), and Lumbar Pain Only (LO) Comparison Subjects (Total N = 204) RFM Group (N = 69)
Variable
LFM Group (N = 48)
Area of injury, n (%)
LO Group (N = 87)
F/χ2
Significance
N/A
7.67
0.22
Lumbar only
2 (3)
2 (4)
Cervical only
6 (9)
11 (23)
Extremity only
10 (15)
6 (13)
Multiple spinal
13 (19)
10 (21)
Spinal plus additional musculoskeletal areas
35 (52)
16 (34)
1 (2)
2 (4)
Sex, n (% male)
30 (45)
24 (51)
62 (65)
7.16
0.01
Age, mean (SD)
49.7 (8.8)
47.6 (9.8)
46.56
2.05
0.13
9.16
0.32
0.89
Other
Ethnicity, n (%) Caucasian
26 (39)
21 (48)
55 (60)
African American
23 (35)
11 (25)
15 (17)
Hispanic
16 (24)
11 (25)
20 (22)
1 (2)
1 (2)
1 (1)
Length of disability in months, mean (SD)
25.9 (27.7)
25.5 (33.1)
28.0 (35.3)
0.12
Total temporary disability in months, mean (SD)
15.9 (15.9)
27.3 (35.6)
20.3 (27.6)
0.20
2 (3)
4 (11)
8 (7)
0.34
N/A
0.93
22 (20)
0.20
Asian
Receiving disability income, n (% yes) Compensable injuries, n (%) 1 injury
15 (22)
14 (30)
More than 1 injury
52 (78)
32 (70)
Attorney retention, n (% yes)
13 (31)
10 (32)
Effect Size*
0.18
0.29
*Effect Size (partial η ): small = 0.1, medium = 0.3, large = 0.5. RFM indicates retained fibromyalgia diagnosis; LFM, lost fibromyalgia diagnosis; LO, lumbar pain only; N/A, not applicable. 2
(r = 0.9, P = 0.001).3 Using the 1990 ACR criteria, patients were diagnosed with comorbid fibromyalgia if they reported chronic widespread pain for at least 3 months and reported 11 of 18 tender points when digitally palpitated at 4 kg of pressure.4 This procedure was repeated at discharge. In addition, psychosocial variables assessing pain, function, health-related quality of life, and depressive symptoms were collected at admission and discharge. These included the Pain Disability Questionnaire (PDQ)38,39; the Oswestry Disability Index (ODI)40,41; the Beck Depression Inventory42; the 36-Item Short Form Health Survey (SF-36)43–45; and the pain visual analogue scale.46 Finally, socioeconomically relevant outcomes were assessed 1 year after discharge in a structured face-to-face or telephone interview. Work outcomes included work return, which assessed whether the patient reentered the work force at any point in the year after discharge, and work Spine
retention, which assessed whether the patient was still working at 1-year follow-up. Health utilization outcomes incorporated the number of surgical procedures to the original site(s) of claimed injury, the percentage of patients seeking health care from a new provider, and the number of visits to a new provider. A third outcome domain involved new WC injury claims to the original body part(s), with lost work time.46
Statistical Analyses Analyses of variance were used to compare the LFM, RFM, and LO groups for continuous variables. Independent χ2 tests were used for categorical variables. Admission-to-discharge changes were calculated using repeated measures analyses of variance for all psychosocial variables. Effect sizes were reported for all significant variables: partial η2 was used for continuous variables and Cohen W47 for categorical variables. www.spinejournal.com
1395
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINEIP1315_LR 1395
7/3/14 9:28 PM
CLINICAL CASE SERIES
Diagnosis Change in Patients With Comorbid Fibromyalgia • Hartzell et al
In addition, to correct for the number of multiple comparisons and to control for type I error, all pairwise comparison tests were corrected for using the Holm-Bonferroni step-down procedure48; however, this procedure did not change the overall significance levels of any tests, and thus unadjusted P values for omnibus tests were reported. A power analysis, conducted with G*Power 3.1,49,50 specified a minimum of 35 patients per group to detect a medium effect size (partial η2 = 0.09); these requirements were fulfilled.
RESULTS Of the 117 patients with comorbid fibromyalgia at treatment admission, and who successfully completed the treatment program, 68 (59%) retained the fibromyalgia diagnosis at discharge (i.e., still had >11 tender points and the presence of chronic widespread pain), whereas 48 (41%) no longer met diagnostic criteria (Figure 1). Of those 48 patients who lost the diagnostic criteria, 24 (50%) had less than 11 tender points but still reported chronic widespread pain. Of the remaining 24, 17 (35%) had less than 11 tender points and no chronic widespread pain, and 7 (15%) had no chronic widespread pain but more than 11 tender points. As expected, a repeated measures analysis of variance revealed a greater decrease in mean number of tender points for the LFM group [M (change) = −7.0, SE (change) = 0.49], relative to the RFM group [M (change) = −0.9, SE (change) = 0.4], P < 0.001, partial η2 = 0.45. As shown in Table 2, few psychosocial differences were found among the LFM, RFM, and LO groups at admission. The RFM group had significantly higher pain intensity (P = 0.04) and more depressive symptoms (P = 0.05). All other psychosocial self-report variables were nonsignificant. However, many psychosocial differences between the LFM, RFM, and LO groups were identified at discharge. Compared with both the LO and LFM groups, the RFM group had significantly higher scores on the Beck Depression Inventory (P < 0.001), pain intensity VAS (P < 0.05), and both disability measures (PDQ [P = 0.05] and ODI [P = 0.04]). The RFM group scored significantly lower than the LO group on the SF-36 Mental Health (P = 0.05) and Physical Health (P < 0.01) Summary scales. The LFM group, in contrast to the RFM group, was statistically similar to the LO group on all discharge measures and did not differ significantly in self-reported pain intensity (P = 1.0), depressive symptoms (P = 0.09), the SF-36 Mental Health (P = 0.5) and Physical Health (P = 0.2) Summary Scales, or perceived disability, as measured by the ODI (P = 0.1) and PDQ (P = 1.0). All 3 groups significantly improved from admission to discharge on both measures of perceived disability (PDQ and ODI), although the RFM group consistently and significantly showed the least amount of improvement. In addition, all groups showed significant improvement on the Beck Depression Inventory and pain intensity VAS, with no significant differences in the amount of improvement among the 3 groups. There was, however, significantly more change on the SF-36 Physical Health Summary from admission to discharge in the 1396
www.spinejournal.com
Figure 1. Number and percentage of patients with chronic disabling occupational musculoskeletal disorder with comorbid FM at treatment admission who retained and who lost the FM diagnosis at treatment discharge. FM indicates fibromyalgia; LFM, lost fibromyalgia diagnosis; RFM, retained fibromyalgia diagnosis.
LO group than in the RFM group but no differences in the SF-36 Mental Health Summary component (P = 0.08). One year postdischarge outcomes are presented in Table 3. Outcomes were significantly different among the 3 groups in work retention (P = 0.03, W = 0.02), with the LO group having the highest work retention rate (82%) and the LFM and RFM groups having similarly low work retention rates (59% and 60%, respectively). Work return was not significantly different among the 3 groups (P = 0.07). There were no significant differences in health care utilization, including the number of new surgical procedures or percent seeking health care from new providers. There was also no difference in number of new claimed lost-time injuries within the year after discharge.
DISCUSSION Previous studies have found a high comorbidity between CDOMDs and the linked diagnoses of chronic widespread pain and fibromyalgia.3,36 The purpose of this present study was to evaluate treatment responsiveness of comorbid fibromyalgia diagnoses in a sample of patients with CDOMDs with spinal pain and active WC claims. These patients initially met criteria for fibromyalgia and subsequently completed an FRP. Nearly half (41%) of the patients with comorbid fibromyalgia at treatment admission no longer met the diagnostic August 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINEIP1315_LR 1396
7/3/14 9:28 PM
CLINICAL CASE SERIES
Diagnosis Change in Patients With Comorbid Fibromyalgia • Hartzell et al
TABLE 2. Psychosocial Self-Report Measures Among Patients With Fibromyalgia Who Lost the
Diagnosis at Discharge (LFM), Retained the Diagnosis at Discharge (RFM), and Lumbar Pain Only (LO) Comparison Subjects (Total N = 204) RFM Group (N = 69)
Variable
LFM Group (N = 48)
LO Group (N = 87)
F
Significance
Effect Size*
Pain Intensity, mean (SD) Pre
7.5 (1.9)
7.2 (1.9)
6.8 (1.7)
3.26
0.04†
0.03
Post
5.9 (2.7)
4.7 (2.2)
4.6 (2.3)
5.29
0.01†,‡
0.05
−1.6 (0.3)§
−2.4 (0.4)§
−2.2 (0.3)§
1.38
0.26
Change
Beck Depression Inventory, mean (SD) Pre
23.5 (11.6)
22.4 (12.8)
19.3 (9.9)
3.01
0.05†
0.03
Post
18.6 (13.1)
13.2 (9.4)
9.6 (7.7)
14.02
