- Email: [email protected]
Prognostic factors for return to work in patients with sciatica
The Spine Journal 13 (2013) 1849–1857
Clinical Study
Prognostic factors for return to work in patients with sciatica Lars Grøvle, MD, PhDa,*, Anne J. Haugen, MD, PhDa, Anne Keller, MD, PhDb, B ard Ntvig, MD, PhDc,d, Jens I. Brox, MD, PhDe, Margreth Grotle, PT, PhDc,f a Department of Rheumatology, Østfold Hospital Trust, Postboks 16, 1603 Fredrikstad, Norway Department of Physical Medicine and Rehabilitation, Oslo University Hospital, Ullevaal, Postboks 4950 Nydalen, 0424 Oslo, Norway c National Resource Centre for Rehabilitation in Rheumatology, Diakonhjemmet Hospital, Postboks 23 Vinderen, 0319 Oslo, Norway d Department of General Practice, Institute of Health and Society, University of Oslo, Postboks 1130 Blindern, 0318 Oslo, Norway e Orthopaedic Department, Section for Back Surgery, Oslo University Hospital, Rikshospitalet, Postboks 4950 Nydalen, 0424 Oslo, Norway f FORMI (Communication Unit for Musculoskeletal Disorders), Division of Neuroscience, Oslo University Hospital, Ullevaal, Postboks 4950 Nydalen, 0424 Oslo, Norway b
Received 25 August 2011; revised 26 June 2013; accepted 10 July 2013
Abstract
BACKGROUND CONTEXT: Little is known about the prognostic factors for work-related outcomes of sciatica caused by disc herniation. PURPOSE: To identify the prognostic factors for return to work (RTW) during a 2-year follow-up among sciatica patients referred to secondary care. STUDY DESIGN/SETTING: Multicenter prospective cohort study including 466 patients. Administrative data from the National Sickness Benefit Register were accessed for 227 patients. PATIENT SAMPLE: Two samples were used. Sample A comprised patients who at the time of inclusion in the cohort reported being on partial sick leave or complete sick leave or were undergoing rehabilitation because of back pain/sciatica. Sample B comprised patients who, according to the sickness benefit register, at the time of inclusion received sickness benefits or rehabilitation allowances because of back pain/sciatica. OUTCOME MEASURES: In Sample A, the outcome was self-reported return to full-time work at the 2-year follow-up. In Sample B, the outcome was time to first sustained RTW, defined as the first period of more than 60 days without receiving benefits from the register. METHODS: Significant baseline predictors of self-reported RTW at 2 years (Analysis A) were identified by multivariate logistic regression. Significant predictors of time to sustained RTW (Analysis B) were identified by multivariate Cox proportional hazard modeling. Both analyses included adjustment for age and sex. To assess the effect of surgery on the probability of RTW, analyses similar to A and B were performed, including the variable surgery (yes/no). RESULTS: One-fourth of the patients were still out of work at the 2-year follow-up. In Sample A (n5237), younger age, better general health, lower baseline sciatica bothersomeness, less fearavoidance work, and a negative straight-leg–raising test result were significantly associated with a higher probability of RTW at the 2-year follow-up. Surgery was not significantly associated with the outcome. In Sample B (n5125), history of sciatica, duration of the current sciatica episode more than 3 months, greater sciatica bothersomeness, fear-avoidance work, and back pain were significantly associated with a longer time to sustained RTW. Surgery was significantly negatively associated with time to sustained RTW both in univariate (hazard ratio [HR] 0.60; 95% confidence interval [CI] 0.39, 0.93; p5.02) and in multivariate (HR 0.49; 95% CI 0.31, 0.79; p5.003) analyses. CONCLUSIONS: The baseline factors associated with RTW identified in multivariate analysis were age, general health, history of sciatica, duration of the current episode, baseline sciatica bothersomeness, fear-avoidance work, back pain, and the straight-leg–raising test result.
FDA device/drug status: Not applicable. Author disclosures: LG: Nothing to disclose. AJH: Nothing to disclose. AK: Nothing to disclose. BN: Nothing to disclose. JIB: Nothing to disclose. MG: Nothing to disclose. 1529-9430/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.spinee.2013.07.433
* Corresponding author. Department of Rheumatology, Østfold Hospital Trust, Postboks 16, 1603 Fredrikstad, Norway. Tel.: (47) 91814029. E-mail address: [email protected] (L. Grøvle)
1850
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
Surgical treatment was associated with slower RTW, but surgical patients were more severely affected than patients treated without surgery; so, this finding should be interpreted with caution. Ó 2013 Elsevier Inc. All rights reserved. Keywords:
Sciatica; Prognosis; Return to work; Sickness absence; Fear-avoidance beliefs
Introduction More than 10% of all annual sickness absence days are taken by people with a diagnosis of low back pain [1,2]. Among those with low back pain, patients with sciatica generally report more severe pain, have longer absence, and lower rates of return to work (RTW) than patients with nonspecific low back pain [2–7]. In about 90% of cases, sciatica, defined as radiating leg pain below the knee, is caused by a herniated disc with nerve root compression. The compression and secondary inflammation of the nearby nerve roots and dorsal root ganglia [8,9] may cause radiating pain, muscular weakness, sensory disturbances, and depressed tendon reflexes. Some patients also experience bladder, bowel, and genital dysfunction. The condition can vary from short-lasting single episodes to a remitting or permanent course over months or years. The point prevalence of sciatica has been estimated as 2% to 5%, and this condition affects mainly the working-age population [10,11]. In a review, Koes et al. [12] recommended referral to secondary care for patients with intractable radicular pain or whose pain does not diminish after 6 to 8 weeks of conservative care. Surgical discectomy is considered to improve the short-term but not the long-term prognosis of pain and perceived disability [13,14]. Although rates vary considerably [15], surgery for sciatica is a common procedure. In the United States, about one discectomy is performed per 1,000 persons each year [16]. A randomized study did not find an effect of surgery on work status [17]. Despite the frequency and severity of this condition, few studies have investigated the prognostic factors for workrelated outcomes in sciatica. The aim of this study was to identify the prognostic factors for RTW during a 2-year follow-up among sciatica patients referred to secondary care.
Methods Setting This study was part of a prospective cohort study of patients with sciatica and disc herniation referred to back clinics at four hospitals in Southeastern Norway. Patients were invited to participate by the clinic staff. Study participation did not involve any specific type of intervention. The consultation included information about the condition and general advise to stay active and use pain medication if necessary. In patients with severe symptoms, surgery
was performed at the discretion of the individual surgeon. The study protocol was approved by the Regional Committee for Medical Research Ethics and the Ombudsman for Privacy in Research at the Norwegian Social Science Data Services. Patients Patients included in the study were 18 years of age or older, had radiating pain below the knee and/or paresis, and had a lumbar disc herniation verified by a magnetic resonance imaging or computed tomography scan of the corresponding level and side. The exclusion criteria were pregnancy, spinal fracture, tumor, infection, previous surgery in the affected disc, and inability to communicate in written Norwegian. After a protocol amendment, patients included after October 2005 were asked to give a consent for us to obtain sickness absence data from the National Sickness Benefit Register, which covers the entire Norwegian population. Procedures On the day of inclusion, a clinical examination was performed by a physician or physiotherapist, and the patient completed a comprehensive questionnaire. Self-reported data A follow-up assessment was conducted at 2 years using a mailed questionnaire that was completed at home and returned in a prepaid envelope. Patients who had not responded 2 weeks after the scheduled date were contacted by telephone or text message. A reminder letter was sent to nonresponders if no reply was obtained after 3 weeks. Self-reported current work status at baseline and at the 2year follow-up included the categories full-time work, partial sick leave, complete sick leave, rehabilitation, disability pension, student, job seeker, old-age retirement, or homemaker. Administrative data Data obtained from the sickness benefit register included the starting and stopping dates of sickness absence, rehabilitation, and disability benefits. To qualify for sickness benefit in Norway, the occupational disability must be documented by a doctor’s sick leave certificate, which is submitted to the register. If the person is still unable to work after 1 year, he or she may be entitled to a rehabilitation allowance or disability benefits. In the present study, the
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
1851
sickness absence or rehabilitation allowance certificate diagnoses indicating back pain or sciatica were classified according to the International Classification of Primary Care [18]: L02 (back symptom/complaint), L04 (low back symptom/complaint), L84 (back syndrome without radiating pain), and L86 (back syndrome with radiating pain).
!60 ), sensation (deemed abnormal if reduced light touch), reflexes (deemed abnormal if depressed patellar or Achilles reflexes), and muscle strength (deemed abnormal if reduced performance of the single limb stance, tiptoe or heel walking, supine knee or ankle flexion/extension, or big toe extension).
Potential prognostic factors
Outcomes
Information about demographic variables, education, and back pain/sciatica history were obtained at baseline. Leg and back pain were rated on a 0 to 100 visual analog scale, and a composite of perceived symptoms (leg pain, numbness, weakness, and back or leg pain while sitting) was measured using the Sciatica Bothersomeness Index [19,20]. Disability was assessed by the Maine-Seattle Back Questionnaire [20,21], a sciatica-specific shortened version of the Roland-Morris Questionnaire. General health was assessed by the Short-Form 36 general health subscale [22]. Selfreported comorbidity was measured by 27 items of the Subjective Health Complaint Inventory [23], a list of common somatic or psychological complaints. The scoring procedure used in this study has been described elsewhere [24]. Emotional distress was measured by the Hopkins Symptom Check List 25 [25]. The work subscale of the FearAvoidance Back Questionnaire [26,27] was used to assess fear-avoidance beliefs regarding work (here, called fearavoidance work). Pain-related fear of movement/reinjury was measured by a 13-item version of the Tampa Scale for Kinesiophobia [28,29]. The clinical tests were dichotomized and included straight-leg raising (deemed abnormal if
RTW at 2 years by self-report Patients who at the time of inclusion reported being on partial sick leave or complete sick leave or undergoing rehabilitation because of back pain/sciatica were included in this data set. The outcome was return to full-time work at the 2-year follow-up. Patients who reported being employed in full-time work were given the code 1, whereas those who reported being on partial sick leave, complete sick leave, rehabilitation, or disability pension were given the code 0. Patients who at the 2-year follow-up reported being a student, old-age pensioner, or homemaker were excluded from the analyses. Time to sustained RTW assessed from the administrative data Patients who at the time of inclusion according to the sickness benefit register received sickness benefits or rehabilitation allowances because of back pain/sciatica were included in this data set. The outcome was time to first sustained RTW, which was defined by being off the register’s sick list as a proxy for RTW. Sustained RTW was chosen to avoid misclassifications because of sickness
Fig. 1. Flow chart: from the whole cohort to analyzed samples.
1852
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
surgery set as the date of inclusion in the study [32]. Multicollinearity was assessed by the variance inflation factor using a value less than 5 as the acceptance criterion [33]. Variables in the final models were examined for significant interaction effects.
absence recurrences [30,31] and was defined as the number of calendar days from inclusion to the first period of more than 60 days without receiving benefits from the register. Statistical analyses Group differences were analyzed using Mann-Whitney U test, Student t test, or chi-square test according to the type of the data. Significant predictors of self-reported RTW at 2 years (Analysis A) were identified by logistic regression. Significant predictors of time to sustained RTW (Analysis B) were identified by Cox proportional hazard modeling. Patients who had not returned to sustained work at 2 years were classified as censored. In both analyses (A and B), multivariate models were built by including potential prognostic factors with p values less than .2 in the univariate analyses. Variables with the highest p value were removed one by one until all remaining variables had p values less than .05. Both analyses included adjustment for age and sex. To assess the effect of surgery on the probability of RTW, analyses similar to A and B were performed, including the variable surgery (yes/no). Cox regression analysis was then performed with the date of
Results During 2005 and 2006, a total of 466 patients, with a mean age of 43.6 (standard deviation [SD] 11.5) years, were enrolled in the prospective cohort study. All patients included after October 2005 (n5227) gave consent to obtain sickness absence data from the National Sickness Benefit Register. A flow chart from the whole cohort to analyzed samples is shown in Figure 1. RTW at 2 years by self-report At the time of inclusion, 297 patients reported being on complete sick leave (n5227), partial sick leave (n554), or undergoing rehabilitation (n516). The 2-year follow-up assessments were completed by 246 patients (82.3%). Those who at 2 years reported being a student (n55), in old-age
Table 1 Patients’ baseline demographic and clinical characteristics according to the type of analysis performed A
B
Patient characteristics
RTW at 2 y, n5237
Time to sustained RTW, n5125
Men Mean (SD) age, y Mean (SD) education, y Married/cohabitant Current smoker Duration back problems !1 y 1–5 y O5 y Number of previous sciatica episodes 0 1–4 $5 Mean (SD) duration of current sciatica episode, wk Mean (SD) back pain (0–100) Mean (SD) leg pain (0–100) Mean (SD) sciatica bothersomeness (0–24) Mean (SD) disability* (0–12) Mean (SD) fear-avoidance beliefs, work (0–42) Mean (SD) fear of movement/reinjury (13–52) Mean (SD) emotional distress (0–4) Mean (SD) subjective health complaints (0–27) Mean (SD) general health (0–100)y Clinical findings Positive straight-leg–raising test Motor weakness Reflexes depressed Sensory decrease
136 43.2 13.2 181 99
68 42.1 13.1 86 55
RTW, return to work; SD, standard deviation. Values are n (%) unless otherwise stated. * Measured by the Maine-Seattle Back Questionnaire. y Higher values indicate better health.
(57.4) (9.2) (3.0) (76.7) (41.9)
73 (30.8) 53 (22.4) 111 (46.8)
(54.4) (10.4) (3.1) (69.9) (44.4)
44 (35.2) 30 (24.0) 51 (40.8)
106 98 32 21.6 43.8 64.6 14.9 8.5 20.1 27.6 1.6 7.3 71.1
(44.9) (41.5) (13.6) (24.1) (30.0) (26.9) (4.8) (2.3) (10.0) (6.8) (0.4) (4.1) (19.4)
61 49 15 19.0 49.0 70.6 15.5 8.8 21.2 28.1 1.6 7.8 68.7
(48.8) (39.2) (12.0) (14.5) (30.3) (24.3) (4.6) (2.3) (9.9) (6.7) (0.4) (4.1) (19.5)
140 108 108 138
(60.6) (47.0) (46.4) (58.7)
88 56 59 76
(71.5) (45.5) (48.4) (61.3)
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
1853
Table 2 Associations between potential prognostic variables and RTW according to the type of analysis A
B
RTW at 2 y by self-report (n5237)
Time to sustained RTW by administrative data (n5125)
Univariate Prognostic factors Sociodemographic characteristics Age (y) Women Married/cohabitant Education (y) Current smoker Sciatica history Duration current sciatica episode O3 mo Had sciatica before Duration of back problems O1 y Self-reported health status Subjective health complaints (0–27) Sciatica bothersomeness (0–24) Disability (0–12)y Fear-avoidance beliefs, work (0–42) Fear of movement/reinjury (13–52) General health (0–100)z Emotional distress (0–4) Back pain (0–100) Leg pain (0–100) Clinical findings Positive straight-leg–raising test Motor weakness Reflexes depressed Sensory decrease
Multivariate*
Univariate
Multivariate*
OR
95% CI
p
OR
95% CI
p
HR
95% CI
p
HR
95% CI
p
0.98 0.51 0.89 1.11 0.64
0.95–1.01 0.29–0.91 0.45–1.78 1.00–1.23 0.36–1.14
.24 .02 .75 .04 .13
0.97 0.61
0.93–1.00 0.31–1.22
.07 .17
1.00 0.66 0.81 1.03 0.77
0.98–1.02 0.45–0.98 0.53–1.24 0.97–1.10 0.52–1.14
.86 .04 .34 .29 .19
1.00 0.81
0.98–1.02 0.54–1.23
.78 .33
0.56 0.60 0.61
0.31–1.03 0.33–1.09 0.31–1.17
.06 .09 .14
0.63 0.64 0.88
0.43–0.93 0.43–0.94 0.59–1.30
.02 .02 .52
0.63 0.66
0.42–0.93 0.44–1.00
.02 .05
0.85 0.86 0.82 0.93 0.94 1.04 0.25 0.99 0.99
0.79–0.92 0.81–0.93 0.72–0.94 0.90–0.96 0.90–0.98 1.02–1.05 0.12–0.51 0.98–1.00 0.97–1.00
.00 .00 .00 .00 .00 .00 .00 .08 .01
0.97 0.93 0.93 0.98 0.98 1.01 0.47 0.99 1.00
0.93–1.01 0.89–0.97 0.86–1.01 0.96–0.99 0.96–1.01 1.00–1.02 0.29–0.77 0.99–1.00 0.99–1.00
.17 .00 .10 .01 .26 .01 .00 .01 .17
0.94
0.90–0.99
.02
0.97
0.95–0.99
.00
0.99
0.99–1.00
.04
0.48 0.56 1.07 1.19
0.26–0.92 0.31–1.02 0.60–1.91 0.67–2.14
.03 .06 .83 .55
0.78 0.81 0.96 1.07
0.51–1.19 0.55–1.20 0.66–1.42 0.73–1.59
.25 .30 .85 .72
0.89
0.82–0.97
.01
0.93
0.90–0.97
.00
1.03
1.01–1.05
.00
0.44
0.20–0.95
.04
RTW, return to work; OR, odds ratios; HR, hazard ratio; CI, confidence interval. * Final result of a stepwise backward selection model. y Measured by the Maine-Seattle Back Questionnaire. z Higher values indicate better health.
retirement (n52), or a homemaker (n52) were excluded, rendering 237 patients eligible for analysis. The baseline characteristics of these patients are shown in Table 1 (A). During the 2-year follow-up period, 88 patients (29.6%) received surgical therapy at a mean (SD) of 13.3 (18.4) weeks after inclusion in the study. Compared with those who did not have surgery, the surgically treated patients reported at baseline more disability (9.4 [2.0] vs. 8.2 [2.4], p!.01), more sciatica bothersomeness (16.6 [4.5] vs. 13.9 [4.7], p!.01), more leg pain (76.7 [22.2] vs. 60.0 [26.6], p!.01), more back pain (52.3 [31.0] vs. 41.6 [29.2], p!.01), and a longer duration of the current sciatica episode (23.2 [19.7] vs. 20.1 [23.0] weeks, p!.05). The proportion of patients with a positive straight-leg–raising test was also higher in the surgically treated group (75.6% vs. 56.6%, p!.01). In total, 54% and 37% of the patients had visited a physiotherapist or a chiropractor during the first and the second year of follow-up, respectively. No patients reported to have received epidural injections. The current work status at 2 years was full-time work n5172 (72.6%), partial sick leave n57 (3.0%), complete sick leave n515 (6.3%), rehabilitation n532 (13.5%), disability pension n510 (4.2%), and job seeker n51
(0.4%). The patients who were included in Sample A and who did not respond at the 2-year follow-up (n551) were younger (39.4 vs. 43.2 years, p!.05), reported higher baseline fear-avoidance work (23.8 vs. 20.1, p!.05), and were less likely to have a positive straight-leg–raising test (51.1% vs. 62.4%, p!.05) than those who responded. None of the other baseline variables differed significantly between responders and nonresponders. According to the sickness benefit register, the responders and nonresponders in Sample A did not differ significantly in terms of days to return to sustained work or the proportion of patients who were off the sick list for more than 60 days at 2 years. In the univariate analyses (Table 2 [A]), the following variables were associated with better outcome: being a man, having a higher educational level, better general health, and fewer subjective health complaints, reporting less sciatica bothersomeness and disability, and having less fear-avoidance work and emotional distress, and a negative straight-leg–raising test result. In the final multivariate analysis the following variables remained as independent and significant factors associated with RTW at the 2-year follow-up: age, general health, sciatica bothersomeness,
1854
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
fear-avoidance work, and the straight-leg–raising test result. Surgery was not significantly associated with the outcome in either the univariate or the multivariate analysis. The explained variance of the final model as assessed by the Nagelkerke R2 was 31%. Time to sustained RTW obtained from administrative data According to the sickness benefit register, 125 of the 227 patients included in the prospective cohort after October 2005 received either sickness benefits (n5121) or rehabilitation benefits (n54). The baseline characteristics of this sample are presented in Table 1 (B). At 2 years, 98 (78.4%) patients had been off the sick list for more than 60 days. Figure 2 shows the proportion of patients returning to sustained work as a function of time. The median time to sustained RTW was 169 days (95% confidence interval [CI] 123, 215). Forty-one (32.8%) patients had received surgical therapy at a mean of 11.2 (SD 14.7) weeks after the date of inclusion. Self-reported work status at the 2-year follow-up was nearly identical in samples A and B (data not shown). Results of the univariate and multivariate Cox regression analyses are shown in Table 2 (B). In the univariate analysis, the following variables were associated with longer time to sustained RTW: being a woman, duration of the current sciatica episode more than 3 months, having had sciatica before, worse general health, and greater sciatica bothersomeness, fear-avoidance work, emotional distress, and back pain. In the final multivariate analysis, the following variables were independently and significantly associated with a longer time to sustained RTW: duration of the
Fig. 2. Data from the National Sickness Benefit Register. Proportion of patients who at baseline received sickness or rehabilitation benefits, returning to sustained work as a function of time.
current sciatica episode more than 3 months, having had sciatica before, greater sciatica bothersomeness, fearavoidance work, and back pain. Surgery was significantly negatively associated with time to sustained RTW in both the univariate analysis (hazard ratio [HR] 0.60; 95% CI 0.39, 0.93; p5.02) and multivariate analysis (HR 0.49; 95% CI 0.31, 0.79; p5.003). For both analyses (A and B), all variance inflation factor values were less than or equal to 2.0, and no significant interaction effects were found.
Discussion This study showed that about three-fourth of patients with sciatica had returned to full-time work at the 2-year follow-up. Less baseline sciatica bothersomeness, fearavoidance work, back pain, an episode duration less than 3 months, and no previous episodes predicted faster RTW, whereas surgical therapy was associated with slower RTW. In addition to less baseline sciatica bothersomeness and fear-avoidance work, younger age, better general health, and a negative straight-leg–raising test result were associated with higher self-reported RTW rates at 2 years. To our knowledge, only two previous articles have reported on prognostic factors for RTW in patients with sciatica, and both focused on the effect of the workers’ baseline compensation status. Neither the Spine Patient Outcome Research Trial (SPORT) [34] nor the Maine Lumbar Spine Study [35] found evidence that workers’ compensation was significantly related to work status at the 2- and 4-year follow-up, respectively. Multivariate analysis of the Maine study data indicated that younger age, better perception of general health at baseline, and less severe low back pain were associated with higher RTW rates 4 years later [35]. In studies restricted to surgical patients, Schade et al. [36] reported that depression and occupational mental stress and Graver et al. [37] reported that being a woman, having low stature and a long duration of sickness absence, and performing strenuous work activities were negative prognostic factors for RTW. In the present study, baseline sciatica bothersomeness was independently related to both the time to sustained RTW and RTW at 2 years. Disability, as measured by the Maine-Seattle Back Questionnaire score, was associated with RTW in the univariate but not in the multivariate analysis. The Oswestry Disability Index has been shown to predict loss of working time after disc surgery [38]. It is not known whether back pain disability instruments differ in their ability to predict work-related outcomes in patients with sciatica. In patients with nonspecific low back pain, it has been postulated that high levels of fear of pain, labeled as fear avoidance, may lead to self-limitations of movement and activity [39,40] and may consequently hinder recovery [41]. So far, there has been little support for this hypothesis about
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
work-related recovery [42–44]. Our results indicate that fear avoidance related to work may be a prognostic factor for RTW in patients with sciatica. However, in a study of workers with musculoskeletal disorders [45], the impact of fear avoidance on the duration of total compensation benefits lost significance when work-related variables were included in the multivariate model. In the present study, these variables were not obtained. Our results relating to fear of movement/reinjury are similar to studies on nonspecific low back pain [46,47] that showed that higher scores on the Tampa Scale for Kinesiophobia were significantly related to RTW in univariate but not in multivariate analysis. In contrast with studies on nonspecific low back pain [7,48], our multivariate analysis study found no association between RTW and emotional distress or psychosomatic complaints. Consistent with the SPORT study [17,49], our study did not find surgery to be related to self-reported RTW at the 2year follow-up. However, surgery was significantly associated with slower return to sustained work. Our estimates of the time to sustained work are conservative because the Cox regression analysis was performed using the time from the date of surgery, not the date of inclusion in the study, as the outcome measure. However, the observational design of this study precludes the ability to draw causal inferences about treatment effects. The surgical patients were more severely affected in terms of symptom duration, pain, disability, and signs of radiculopathy assessed by the straight-leg–raising test. The statistical analyses were adjusted for some of these differences, but it is possible that surgery acted as a proxy for variables that were unmeasured or not included in the analyses. Thus, our results indicating slower return to sustained work in surgical patients should be interpreted with caution. To fully assess the effect of surgery on RTW in sciatica, a randomized controlled trial is necessary. However, an operative treatment protocol standardized across all patients has proven to be difficult to achieve in sciatica. The few randomized trials that have compared the effectiveness of surgical versus conservative treatment [17,50,51] have all high rates of nonadherence to the assigned treatment groups. In the SPORT study [17], 60% of those who initially had been randomized to surgery actually became operated, whereas 45% of those assigned to conservative therapy did receive surgery. The abundance of existing conservative treatments makes them even harder to standardize. To assess the time to sustained RTW, we used the definition more than 60 days without receiving sickness benefits as the outcome. The time to the end of the first sickness absence episode, which is used frequently in the literature, ignores multiple absences and underestimates the total duration of work disability in people with low back pain [30]. No consensus exists for the definition of sustained RTW. Our choice of 60 days was based on a previous Norwegian study on low back pain [52]. Kuijer et al. [53] recommended using a definition of full return to regular work with a minimum duration of 6 weeks.
1855
Collection of administrative data from the sickness benefit register began after about half of the patients were included in this study. Because the baseline variables and self-reported work status at 2 years were almost identical in patients with and without sickness benefit register data, we believe that the results of the analysis of the time to sustained RTW are representative for the total cohort. A small discrepancy was found between the proportion of patients who, according to the register, were off the sick list at 2 years (78.4%) and the proportion who, according to selfreport, were working full time (71.1%). This difference may reflect differences between responders and nonresponders, uncertainty by the patients in reporting work status, or coding failures at the register. Only patients who were off the register’s sick list because of back pain or sciatica were counted, and some patients may have been sick listed by other diagnoses. Strengths and limitations The strengths of this study are the relatively large sample size, multicenter design, use of validated instruments, strict inclusion criteria, high response rate, and access to data from a public registry. The main limitations are that our cohort was a sample of patients referred to the clinics and did not include all sick-listed sciatica patients in need of secondary care in the target population. We have no data for patients who were not referred to the participating centers. The patients included in the present study may have had either more severe symptoms or a lower threshold for seeking medical assistance than patients who were not referred. Patient demographics, symptoms, and signs at baseline closely resemble those of the Maine Lumbar Spine Study [54] and the observational part of the SPORT study [49]. We did query about the use of pain and sleep medication, visits to general practitioners, physiotherapists, chiropractors, admissions to hospitals, and rehabilitation centers. However, the details or the exact nature of the treatments or advise given were not recorded. This precludes any conjecture about the impact of conservative care on the observed outcomes. Although available evidence suggests that the effect of epidural corticosteroid injections is small and of short duration [55], they are a common treatment in many secondary treatment centers. The fact that patients in this study not received epidural injections may limit the generalizability of our findings. Because the study took place in secondary care, the results may not be valid for patients in primary care. Administrative data were collected for only half of the cohort, potentially reducing the strength of the Cox regression analysis. More than 100 different determinants of prolonged disability and/or RTW have been identified in the literature [56], and it was not possible to include all these in the present study. In particular, socioeconomic and workplace factors have been shown to influence the RTW process [6,7,56]. Our results might have
1856
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
been different if more potential prognostic factors had been included. In conclusion, one-fourth of sick-listed patients with sciatica were still out of work at the 2-year follow-up. Factors associated with RTW identified in multivariate analysis were age, general health, sciatica bothersomeness, fearavoidance work, the straight-leg–raising test result, back pain, duration of the current episode, and the number of previous sciatica episodes. Surgical treatment was associated with slower RTW. Acknowledgments Funding for this study was received from the SouthEastern Norway Regional Health Authority; no support was received from commercial sources. None of the authors has financial interests that could create a potential conflict of interest with regard to the work. We thank Eli Molde Hagen, Dag Soldal, Knut Morten Huneide, Anett Bjørnødegard H angmann, and Bjarte Justnæs for their help with collecting data and Professor Leiv Sandvik for statistical advice. References [1] Andersson GB. Epidemiological features of chronic low-back pain. Lancet 1999;354:581–5. [2] Hagen KB, Thune O. Work incapacity from low back pain in the general population. Spine 1998;23:2091. [3] Selim AJ, Ren XS, Fincke G, et al. The importance of radiating leg pain in assessing health outcomes among patients with low back pain. Results from the veterans health study. Spine 1998;23:470–4. [4] Jensen OK, Nielsen CV, Stengaard-Pedersen K. One-year prognosis in sick-listed low back pain patients with and without radiculopathy. Prognostic factors influencing pain and disability. Spine J 2010;10: 659–75. [5] Andersson GB, Svensson HO, Oden A. The intensity of work recovery in low back pain. Spine 1983;8:880–4. [6] Steenstra IA, Verbeek JH, Heymans MW, Bongers PM. Prognostic factors for duration of sick leave in patients sick listed with acute low back pain: a systematic review of the literature. Occup Environ Med 2005;62:851–60. [7] Crook J, Milner R, Schultz IZ, Stringer B. Determinants of occupational disability following a low back injury: a critical review of the literature. J Occup Rehabil 2002;12:277–95. [8] Rydevik B, Brown MD, Lundborg G. Pathoanatomy and pathophysiology of nerve root compression. Spine 1984;9:7–15. [9] Mulleman D, Mammou S, Griffoul I, et al. Pathophysiology of diskrelated sciatica. I.—evidence supporting a chemical component. Joint Bone Spine 2006;73:151–8. [10] Heliovaara M, Impivaara O, Sievers K, et al. Lumbar disc syndrome in Finland. J Epidemiol Community Health 1987;41:251–8. [11] Younes M, Bejia I, Aguir Z, et al. Prevalence and risk factors of disk-related sciatica in an urban population in Tunisia. Joint Bone Spine 2006;73:538–42. [12] Koes BW, van Tulder MW, Peul WC. Diagnosis and treatment of sciatica. BMJ 2007;334:1313–7. [13] Chou R, Baisden J, Carragee EJ, et al. Surgery for low back pain: a review of the evidence for an American pain society clinical practice guideline. Spine 2009;34:1094–109. [14] Gibson JN, Waddell G. Surgical interventions for lumbar disc prolapse: updated Cochrane review. Spine 2007;32:1735–47.
[15] Weinstein JN, Lurie JD, Olson PR, et al. United States’ trends and regional variations in lumbar spine surgery: 1992-2003. Spine 2006;31:2707–14. [16] Sherman J, Cauthen J, Schoenberg D, et al. Economic impact of improving outcomes of lumbar discectomy. Spine J 2010;10: 108–16. [17] Weinstein JN, Tosteson TD, Lurie JD, et al. Surgical vs nonoperative treatment for lumbar disk herniation: the spine patient outcomes research trial (SPORT): a randomized trial. JAMA 2006;296: 2441–50. [18] International Classification Committee of WONCA. ICPC-2 International Classification of Primary Care. 2nd ed. Oxford, UK: Oxford University Press, 1998. [19] Patrick DL, Deyo RA, Atlas SJ, et al. Assessing health-related quality of life in patients with sciatica. Spine 1995;20:1899–908. [20] Grovle L, Haugen AJ, Keller A, et al. Reliability, validity, and responsiveness of the Norwegian versions of the Maine-Seattle back questionnaire and the sciatica bothersomeness and frequency indices. Spine 2008;33:2347–53. [21] Atlas SJ, Deyo RA, van den Ancker M, et al. The Maine-Seattle back questionnaire: a 12-item disability questionnaire for evaluating patients with lumbar sciatica or stenosis: results of a derivation and validation cohort analysis. Spine 2003;28:1869–76. [22] Ware JE Jr. SF-36 health survey update. Spine 2000;25:3130–9. [23] Eriksen HR, Ihlebaek C, Ursin H. A scoring system for subjective health complaints (SHC). Scand J Public Health 1999;27:63–72. [24] Grovle L, Haugen AJ, Ihlebaek CM, et al. Comorbid subjective health complaints in patients with sciatica: a prospective study including comparison with the general population. J Psychosom Res 2011;70:548–56. [25] Derogatis LR, Lipman RS, Rickels K, et al. The Hopkins symptom checklist (HSCL): a self-report symptom inventory. Behav Sci 1974;19:1–15. [26] Waddell G, Newton M, Henderson I, et al. A fear-avoidance beliefs questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain 1993;52:157–68. [27] Grotle M, Brox JI, Vollestad NK. Reliability, validity and responsiveness of the fear-avoidance beliefs questionnaire: methodological aspects of the Norwegian version. J Rehabil Med 2006;38:346–53. [28] Clark ME, Kori SH, Brockel J. Kinesiophobia and chronic pain: psychometric characteristics and factor analysis of the Tampa scale. Paper presented at: 15th Annual Scientific Meeting of the American Pain Society; September 1996; Washington, DC. [29] Haugen AJ, Grovle L, Keller A, Grotle M. Cross-cultural adaptation and validation of the Norwegian version of the Tampa scale for kinesiophobia. Spine 2008;33:E595–601. [30] Krause N, Dasinger LK, Deegan LJ, et al. Alternative approaches for measuring duration of work disability after low back injury based on administrative workers’ compensation data. Am J Ind Med 1999;35:604–18. [31] Roelen CA, Koopmans PC, Anema JR, van der Beek AJ. Recurrence of medically certified sickness absence according to diagnosis: a sickness absence register study. J Occup Rehabil 2010;20: 113–21. [32] Fisher LD, Lin DY. Time-dependent covariates in the Cox proportional-hazards regression model. Annu Rev Public Health 1999;20:145–57. [33] O’Brien RM. A caution regarding rules of thumb for variance inflation factors. Qual Quant 2007;41:673–90. [34] Atlas SJ, Tosteson TD, Blood EA, et al. The impact of workers’ compensation on outcomes of surgical and nonoperative therapy for patients with a lumbar disc herniation: SPORT. Spine 2010;35: 89–97. [35] Atlas SJ, Chang Y, Keller RB, et al. The impact of disability compensation on long-term treatment outcomes of patients with sciatica due to a lumbar disc herniation. Spine 2006;31: 3061–9.
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857 [36] Schade V, Semmer N, Main CJ, et al. The impact of clinical, morphological, psychosocial and work-related factors on the outcome of lumbar discectomy. Pain 1999;80:239–49. [37] Graver V, Ljunggren AE, Loeb M, et al. Background variables (medical history, anthropometric and biological factors) in relation to the outcome of lumbar disc surgery. Scand J Rehabil Med 1998;30:221–5. [38] Puolakka K, Ylinen J, Neva MH, et al. Risk factors for back painrelated loss of working time after surgery for lumbar disc herniation: a 5-year follow-up study. Eur Spine J 2008;17:386–92. [39] Lethem J, Slade PD, Troup JD, Bentley G. Outline of a fearavoidance model of exaggerated pain perception—I. Behav Res Ther 1983;21:401–8. [40] Vlaeyen JW, Linton SJ. Fear-avoidance and its consequences in chronic musculoskeletal pain: a state of the art. Pain 2000;85: 317–32. [41] Leeuw M, Goossens MEJB, Linton SJ, et al. The fear-avoidance model of musculoskeletal pain: current state of scientific evidence. J Behav Med 2007;30:77–94. [42] Pincus T, Smeets RJ, Simmonds MJ, Sullivan MJ. The fear avoidance model disentangled: improving the clinical utility of the fear avoidance model. Clin J Pain 2010;26:739–46. [43] Iles RA, Davidson M, Taylor NF. Psychosocial predictors of failure to return to work in non-chronic non-specific low back pain: a systematic review. Occup Environ Med 2008;65:507–17. [44] Pincus T, Vogel S, Burton AK, et al. Fear avoidance and prognosis in back pain: a systematic review and synthesis of current evidence. Arthritis Rheum 2006;54:3999–4010. [45] Lotters F, Franche RL, Hogg-Johnson S, et al. The prognostic value of depressive symptoms, fear-avoidance, and self-efficacy for duration of lost-time benefits in workers with musculoskeletal disorders. Occup Environ Med 2006;63:794–801. [46] Heymans MW, de Vet HCW, Knol DL, et al. Workers’ beliefs and expectations affect return to work over 12 months. J Occup Rehabil 2006;16:685–95.
1857
[47] Werneke M, Hart DL. Centralization phenomenon as a prognostic factor for chronic low back pain and disability. Spine 2001;26: 758–64; discussion 765. [48] Linton SJ, Gross D, Schultz IZ, et al. Prognosis and the identification of workers risking disability: research issues and directions for future research. J Occup Rehabil 2005;15:459–74. [49] Weinstein JN, Lurie JD, Tosteson TD, et al. Surgical vs nonoperative treatment for lumbar disk herniation: the spine patient outcomes research trial (SPORT) observational cohort. JAMA 2006;296: 2451–9. [50] Osterman H, Seitsalo S, Karppinen J, Malmivaara A. Effectiveness of microdiscectomy for lumbar disc herniation: a randomized controlled trial with 2 years of follow-up. Spine 2006;31:2409–14. [51] Peul WC, van den Hout WB, Brand R, et al, Leiden-The Hague Spine Intervention Prognostic Study Group. Prolonged conservative care versus early surgery in patients with sciatica caused by lumbar disc herniation: two year results of a randomised controlled trial. BMJ 2008;336:1355–8. [52] Reiso H, Nygard JF, Jorgensen GS, et al. Back to work: predictors of return to work among patients with back disorders certified as sick: a two-year follow-up study. Spine 2003;28:1468–73; discussion 1473–4. [53] Kuijer W, Groothoff JW, Brouwer S, et al. Prediction of sickness absence in patients with chronic low back pain: a systematic review. J Occup Rehabil 2006;16:439–67. [54] Atlas SJ, Keller RB, Chang Y, et al. Surgical and nonsurgical management of sciatica secondary to a lumbar disc herniation: five-year outcomes from the Maine lumbar spine study. Spine 2001;26: 1179–87. [55] Pinto RZ, Maher CG, Ferreira ML, et al. Epidural corticosteroid injections in the management of sciatica: a systematic review and meta-analysis. Ann Intern Med 2012;157:865–77. [56] Krause N, Frank JW, Dasinger LK, et al. Determinants of duration of disability and return-to-work after work-related injury and illness: challenges for future research. Am J Ind Med 2001;40:464–84.
Clinical Study
Prognostic factors for return to work in patients with sciatica Lars Grøvle, MD, PhDa,*, Anne J. Haugen, MD, PhDa, Anne Keller, MD, PhDb, B ard Ntvig, MD, PhDc,d, Jens I. Brox, MD, PhDe, Margreth Grotle, PT, PhDc,f a Department of Rheumatology, Østfold Hospital Trust, Postboks 16, 1603 Fredrikstad, Norway Department of Physical Medicine and Rehabilitation, Oslo University Hospital, Ullevaal, Postboks 4950 Nydalen, 0424 Oslo, Norway c National Resource Centre for Rehabilitation in Rheumatology, Diakonhjemmet Hospital, Postboks 23 Vinderen, 0319 Oslo, Norway d Department of General Practice, Institute of Health and Society, University of Oslo, Postboks 1130 Blindern, 0318 Oslo, Norway e Orthopaedic Department, Section for Back Surgery, Oslo University Hospital, Rikshospitalet, Postboks 4950 Nydalen, 0424 Oslo, Norway f FORMI (Communication Unit for Musculoskeletal Disorders), Division of Neuroscience, Oslo University Hospital, Ullevaal, Postboks 4950 Nydalen, 0424 Oslo, Norway b
Received 25 August 2011; revised 26 June 2013; accepted 10 July 2013
Abstract
BACKGROUND CONTEXT: Little is known about the prognostic factors for work-related outcomes of sciatica caused by disc herniation. PURPOSE: To identify the prognostic factors for return to work (RTW) during a 2-year follow-up among sciatica patients referred to secondary care. STUDY DESIGN/SETTING: Multicenter prospective cohort study including 466 patients. Administrative data from the National Sickness Benefit Register were accessed for 227 patients. PATIENT SAMPLE: Two samples were used. Sample A comprised patients who at the time of inclusion in the cohort reported being on partial sick leave or complete sick leave or were undergoing rehabilitation because of back pain/sciatica. Sample B comprised patients who, according to the sickness benefit register, at the time of inclusion received sickness benefits or rehabilitation allowances because of back pain/sciatica. OUTCOME MEASURES: In Sample A, the outcome was self-reported return to full-time work at the 2-year follow-up. In Sample B, the outcome was time to first sustained RTW, defined as the first period of more than 60 days without receiving benefits from the register. METHODS: Significant baseline predictors of self-reported RTW at 2 years (Analysis A) were identified by multivariate logistic regression. Significant predictors of time to sustained RTW (Analysis B) were identified by multivariate Cox proportional hazard modeling. Both analyses included adjustment for age and sex. To assess the effect of surgery on the probability of RTW, analyses similar to A and B were performed, including the variable surgery (yes/no). RESULTS: One-fourth of the patients were still out of work at the 2-year follow-up. In Sample A (n5237), younger age, better general health, lower baseline sciatica bothersomeness, less fearavoidance work, and a negative straight-leg–raising test result were significantly associated with a higher probability of RTW at the 2-year follow-up. Surgery was not significantly associated with the outcome. In Sample B (n5125), history of sciatica, duration of the current sciatica episode more than 3 months, greater sciatica bothersomeness, fear-avoidance work, and back pain were significantly associated with a longer time to sustained RTW. Surgery was significantly negatively associated with time to sustained RTW both in univariate (hazard ratio [HR] 0.60; 95% confidence interval [CI] 0.39, 0.93; p5.02) and in multivariate (HR 0.49; 95% CI 0.31, 0.79; p5.003) analyses. CONCLUSIONS: The baseline factors associated with RTW identified in multivariate analysis were age, general health, history of sciatica, duration of the current episode, baseline sciatica bothersomeness, fear-avoidance work, back pain, and the straight-leg–raising test result.
FDA device/drug status: Not applicable. Author disclosures: LG: Nothing to disclose. AJH: Nothing to disclose. AK: Nothing to disclose. BN: Nothing to disclose. JIB: Nothing to disclose. MG: Nothing to disclose. 1529-9430/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.spinee.2013.07.433
* Corresponding author. Department of Rheumatology, Østfold Hospital Trust, Postboks 16, 1603 Fredrikstad, Norway. Tel.: (47) 91814029. E-mail address: [email protected] (L. Grøvle)
1850
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
Surgical treatment was associated with slower RTW, but surgical patients were more severely affected than patients treated without surgery; so, this finding should be interpreted with caution. Ó 2013 Elsevier Inc. All rights reserved. Keywords:
Sciatica; Prognosis; Return to work; Sickness absence; Fear-avoidance beliefs
Introduction More than 10% of all annual sickness absence days are taken by people with a diagnosis of low back pain [1,2]. Among those with low back pain, patients with sciatica generally report more severe pain, have longer absence, and lower rates of return to work (RTW) than patients with nonspecific low back pain [2–7]. In about 90% of cases, sciatica, defined as radiating leg pain below the knee, is caused by a herniated disc with nerve root compression. The compression and secondary inflammation of the nearby nerve roots and dorsal root ganglia [8,9] may cause radiating pain, muscular weakness, sensory disturbances, and depressed tendon reflexes. Some patients also experience bladder, bowel, and genital dysfunction. The condition can vary from short-lasting single episodes to a remitting or permanent course over months or years. The point prevalence of sciatica has been estimated as 2% to 5%, and this condition affects mainly the working-age population [10,11]. In a review, Koes et al. [12] recommended referral to secondary care for patients with intractable radicular pain or whose pain does not diminish after 6 to 8 weeks of conservative care. Surgical discectomy is considered to improve the short-term but not the long-term prognosis of pain and perceived disability [13,14]. Although rates vary considerably [15], surgery for sciatica is a common procedure. In the United States, about one discectomy is performed per 1,000 persons each year [16]. A randomized study did not find an effect of surgery on work status [17]. Despite the frequency and severity of this condition, few studies have investigated the prognostic factors for workrelated outcomes in sciatica. The aim of this study was to identify the prognostic factors for RTW during a 2-year follow-up among sciatica patients referred to secondary care.
Methods Setting This study was part of a prospective cohort study of patients with sciatica and disc herniation referred to back clinics at four hospitals in Southeastern Norway. Patients were invited to participate by the clinic staff. Study participation did not involve any specific type of intervention. The consultation included information about the condition and general advise to stay active and use pain medication if necessary. In patients with severe symptoms, surgery
was performed at the discretion of the individual surgeon. The study protocol was approved by the Regional Committee for Medical Research Ethics and the Ombudsman for Privacy in Research at the Norwegian Social Science Data Services. Patients Patients included in the study were 18 years of age or older, had radiating pain below the knee and/or paresis, and had a lumbar disc herniation verified by a magnetic resonance imaging or computed tomography scan of the corresponding level and side. The exclusion criteria were pregnancy, spinal fracture, tumor, infection, previous surgery in the affected disc, and inability to communicate in written Norwegian. After a protocol amendment, patients included after October 2005 were asked to give a consent for us to obtain sickness absence data from the National Sickness Benefit Register, which covers the entire Norwegian population. Procedures On the day of inclusion, a clinical examination was performed by a physician or physiotherapist, and the patient completed a comprehensive questionnaire. Self-reported data A follow-up assessment was conducted at 2 years using a mailed questionnaire that was completed at home and returned in a prepaid envelope. Patients who had not responded 2 weeks after the scheduled date were contacted by telephone or text message. A reminder letter was sent to nonresponders if no reply was obtained after 3 weeks. Self-reported current work status at baseline and at the 2year follow-up included the categories full-time work, partial sick leave, complete sick leave, rehabilitation, disability pension, student, job seeker, old-age retirement, or homemaker. Administrative data Data obtained from the sickness benefit register included the starting and stopping dates of sickness absence, rehabilitation, and disability benefits. To qualify for sickness benefit in Norway, the occupational disability must be documented by a doctor’s sick leave certificate, which is submitted to the register. If the person is still unable to work after 1 year, he or she may be entitled to a rehabilitation allowance or disability benefits. In the present study, the
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
1851
sickness absence or rehabilitation allowance certificate diagnoses indicating back pain or sciatica were classified according to the International Classification of Primary Care [18]: L02 (back symptom/complaint), L04 (low back symptom/complaint), L84 (back syndrome without radiating pain), and L86 (back syndrome with radiating pain).
!60 ), sensation (deemed abnormal if reduced light touch), reflexes (deemed abnormal if depressed patellar or Achilles reflexes), and muscle strength (deemed abnormal if reduced performance of the single limb stance, tiptoe or heel walking, supine knee or ankle flexion/extension, or big toe extension).
Potential prognostic factors
Outcomes
Information about demographic variables, education, and back pain/sciatica history were obtained at baseline. Leg and back pain were rated on a 0 to 100 visual analog scale, and a composite of perceived symptoms (leg pain, numbness, weakness, and back or leg pain while sitting) was measured using the Sciatica Bothersomeness Index [19,20]. Disability was assessed by the Maine-Seattle Back Questionnaire [20,21], a sciatica-specific shortened version of the Roland-Morris Questionnaire. General health was assessed by the Short-Form 36 general health subscale [22]. Selfreported comorbidity was measured by 27 items of the Subjective Health Complaint Inventory [23], a list of common somatic or psychological complaints. The scoring procedure used in this study has been described elsewhere [24]. Emotional distress was measured by the Hopkins Symptom Check List 25 [25]. The work subscale of the FearAvoidance Back Questionnaire [26,27] was used to assess fear-avoidance beliefs regarding work (here, called fearavoidance work). Pain-related fear of movement/reinjury was measured by a 13-item version of the Tampa Scale for Kinesiophobia [28,29]. The clinical tests were dichotomized and included straight-leg raising (deemed abnormal if
RTW at 2 years by self-report Patients who at the time of inclusion reported being on partial sick leave or complete sick leave or undergoing rehabilitation because of back pain/sciatica were included in this data set. The outcome was return to full-time work at the 2-year follow-up. Patients who reported being employed in full-time work were given the code 1, whereas those who reported being on partial sick leave, complete sick leave, rehabilitation, or disability pension were given the code 0. Patients who at the 2-year follow-up reported being a student, old-age pensioner, or homemaker were excluded from the analyses. Time to sustained RTW assessed from the administrative data Patients who at the time of inclusion according to the sickness benefit register received sickness benefits or rehabilitation allowances because of back pain/sciatica were included in this data set. The outcome was time to first sustained RTW, which was defined by being off the register’s sick list as a proxy for RTW. Sustained RTW was chosen to avoid misclassifications because of sickness
Fig. 1. Flow chart: from the whole cohort to analyzed samples.
1852
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
surgery set as the date of inclusion in the study [32]. Multicollinearity was assessed by the variance inflation factor using a value less than 5 as the acceptance criterion [33]. Variables in the final models were examined for significant interaction effects.
absence recurrences [30,31] and was defined as the number of calendar days from inclusion to the first period of more than 60 days without receiving benefits from the register. Statistical analyses Group differences were analyzed using Mann-Whitney U test, Student t test, or chi-square test according to the type of the data. Significant predictors of self-reported RTW at 2 years (Analysis A) were identified by logistic regression. Significant predictors of time to sustained RTW (Analysis B) were identified by Cox proportional hazard modeling. Patients who had not returned to sustained work at 2 years were classified as censored. In both analyses (A and B), multivariate models were built by including potential prognostic factors with p values less than .2 in the univariate analyses. Variables with the highest p value were removed one by one until all remaining variables had p values less than .05. Both analyses included adjustment for age and sex. To assess the effect of surgery on the probability of RTW, analyses similar to A and B were performed, including the variable surgery (yes/no). Cox regression analysis was then performed with the date of
Results During 2005 and 2006, a total of 466 patients, with a mean age of 43.6 (standard deviation [SD] 11.5) years, were enrolled in the prospective cohort study. All patients included after October 2005 (n5227) gave consent to obtain sickness absence data from the National Sickness Benefit Register. A flow chart from the whole cohort to analyzed samples is shown in Figure 1. RTW at 2 years by self-report At the time of inclusion, 297 patients reported being on complete sick leave (n5227), partial sick leave (n554), or undergoing rehabilitation (n516). The 2-year follow-up assessments were completed by 246 patients (82.3%). Those who at 2 years reported being a student (n55), in old-age
Table 1 Patients’ baseline demographic and clinical characteristics according to the type of analysis performed A
B
Patient characteristics
RTW at 2 y, n5237
Time to sustained RTW, n5125
Men Mean (SD) age, y Mean (SD) education, y Married/cohabitant Current smoker Duration back problems !1 y 1–5 y O5 y Number of previous sciatica episodes 0 1–4 $5 Mean (SD) duration of current sciatica episode, wk Mean (SD) back pain (0–100) Mean (SD) leg pain (0–100) Mean (SD) sciatica bothersomeness (0–24) Mean (SD) disability* (0–12) Mean (SD) fear-avoidance beliefs, work (0–42) Mean (SD) fear of movement/reinjury (13–52) Mean (SD) emotional distress (0–4) Mean (SD) subjective health complaints (0–27) Mean (SD) general health (0–100)y Clinical findings Positive straight-leg–raising test Motor weakness Reflexes depressed Sensory decrease
136 43.2 13.2 181 99
68 42.1 13.1 86 55
RTW, return to work; SD, standard deviation. Values are n (%) unless otherwise stated. * Measured by the Maine-Seattle Back Questionnaire. y Higher values indicate better health.
(57.4) (9.2) (3.0) (76.7) (41.9)
73 (30.8) 53 (22.4) 111 (46.8)
(54.4) (10.4) (3.1) (69.9) (44.4)
44 (35.2) 30 (24.0) 51 (40.8)
106 98 32 21.6 43.8 64.6 14.9 8.5 20.1 27.6 1.6 7.3 71.1
(44.9) (41.5) (13.6) (24.1) (30.0) (26.9) (4.8) (2.3) (10.0) (6.8) (0.4) (4.1) (19.4)
61 49 15 19.0 49.0 70.6 15.5 8.8 21.2 28.1 1.6 7.8 68.7
(48.8) (39.2) (12.0) (14.5) (30.3) (24.3) (4.6) (2.3) (9.9) (6.7) (0.4) (4.1) (19.5)
140 108 108 138
(60.6) (47.0) (46.4) (58.7)
88 56 59 76
(71.5) (45.5) (48.4) (61.3)
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
1853
Table 2 Associations between potential prognostic variables and RTW according to the type of analysis A
B
RTW at 2 y by self-report (n5237)
Time to sustained RTW by administrative data (n5125)
Univariate Prognostic factors Sociodemographic characteristics Age (y) Women Married/cohabitant Education (y) Current smoker Sciatica history Duration current sciatica episode O3 mo Had sciatica before Duration of back problems O1 y Self-reported health status Subjective health complaints (0–27) Sciatica bothersomeness (0–24) Disability (0–12)y Fear-avoidance beliefs, work (0–42) Fear of movement/reinjury (13–52) General health (0–100)z Emotional distress (0–4) Back pain (0–100) Leg pain (0–100) Clinical findings Positive straight-leg–raising test Motor weakness Reflexes depressed Sensory decrease
Multivariate*
Univariate
Multivariate*
OR
95% CI
p
OR
95% CI
p
HR
95% CI
p
HR
95% CI
p
0.98 0.51 0.89 1.11 0.64
0.95–1.01 0.29–0.91 0.45–1.78 1.00–1.23 0.36–1.14
.24 .02 .75 .04 .13
0.97 0.61
0.93–1.00 0.31–1.22
.07 .17
1.00 0.66 0.81 1.03 0.77
0.98–1.02 0.45–0.98 0.53–1.24 0.97–1.10 0.52–1.14
.86 .04 .34 .29 .19
1.00 0.81
0.98–1.02 0.54–1.23
.78 .33
0.56 0.60 0.61
0.31–1.03 0.33–1.09 0.31–1.17
.06 .09 .14
0.63 0.64 0.88
0.43–0.93 0.43–0.94 0.59–1.30
.02 .02 .52
0.63 0.66
0.42–0.93 0.44–1.00
.02 .05
0.85 0.86 0.82 0.93 0.94 1.04 0.25 0.99 0.99
0.79–0.92 0.81–0.93 0.72–0.94 0.90–0.96 0.90–0.98 1.02–1.05 0.12–0.51 0.98–1.00 0.97–1.00
.00 .00 .00 .00 .00 .00 .00 .08 .01
0.97 0.93 0.93 0.98 0.98 1.01 0.47 0.99 1.00
0.93–1.01 0.89–0.97 0.86–1.01 0.96–0.99 0.96–1.01 1.00–1.02 0.29–0.77 0.99–1.00 0.99–1.00
.17 .00 .10 .01 .26 .01 .00 .01 .17
0.94
0.90–0.99
.02
0.97
0.95–0.99
.00
0.99
0.99–1.00
.04
0.48 0.56 1.07 1.19
0.26–0.92 0.31–1.02 0.60–1.91 0.67–2.14
.03 .06 .83 .55
0.78 0.81 0.96 1.07
0.51–1.19 0.55–1.20 0.66–1.42 0.73–1.59
.25 .30 .85 .72
0.89
0.82–0.97
.01
0.93
0.90–0.97
.00
1.03
1.01–1.05
.00
0.44
0.20–0.95
.04
RTW, return to work; OR, odds ratios; HR, hazard ratio; CI, confidence interval. * Final result of a stepwise backward selection model. y Measured by the Maine-Seattle Back Questionnaire. z Higher values indicate better health.
retirement (n52), or a homemaker (n52) were excluded, rendering 237 patients eligible for analysis. The baseline characteristics of these patients are shown in Table 1 (A). During the 2-year follow-up period, 88 patients (29.6%) received surgical therapy at a mean (SD) of 13.3 (18.4) weeks after inclusion in the study. Compared with those who did not have surgery, the surgically treated patients reported at baseline more disability (9.4 [2.0] vs. 8.2 [2.4], p!.01), more sciatica bothersomeness (16.6 [4.5] vs. 13.9 [4.7], p!.01), more leg pain (76.7 [22.2] vs. 60.0 [26.6], p!.01), more back pain (52.3 [31.0] vs. 41.6 [29.2], p!.01), and a longer duration of the current sciatica episode (23.2 [19.7] vs. 20.1 [23.0] weeks, p!.05). The proportion of patients with a positive straight-leg–raising test was also higher in the surgically treated group (75.6% vs. 56.6%, p!.01). In total, 54% and 37% of the patients had visited a physiotherapist or a chiropractor during the first and the second year of follow-up, respectively. No patients reported to have received epidural injections. The current work status at 2 years was full-time work n5172 (72.6%), partial sick leave n57 (3.0%), complete sick leave n515 (6.3%), rehabilitation n532 (13.5%), disability pension n510 (4.2%), and job seeker n51
(0.4%). The patients who were included in Sample A and who did not respond at the 2-year follow-up (n551) were younger (39.4 vs. 43.2 years, p!.05), reported higher baseline fear-avoidance work (23.8 vs. 20.1, p!.05), and were less likely to have a positive straight-leg–raising test (51.1% vs. 62.4%, p!.05) than those who responded. None of the other baseline variables differed significantly between responders and nonresponders. According to the sickness benefit register, the responders and nonresponders in Sample A did not differ significantly in terms of days to return to sustained work or the proportion of patients who were off the sick list for more than 60 days at 2 years. In the univariate analyses (Table 2 [A]), the following variables were associated with better outcome: being a man, having a higher educational level, better general health, and fewer subjective health complaints, reporting less sciatica bothersomeness and disability, and having less fear-avoidance work and emotional distress, and a negative straight-leg–raising test result. In the final multivariate analysis the following variables remained as independent and significant factors associated with RTW at the 2-year follow-up: age, general health, sciatica bothersomeness,
1854
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
fear-avoidance work, and the straight-leg–raising test result. Surgery was not significantly associated with the outcome in either the univariate or the multivariate analysis. The explained variance of the final model as assessed by the Nagelkerke R2 was 31%. Time to sustained RTW obtained from administrative data According to the sickness benefit register, 125 of the 227 patients included in the prospective cohort after October 2005 received either sickness benefits (n5121) or rehabilitation benefits (n54). The baseline characteristics of this sample are presented in Table 1 (B). At 2 years, 98 (78.4%) patients had been off the sick list for more than 60 days. Figure 2 shows the proportion of patients returning to sustained work as a function of time. The median time to sustained RTW was 169 days (95% confidence interval [CI] 123, 215). Forty-one (32.8%) patients had received surgical therapy at a mean of 11.2 (SD 14.7) weeks after the date of inclusion. Self-reported work status at the 2-year follow-up was nearly identical in samples A and B (data not shown). Results of the univariate and multivariate Cox regression analyses are shown in Table 2 (B). In the univariate analysis, the following variables were associated with longer time to sustained RTW: being a woman, duration of the current sciatica episode more than 3 months, having had sciatica before, worse general health, and greater sciatica bothersomeness, fear-avoidance work, emotional distress, and back pain. In the final multivariate analysis, the following variables were independently and significantly associated with a longer time to sustained RTW: duration of the
Fig. 2. Data from the National Sickness Benefit Register. Proportion of patients who at baseline received sickness or rehabilitation benefits, returning to sustained work as a function of time.
current sciatica episode more than 3 months, having had sciatica before, greater sciatica bothersomeness, fearavoidance work, and back pain. Surgery was significantly negatively associated with time to sustained RTW in both the univariate analysis (hazard ratio [HR] 0.60; 95% CI 0.39, 0.93; p5.02) and multivariate analysis (HR 0.49; 95% CI 0.31, 0.79; p5.003). For both analyses (A and B), all variance inflation factor values were less than or equal to 2.0, and no significant interaction effects were found.
Discussion This study showed that about three-fourth of patients with sciatica had returned to full-time work at the 2-year follow-up. Less baseline sciatica bothersomeness, fearavoidance work, back pain, an episode duration less than 3 months, and no previous episodes predicted faster RTW, whereas surgical therapy was associated with slower RTW. In addition to less baseline sciatica bothersomeness and fear-avoidance work, younger age, better general health, and a negative straight-leg–raising test result were associated with higher self-reported RTW rates at 2 years. To our knowledge, only two previous articles have reported on prognostic factors for RTW in patients with sciatica, and both focused on the effect of the workers’ baseline compensation status. Neither the Spine Patient Outcome Research Trial (SPORT) [34] nor the Maine Lumbar Spine Study [35] found evidence that workers’ compensation was significantly related to work status at the 2- and 4-year follow-up, respectively. Multivariate analysis of the Maine study data indicated that younger age, better perception of general health at baseline, and less severe low back pain were associated with higher RTW rates 4 years later [35]. In studies restricted to surgical patients, Schade et al. [36] reported that depression and occupational mental stress and Graver et al. [37] reported that being a woman, having low stature and a long duration of sickness absence, and performing strenuous work activities were negative prognostic factors for RTW. In the present study, baseline sciatica bothersomeness was independently related to both the time to sustained RTW and RTW at 2 years. Disability, as measured by the Maine-Seattle Back Questionnaire score, was associated with RTW in the univariate but not in the multivariate analysis. The Oswestry Disability Index has been shown to predict loss of working time after disc surgery [38]. It is not known whether back pain disability instruments differ in their ability to predict work-related outcomes in patients with sciatica. In patients with nonspecific low back pain, it has been postulated that high levels of fear of pain, labeled as fear avoidance, may lead to self-limitations of movement and activity [39,40] and may consequently hinder recovery [41]. So far, there has been little support for this hypothesis about
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
work-related recovery [42–44]. Our results indicate that fear avoidance related to work may be a prognostic factor for RTW in patients with sciatica. However, in a study of workers with musculoskeletal disorders [45], the impact of fear avoidance on the duration of total compensation benefits lost significance when work-related variables were included in the multivariate model. In the present study, these variables were not obtained. Our results relating to fear of movement/reinjury are similar to studies on nonspecific low back pain [46,47] that showed that higher scores on the Tampa Scale for Kinesiophobia were significantly related to RTW in univariate but not in multivariate analysis. In contrast with studies on nonspecific low back pain [7,48], our multivariate analysis study found no association between RTW and emotional distress or psychosomatic complaints. Consistent with the SPORT study [17,49], our study did not find surgery to be related to self-reported RTW at the 2year follow-up. However, surgery was significantly associated with slower return to sustained work. Our estimates of the time to sustained work are conservative because the Cox regression analysis was performed using the time from the date of surgery, not the date of inclusion in the study, as the outcome measure. However, the observational design of this study precludes the ability to draw causal inferences about treatment effects. The surgical patients were more severely affected in terms of symptom duration, pain, disability, and signs of radiculopathy assessed by the straight-leg–raising test. The statistical analyses were adjusted for some of these differences, but it is possible that surgery acted as a proxy for variables that were unmeasured or not included in the analyses. Thus, our results indicating slower return to sustained work in surgical patients should be interpreted with caution. To fully assess the effect of surgery on RTW in sciatica, a randomized controlled trial is necessary. However, an operative treatment protocol standardized across all patients has proven to be difficult to achieve in sciatica. The few randomized trials that have compared the effectiveness of surgical versus conservative treatment [17,50,51] have all high rates of nonadherence to the assigned treatment groups. In the SPORT study [17], 60% of those who initially had been randomized to surgery actually became operated, whereas 45% of those assigned to conservative therapy did receive surgery. The abundance of existing conservative treatments makes them even harder to standardize. To assess the time to sustained RTW, we used the definition more than 60 days without receiving sickness benefits as the outcome. The time to the end of the first sickness absence episode, which is used frequently in the literature, ignores multiple absences and underestimates the total duration of work disability in people with low back pain [30]. No consensus exists for the definition of sustained RTW. Our choice of 60 days was based on a previous Norwegian study on low back pain [52]. Kuijer et al. [53] recommended using a definition of full return to regular work with a minimum duration of 6 weeks.
1855
Collection of administrative data from the sickness benefit register began after about half of the patients were included in this study. Because the baseline variables and self-reported work status at 2 years were almost identical in patients with and without sickness benefit register data, we believe that the results of the analysis of the time to sustained RTW are representative for the total cohort. A small discrepancy was found between the proportion of patients who, according to the register, were off the sick list at 2 years (78.4%) and the proportion who, according to selfreport, were working full time (71.1%). This difference may reflect differences between responders and nonresponders, uncertainty by the patients in reporting work status, or coding failures at the register. Only patients who were off the register’s sick list because of back pain or sciatica were counted, and some patients may have been sick listed by other diagnoses. Strengths and limitations The strengths of this study are the relatively large sample size, multicenter design, use of validated instruments, strict inclusion criteria, high response rate, and access to data from a public registry. The main limitations are that our cohort was a sample of patients referred to the clinics and did not include all sick-listed sciatica patients in need of secondary care in the target population. We have no data for patients who were not referred to the participating centers. The patients included in the present study may have had either more severe symptoms or a lower threshold for seeking medical assistance than patients who were not referred. Patient demographics, symptoms, and signs at baseline closely resemble those of the Maine Lumbar Spine Study [54] and the observational part of the SPORT study [49]. We did query about the use of pain and sleep medication, visits to general practitioners, physiotherapists, chiropractors, admissions to hospitals, and rehabilitation centers. However, the details or the exact nature of the treatments or advise given were not recorded. This precludes any conjecture about the impact of conservative care on the observed outcomes. Although available evidence suggests that the effect of epidural corticosteroid injections is small and of short duration [55], they are a common treatment in many secondary treatment centers. The fact that patients in this study not received epidural injections may limit the generalizability of our findings. Because the study took place in secondary care, the results may not be valid for patients in primary care. Administrative data were collected for only half of the cohort, potentially reducing the strength of the Cox regression analysis. More than 100 different determinants of prolonged disability and/or RTW have been identified in the literature [56], and it was not possible to include all these in the present study. In particular, socioeconomic and workplace factors have been shown to influence the RTW process [6,7,56]. Our results might have
1856
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857
been different if more potential prognostic factors had been included. In conclusion, one-fourth of sick-listed patients with sciatica were still out of work at the 2-year follow-up. Factors associated with RTW identified in multivariate analysis were age, general health, sciatica bothersomeness, fearavoidance work, the straight-leg–raising test result, back pain, duration of the current episode, and the number of previous sciatica episodes. Surgical treatment was associated with slower RTW. Acknowledgments Funding for this study was received from the SouthEastern Norway Regional Health Authority; no support was received from commercial sources. None of the authors has financial interests that could create a potential conflict of interest with regard to the work. We thank Eli Molde Hagen, Dag Soldal, Knut Morten Huneide, Anett Bjørnødegard H angmann, and Bjarte Justnæs for their help with collecting data and Professor Leiv Sandvik for statistical advice. References [1] Andersson GB. Epidemiological features of chronic low-back pain. Lancet 1999;354:581–5. [2] Hagen KB, Thune O. Work incapacity from low back pain in the general population. Spine 1998;23:2091. [3] Selim AJ, Ren XS, Fincke G, et al. The importance of radiating leg pain in assessing health outcomes among patients with low back pain. Results from the veterans health study. Spine 1998;23:470–4. [4] Jensen OK, Nielsen CV, Stengaard-Pedersen K. One-year prognosis in sick-listed low back pain patients with and without radiculopathy. Prognostic factors influencing pain and disability. Spine J 2010;10: 659–75. [5] Andersson GB, Svensson HO, Oden A. The intensity of work recovery in low back pain. Spine 1983;8:880–4. [6] Steenstra IA, Verbeek JH, Heymans MW, Bongers PM. Prognostic factors for duration of sick leave in patients sick listed with acute low back pain: a systematic review of the literature. Occup Environ Med 2005;62:851–60. [7] Crook J, Milner R, Schultz IZ, Stringer B. Determinants of occupational disability following a low back injury: a critical review of the literature. J Occup Rehabil 2002;12:277–95. [8] Rydevik B, Brown MD, Lundborg G. Pathoanatomy and pathophysiology of nerve root compression. Spine 1984;9:7–15. [9] Mulleman D, Mammou S, Griffoul I, et al. Pathophysiology of diskrelated sciatica. I.—evidence supporting a chemical component. Joint Bone Spine 2006;73:151–8. [10] Heliovaara M, Impivaara O, Sievers K, et al. Lumbar disc syndrome in Finland. J Epidemiol Community Health 1987;41:251–8. [11] Younes M, Bejia I, Aguir Z, et al. Prevalence and risk factors of disk-related sciatica in an urban population in Tunisia. Joint Bone Spine 2006;73:538–42. [12] Koes BW, van Tulder MW, Peul WC. Diagnosis and treatment of sciatica. BMJ 2007;334:1313–7. [13] Chou R, Baisden J, Carragee EJ, et al. Surgery for low back pain: a review of the evidence for an American pain society clinical practice guideline. Spine 2009;34:1094–109. [14] Gibson JN, Waddell G. Surgical interventions for lumbar disc prolapse: updated Cochrane review. Spine 2007;32:1735–47.
[15] Weinstein JN, Lurie JD, Olson PR, et al. United States’ trends and regional variations in lumbar spine surgery: 1992-2003. Spine 2006;31:2707–14. [16] Sherman J, Cauthen J, Schoenberg D, et al. Economic impact of improving outcomes of lumbar discectomy. Spine J 2010;10: 108–16. [17] Weinstein JN, Tosteson TD, Lurie JD, et al. Surgical vs nonoperative treatment for lumbar disk herniation: the spine patient outcomes research trial (SPORT): a randomized trial. JAMA 2006;296: 2441–50. [18] International Classification Committee of WONCA. ICPC-2 International Classification of Primary Care. 2nd ed. Oxford, UK: Oxford University Press, 1998. [19] Patrick DL, Deyo RA, Atlas SJ, et al. Assessing health-related quality of life in patients with sciatica. Spine 1995;20:1899–908. [20] Grovle L, Haugen AJ, Keller A, et al. Reliability, validity, and responsiveness of the Norwegian versions of the Maine-Seattle back questionnaire and the sciatica bothersomeness and frequency indices. Spine 2008;33:2347–53. [21] Atlas SJ, Deyo RA, van den Ancker M, et al. The Maine-Seattle back questionnaire: a 12-item disability questionnaire for evaluating patients with lumbar sciatica or stenosis: results of a derivation and validation cohort analysis. Spine 2003;28:1869–76. [22] Ware JE Jr. SF-36 health survey update. Spine 2000;25:3130–9. [23] Eriksen HR, Ihlebaek C, Ursin H. A scoring system for subjective health complaints (SHC). Scand J Public Health 1999;27:63–72. [24] Grovle L, Haugen AJ, Ihlebaek CM, et al. Comorbid subjective health complaints in patients with sciatica: a prospective study including comparison with the general population. J Psychosom Res 2011;70:548–56. [25] Derogatis LR, Lipman RS, Rickels K, et al. The Hopkins symptom checklist (HSCL): a self-report symptom inventory. Behav Sci 1974;19:1–15. [26] Waddell G, Newton M, Henderson I, et al. A fear-avoidance beliefs questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain 1993;52:157–68. [27] Grotle M, Brox JI, Vollestad NK. Reliability, validity and responsiveness of the fear-avoidance beliefs questionnaire: methodological aspects of the Norwegian version. J Rehabil Med 2006;38:346–53. [28] Clark ME, Kori SH, Brockel J. Kinesiophobia and chronic pain: psychometric characteristics and factor analysis of the Tampa scale. Paper presented at: 15th Annual Scientific Meeting of the American Pain Society; September 1996; Washington, DC. [29] Haugen AJ, Grovle L, Keller A, Grotle M. Cross-cultural adaptation and validation of the Norwegian version of the Tampa scale for kinesiophobia. Spine 2008;33:E595–601. [30] Krause N, Dasinger LK, Deegan LJ, et al. Alternative approaches for measuring duration of work disability after low back injury based on administrative workers’ compensation data. Am J Ind Med 1999;35:604–18. [31] Roelen CA, Koopmans PC, Anema JR, van der Beek AJ. Recurrence of medically certified sickness absence according to diagnosis: a sickness absence register study. J Occup Rehabil 2010;20: 113–21. [32] Fisher LD, Lin DY. Time-dependent covariates in the Cox proportional-hazards regression model. Annu Rev Public Health 1999;20:145–57. [33] O’Brien RM. A caution regarding rules of thumb for variance inflation factors. Qual Quant 2007;41:673–90. [34] Atlas SJ, Tosteson TD, Blood EA, et al. The impact of workers’ compensation on outcomes of surgical and nonoperative therapy for patients with a lumbar disc herniation: SPORT. Spine 2010;35: 89–97. [35] Atlas SJ, Chang Y, Keller RB, et al. The impact of disability compensation on long-term treatment outcomes of patients with sciatica due to a lumbar disc herniation. Spine 2006;31: 3061–9.
L. Grøvle et al. / The Spine Journal 13 (2013) 1849–1857 [36] Schade V, Semmer N, Main CJ, et al. The impact of clinical, morphological, psychosocial and work-related factors on the outcome of lumbar discectomy. Pain 1999;80:239–49. [37] Graver V, Ljunggren AE, Loeb M, et al. Background variables (medical history, anthropometric and biological factors) in relation to the outcome of lumbar disc surgery. Scand J Rehabil Med 1998;30:221–5. [38] Puolakka K, Ylinen J, Neva MH, et al. Risk factors for back painrelated loss of working time after surgery for lumbar disc herniation: a 5-year follow-up study. Eur Spine J 2008;17:386–92. [39] Lethem J, Slade PD, Troup JD, Bentley G. Outline of a fearavoidance model of exaggerated pain perception—I. Behav Res Ther 1983;21:401–8. [40] Vlaeyen JW, Linton SJ. Fear-avoidance and its consequences in chronic musculoskeletal pain: a state of the art. Pain 2000;85: 317–32. [41] Leeuw M, Goossens MEJB, Linton SJ, et al. The fear-avoidance model of musculoskeletal pain: current state of scientific evidence. J Behav Med 2007;30:77–94. [42] Pincus T, Smeets RJ, Simmonds MJ, Sullivan MJ. The fear avoidance model disentangled: improving the clinical utility of the fear avoidance model. Clin J Pain 2010;26:739–46. [43] Iles RA, Davidson M, Taylor NF. Psychosocial predictors of failure to return to work in non-chronic non-specific low back pain: a systematic review. Occup Environ Med 2008;65:507–17. [44] Pincus T, Vogel S, Burton AK, et al. Fear avoidance and prognosis in back pain: a systematic review and synthesis of current evidence. Arthritis Rheum 2006;54:3999–4010. [45] Lotters F, Franche RL, Hogg-Johnson S, et al. The prognostic value of depressive symptoms, fear-avoidance, and self-efficacy for duration of lost-time benefits in workers with musculoskeletal disorders. Occup Environ Med 2006;63:794–801. [46] Heymans MW, de Vet HCW, Knol DL, et al. Workers’ beliefs and expectations affect return to work over 12 months. J Occup Rehabil 2006;16:685–95.
1857
[47] Werneke M, Hart DL. Centralization phenomenon as a prognostic factor for chronic low back pain and disability. Spine 2001;26: 758–64; discussion 765. [48] Linton SJ, Gross D, Schultz IZ, et al. Prognosis and the identification of workers risking disability: research issues and directions for future research. J Occup Rehabil 2005;15:459–74. [49] Weinstein JN, Lurie JD, Tosteson TD, et al. Surgical vs nonoperative treatment for lumbar disk herniation: the spine patient outcomes research trial (SPORT) observational cohort. JAMA 2006;296: 2451–9. [50] Osterman H, Seitsalo S, Karppinen J, Malmivaara A. Effectiveness of microdiscectomy for lumbar disc herniation: a randomized controlled trial with 2 years of follow-up. Spine 2006;31:2409–14. [51] Peul WC, van den Hout WB, Brand R, et al, Leiden-The Hague Spine Intervention Prognostic Study Group. Prolonged conservative care versus early surgery in patients with sciatica caused by lumbar disc herniation: two year results of a randomised controlled trial. BMJ 2008;336:1355–8. [52] Reiso H, Nygard JF, Jorgensen GS, et al. Back to work: predictors of return to work among patients with back disorders certified as sick: a two-year follow-up study. Spine 2003;28:1468–73; discussion 1473–4. [53] Kuijer W, Groothoff JW, Brouwer S, et al. Prediction of sickness absence in patients with chronic low back pain: a systematic review. J Occup Rehabil 2006;16:439–67. [54] Atlas SJ, Keller RB, Chang Y, et al. Surgical and nonsurgical management of sciatica secondary to a lumbar disc herniation: five-year outcomes from the Maine lumbar spine study. Spine 2001;26: 1179–87. [55] Pinto RZ, Maher CG, Ferreira ML, et al. Epidural corticosteroid injections in the management of sciatica: a systematic review and meta-analysis. Ann Intern Med 2012;157:865–77. [56] Krause N, Frank JW, Dasinger LK, et al. Determinants of duration of disability and return-to-work after work-related injury and illness: challenges for future research. Am J Ind Med 2001;40:464–84.