• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of cmajCMAJ Information for AuthorsCMAJ Home Page
CMAJ. Jun 7, 2005; 172(12): 1559–1567.
PMCID: PMC558170

A clinical return-to-work rule for patients with back pain

Abstract

Background

Tools for early identification of workers with back pain who are at high risk of adverse occupational outcome would help concentrate clinical attention on the patients who need it most, while helping reduce unnecessary interventions (and costs) among the others. This study was conducted to develop and validate clinical rules to predict the 2-year work disability status of people consulting for nonspecific back pain in primary care settings.

Methods

This was a 2-year prospective cohort study conducted in 7 primary care settings in the Quebec City area. The study enrolled 1007 workers (participation, 68.4% of potential participants expected to be eligible) aged 18–64 years who consulted for nonspecific back pain associated with at least 1 day's absence from work. The majority (86%) completed 5 telephone interviews documenting a large array of variables. Clinical information was abstracted from the medical files. The outcome measure was “return to work in good health” at 2 years, a variable that combined patients' occupational status, functional limitations and recurrences of work absence. Predictive models of 2-year outcome were developed with a recursive partitioning approach on a 40% random sample of our study subjects, then validated on the rest.

Results

The best predictive model included 7 baseline variables (patient's recovery expectations, radiating pain, previous back surgery, pain intensity, frequent change of position because of back pain, irritability and bad temper, and difficulty sleeping) and was particularly efficient at identifying patients with no adverse occupational outcome (negative predictive value 78%– 94%).

Interpretation

A clinical prediction rule accurately identified a large proportion of workers with back pain consulting in a primary care setting who were at a low risk of an adverse occupational outcome.

Since the 1950s, back pain has taken on the proportions of a veritable epidemic, counting now among the 5 most frequent reasons for visits to physicians' offices in North America1,2,3 and ranking sixth among health problems generating the highest direct medical costs.4 Because of its high incidence and associated expense, effective intervention for back pain has great potential for improving population health and for freeing up extensive societal resources.

So-called red flags to identify pain that is specific (i.e., pain in the back originating from tumours, fractures, infections, cauda equina syndrome, visceral pain and systemic disease)5 account for about 3% of all cases of back pain.6 The overwhelming majority of back-pain problems are thus nonspecific. One important feature of nonspecific back pain among workers is that a small proportion of cases (< 10%) accounts for most of the costs (> 70%).7,8,9,10,11,12,13,14 This fact has led investigators to focus on the early identification of patients who are at higher risk of disability, so that specialized interventions can be provided earlier, whereas other patients can be expected to recover with conservative care.9,15,16,17,18,19,20,21,22,23,24,25 Although this goal has become much sought-after in back-pain research, most available studies in this area have 3 methodological problems:

  • Potential predictors are often limited to administrative or clinical data, whereas it is clear that back pain is a multidimensional health problem.
  • The outcome variable is most often a 1-point dichotomous measure of return to work, time off work or duration of compensation, although some authors have warned against the use of first return to work as a measure of recovery. Baldwin and colleagues,26 for instance, point out that first return to work is frequently followed by recurrences of work absence.
  • Most published prediction rules developed for back pain have not been successfully validated on any additional samples of patients.

Our study aimed to build a simple predictive tool that could be used by primary care physicians to identify workers with nonspecific back pain who are at higher risk of long-term adverse occupational outcomes, and then to validate this tool on a fresh sample of subjects.

Methods

From 1998 though 2002, we conducted a prospective cohort study called the RAMS-Prognosis Study (RAMS stands for recherche sur les affections musculo-squelettiques), which had a qualitative and a quantitative phase. The study was approved annually by the ethics committees of the participating hospitals.

In the first phase, 2 focus groups composed of people who have had severe back pain were held in November and December, 1998. The focus groups enriched the list of potential predictors (Appendix 1)10,27 to be investigated quantitatively in the interview-based second phase. The variables they identified were added to a list of those collected from the existing literature, to document as many potential predictors as possible.

Subjects for the second phase were recruited in 7 primary care settings of the Quebec City area: 4 emergency departments and 3 family medicine units, all having medical teaching responsibilities. Subjects of interest were adult workers aged 18–64 years who consulted for back pain, whatever its character or duration, between June 1999 and September 2000. Patients were eligible if the back pain was nonspecific and had caused them to be absent from their regular job for at least 1 day. Excluded were patients with any other condition that could affect their work capacity (pregnancy, for example, or a serious comorbidity) and those whose pain was located in the cervical spine only or had a specific cause.

Lists of potential subjects at each site were submitted weekly. A medical archivist telephoned these patients to request their participation in the study; those who agreed and were eligible were mailed an informed consent form to sign and return.

Participants were telephoned by trained interviewers for a baseline interview at around 3 weeks after their medical consultation, with repeated measurements at 6 and 12 weeks and 1 and 2 years. (Note that all timeframes are relative to each patient's index medical consultation.) The data collected described demographic, socioeconomic, behavioural, anthropometric, clinical, occupational and psychosocial variables, as well as information on the utilization of health services for back pain. Clinical information was drawn by a single individual from study participants' medical files. An online appendix (available at www.cmaj.ca/cgi/content/full/172/12/1559/DC1) summarizes the standardized instruments used to measure key constructs.28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60 When an instrument was unavailable in French, Vallerand's double-inverse translation method61 was applied.

Return to work in good health (RWGH) is an index of back- pain outcome that takes into account work status, functional limitations and number of days of work-absence since the medical consultation. This variable has 4 categories based on the work of Baldwin's group26 (presented in detail in Table 1): success, partial success, failure after attempt, and failure. Our work-absence data were self-reported; however, a validation study was conducted with data obtained from the employers of 40 volunteers. The type 1 intraclass coefficient correlating the 2 sets of data was 0.97 (95% confidence interval [CI] 0.93–0.98) for number of days of modified work, and 0.99 (95% CI 0.98–1.00) for number of days off any work.

Table thumbnail
Table 1

Subsequent to descriptive analysis, participants were randomly assigned (by means of a randomization function in SAS software) into 2 subgroups: about 40% into a “training sample” and the rest into a “validation sample.” Predictive analyses were conducted to identify, not causal relationships, but only prognostic indicators that could be clinically useful.62 Recursive partitioning with KnowledgeSeeker software (version 3.0, Angoss Software International Ltd., Toronto, Ont.) was used to build a predictive model of 2- year RWGH in the training sample.63 Because the distinction between the “failure after attempt” and “failure” groups has no clinical pertinence, data for these 2 categories of outcome were lumped together. The threshold for statistical significance was fixed first at 0.01, to favour the selection of the strongest predictors and simplicity in the model built. Afterward, it was relaxed to p ≤ 0.05 to permit us to identify complementary and alternative associations.

Classification error rate, sensitivity, specificity, and positive and negative predictive values (the probability, respectively, of having and not having an actual adverse occupational outcome when so predicted by the model) were computed for each model by comparing the predicted classification of subjects to their actual status at 2 years. In these analyses, the subjects were dichotomized along different groupings of the outcome categories. The “success” category was always considered among the “non-diseased” group. Area under the receiver operating characteristic curve64 was used as the main criterion to select one model over another, followed by the number of variables. Each selected model was then applied to the validation sample. We “pruned” the final model to try to keep it as simple as possible. All baseline, 6-week and 12-week variables were candidate predictors, including individual items or questions of specific measurement tools (Appendix 1).

Results

Numbers of eligible subjects, refusals and participants are shown in Fig. 1. The proportion of eligibility among subjects who were reached (37.6%) was applied to those not reached, to estimate a total number of eligible subjects. A total of 1007 subjects (68.4% of the 1471 expected to have been eligible) participated in the baseline interview, conducted an average of 25 days after the index consultation (standard deviation [SD] 10.2 d). Of those who participated in the baseline interview, 923 (91.7%) completed the interview at 6 weeks of follow-up, 907 (90.1%) at 12 weeks, 913 (90.7%) at 1 year, and 864 (85.8%) at 2 years. Three patients died during the follow-up period. Completed records of all 5 interviews were available for 860 participants, whose data were used for the recursive partitioning analysis.

figure 25FF1
Fig. 1: Eligible subjects, refusals and participants in the study. *Eligibility unknown. R = randomization.

The mean age of the participants was just under 39 years; a majority (58.5%) were male (Table 2). Almost half (47.8%) had earned a postsecondary diploma, from either a community college or a university.

Table thumbnail
Table 2

At baseline, a majority of subjects reported their back pain to be recurrent (“it comes and goes”) or persistent (“the pain is always there, to different degrees”); fewer than one-quarter reported that theirs was a one-time problem (“never had back pain before”). The median time since the beginning of patients' first episode was nevertheless 6 years. Pain was mostly situated in the lumbar and lumbosacral areas. Over half of our study subjects reported pain radiating to the arms or legs.

The 2-year evolution of RWGH among study subjects is illustrated i n Fig. 2. The most important changes in RWGH occurred at about 12 weeks, at which time about 50% of subjects were in the RWGH success category (compared with 18% at 6 weeks). At 2 years, close to 20% were still in the “failure after attempt(s)” and “failure” groups.

figure 25FF2
Fig. 2: Progress among study participants toward return to work in good health. Note that subjects could not go back into the “failure” group; this category could thus only diminish over time.

Fig. 3 presents the final predictive model as a clinical algorithm that allows estimation of the probability, for a given individual, of RWGH success, partial success and failure (which includes failure after attempt). The classification error rate was 37.0% in the training sample and 40.5% in the validation sample. Measures of validity to detect RWGH failure, partial success or both outcomes taken together for this model are presented for both samples in Table 3, with a set of example calculations shown in Fig. 4. All validity measures were quite stable when applied to the validation sample. In all cases, findings for negative predictive value were high (74%–91%). It was highest for predicting “failure after attempt/failure” (91%), whereas the highest positive predictive value was for detecting the combined outcome of “failure after attempt/failure” or “partial success” simultaneously (57%).

Table thumbnail
Table 3
figure 25FF4
Fig. 4: Example of calculations of the measures of validity presented in Table 3.
figure 25FF3
Fig. 3: Clinical algorithm to predict an outcome at 2 years of return to work in good health (RWGH) among workers consulting in primary care settings for back pain. All values shown are percentages. High-probability categories in each group, as were used ...

Interpretation

This study corroborates the complex nature of back pain and the inherent difficulty in developing clinical prediction tools for such conditions. With a large coverage of potential predictors measured at baseline and at 6 and 12 weeks, the best model we obtained contains 7 predictors measured at baseline (patient's recovery expectations, radiating pain, previous back surgery, intense pain, frequent change of position because of back pain, irritability and bad temper, and difficulty sleeping). It is far from perfect; nonetheless, its high negative predictive value may constitute a strong advantage.

Most previous studies on the prediction of the long-term outcome of back pain used a dichotomous measure of return to work, duration of work absence or compensation data. Because of methodological differences and the nature of predictive analyses (which are not intended to identify causal relationships, and should not be interpreted as doing so),64 it is not relevant to directly compare the predictors. In fact, 2 studies could end up with different sets of predictors; but the nature of the predictors is not very important, as long as they constitute an efficient and reproducible prognostic tool. The comparison must be made on the predictive validity of the models, that is, on their capacity to classify subjects correctly with respect to their outcome and on demonstration of the reproducibility of this validity. The few existing previously determined models explained some 25%–30% of the variance of continuous outcomes,25,65 similar to what we observed when we applied our final predictive model to long-term absence (data not shown). With regard to other measures of predictive validity, it is quite often negative predictive values that are highest,65,66,67 as in our study.

This study had several strengths: the inclusion of all workers consulting for back pain, whatever the source of pain (i.e., not only workers' compensation cases); the large sample; the prospective design, with repeated measures taken at key points in the natural history of the disease; a high participation rate; coverage of numerous variables that were considered; the use of a more specific and eventually more valid measure of occupational outcome; and the use of recursive partitioning.

Classically in prospective cohort studies, subjects must be free of the disease at the beginning of observation, so that only new (incident) cases are observed and the directionality of associations is non-equivocal. Since only about 1 in 5 back-pain patients consulting in primary care settings have never had back pain before, a so-called inception cohort study of incident cases would require great resources and include only subjects with homogeneous characteristics, which would limit its external validity. Primary care physicians meet with a heterogeneous population of back- pain patients (most with recurrent or persistent pain) for whom they must give a prognosis. A prediction tool that works for a subgroup of patients only is unlikely to be useful to clinicians, especially if the subgroup includes only a minority of their clients.

In this study, several measures of the type and severity of pain were used as potential predictors. If these variables had been important to the prognosis, they would have been retained. Our model is thus applicable to all sorts of back-pain problems, as seen in day-to-day primary care practice. It can be used systematically to assist the physician in deciding the best allocation of clinical resources for patients with back pain (see the example cases accompanying Fig. 3).

Reproducibility of predictors is an important consideration in building a prognostic instrument.68 The fact that 5 of the variables included in our final model are items drawn from well-validated measurement instruments is reassuring. The 2 other items, radiating pain and previous back surgery, are relatively “hard” events, measurement of which is likely to be highly reproducible.

Because the baseline interviews were conducted some 3 weeks after the related medical consultation, it is possible that some variables may have changed during that period, improving the predictive potential over data that would have been collected, had the interviews been held at the time of consultation. However, variables measured at 6 weeks and 12 weeks offered no better prediction than the baseline measures, which is reassuring.

Traditionally, physicians look for clinical decision rules that have high positive predictive value, and attribute generally less importance to negative predictive value. However, it is the nature of the outcome that must determine the most important measures of validity for a given clinical decision tool. Considering the frequency of back pain and the resources that are spent on benign cases, an instrument that allows identification of a group of subjects who are at low risk of adverse outcomes may be quite useful.

β See related article page 1575

Supplementary Material

[Online Appendix]

Acknowledgments

We thank all of the study participants, the staff of the Unité de recherche en santé des populations and all of the research assistants who worked on this study. Thanks also to physicians Stéphane Bergeron, Alexandra Dansereau, Georges Dufresne, Louis Larue, Natalie Le Sage, Jean Maziade and Jean Ouellet for their help with the recruitment of subjects. Special thanks to Arie Nouwen for his help with the questionnaire on self-efficacy, Julie Soucy for her participation in the coordination of the study and Eric Demers for his contribution to the statistical analyses.

An external file that holds a picture, illustration, etc.
Object name is 25TTUA1A.jpg

Appendix 1.

An external file that holds a picture, illustration, etc.
Object name is 25TTUA1B.jpg

Appendix 1. Continued.

Footnotes

This article has been peer reviewed.

Contributors: Clermont Dionne initiated the study. Clermont Dionne, Renée Bourbonnais, Pierre Frémont, Michel Rossignol and Susan Stock designed the study protocol. All of the authors discussed core ideas, participated in analysis and interpretation of the data, and contributed to the writing of the paper under Clermont Dionne's lead. Clermont Dionne and Isabelle Larocque coordinated the study and data collection. All of the authors are guarantors for the paper.

This study was supported by grant 97-061 from the Quebec Institute for Occupational Safety and Health (Institut de recherche Robert-Sauvé en santé et en sécurité du travail du Québec), which did not interfere in any way in the scientific and publication processes. Clermont Dionne and Renée Bourbonnais are Quebec Health Research Fund (Fonds de la recherche en santé du Québec) Scholars.

Competing interests: None declared.

Correspondence to: Dr. Clermont E. Dionne, Unité de recherche en santé des populations, Centre de recherche du Centre hospitalier affilié universitaire de Québec, Hôpital du Saint-Sacrement, 1050, chemin Ste-Foy, Québec QC G1S 4L8; fax 418 682-7949; ac.lavalu.pseru@ennoid.tnomrelc

References

1. Hart LG, Deyo RA, Cherkin DC. Physician office visits for low back pain: frequency, clinical evaluation, and treatment patterns from a US national survey. Spine 1995;20:11-9. [PubMed]
2. Cypress BK. Characteristics of physician visits for back symptoms: a national perspective. Am J Public Health 1983;73(4):389-95. [PMC free article] [PubMed]
3. Rapoport J, Jacobs P, Bell NR, Klarenbach S. Refining the measurement of the economic burden of chronic diseases in Canada. Chronic Dis Can Winter 2004;25(1):13-21. [PubMed]
4. Druss B. The most expensive medical conditions in America. Health Affairs 2002;21(4):105-11. [PubMed]
5. Bigos S, Bowyer O, Braen G, et al. Acute low back problems in adults. [Clinical Practice Guideline no 14.] Rockville (MD): Agency for Health Care Policy and Research, Public Health Service, US Department of Health and Human Services; 1994. Publication no 95-0642.
6. Deyo RA, Weinstein JN. Low back pain. N Engl J Med 2001;344(5):363-70. [PubMed]
7. Deyo RA, Cherkin D, Conrad D, Volinn E. Cost, controversy, crisis: low back pain and the health of the public. Annu Rev Public Health 1991;12:141-56. [PubMed]
8. Volinn E, Van Koevering D, Loeser JD. Back sprain in industry: the role of socioeconomic factors in chronicity. Spine 1991;16(5):542-8. [PubMed]
9. Gatchel RJ, Polatin PB, Mayer TG. The dominant role of psychosocial risk factors in the development of chronic low back pain disability. Spine 1995;20(24):2702-9. [PubMed]
10. Spitzer WO. Scientific approach to the assessment and management of activity-related spinal disorders: a monograph for clinicians. Report of the Quebec Task Force on Spinal Disorders. Spine 1987;12(7 Suppl):S1-59. [PubMed]
11. Abenhaim L, Rossignol M, Gobeille D, Bonvalot Y, Fines P, Scott S. The prognostic consequences in the making of the initial medical diagnosis of work-related back injuries. Spine 1995;20(7):791-5. [PubMed]
12. De Girolamo G. Epidemiology and social costs of low back pain and fibromyalgia. Clin J Pain 1991;7(Suppl 1):S1-7. [PubMed]
13. Spengler DM, Bigos SJ, Martin NA, Zeh J, Fisher L, Nachemson A. Back injuries in industry: a retrospective study. I. Overview and cost analysis. Spine 1986;11(3):241-5. [PubMed]
14. Hashemi L, Webster BS, Clancy EA, Volinn E. Length of disability and cost of workers' compensation low back pain claims. J Occup Environ Med 1997;39(10):937-45. [PubMed]
15. Butterfield PG, Spencer PS, Redmond N, Feldstein A, Perrin N. Low back pain: predictors of absenteeism, residual symptoms, functional impairment, and medical costs in Oregon workers' compensation recipients. Am J Ind Med 1998;34(6):559-67. [PubMed]
16. Van der Giezen AM, Bouter LM, Nijhuis FJ. Prediction of return-to-work of low back pain patients sicklisted for 3–4 months. Pain 2000;87(3):285-94. [PubMed]
17. Feuerstein M, Berkowitz SM, Huang GD. Predictors of occupational low back disability: implications for secondary prevention. J Occup Environ Med 1999;41(12):1024-31. [PubMed]
18. Boos N, Semmer N, Elfering A, Schade V, Gal I, Zanetti M, et al. Natural history of individuals with asymptomatic disc abnormalities in magnetic resonance imaging: predictors of low back pain–related medical consultation and work incapacity. Spine 2000;25(12):1484-92. [PubMed]
19. Viikari-Juntura E, Takala E, Riihimaki H, Martikainen R, Jappinen P. Predictive validity of symptoms and signs in the neck and shoulders. J Clin Epidemiol 2000;53(8):800-8. [PubMed]
20. Linton SJ, Hallden K. Can we screen for problematic back pain? A screening questionnaire for predicting outcome in acute and subacute back pain. Clin J Pain 1998;14(3):209-15. [PubMed]
21. Lancourt J, Kettelhut M. Predicting return to work for lower back pain patients receiving worker's compensation. Spine 1992;17(6):629-40. [PubMed]
22. Cats-Baril WL, Frymoyer JW. Identifying patients at risk of becoming disabled because of low-back pain: the Vermont Rehabilitation Engineering Center predictive model. Spine 1991;16(6):605-7. [PubMed]
23. Lehmann TR, Spratt KF, Lehmann KK. Predicting long-term disability in low back injured workers presenting to a spine consultant. Spine 1993;18: 1103-12. [PubMed]
24. Hogg-Johnson S, Cole DC. Early prognostic factors for duration on temporary total benefits in the first year among workers with compensated occupational soft tissue injuries. Occup Environ Med 2003;60(4):244-53. [PMC free article] [PubMed]
25. McIntosh G, Frank J, Hogg-Johnson S, Bombardier C, Hall H. Prognostic factors for time receiving workers' compensation benefits in a cohort of patients with low back pain. Spine 2000;25(2):147-57. [PubMed]
26. Baldwin ML, Johnson WG, Butler RJ. The error of using returns-to-work to measure the outcomes of health care. Am J Industr Med 1996;29:632-41. [PubMed]
27. Atlas SJ, Deyo RA, Patrick DL, Convery K, Keller RB, Singer DE. The Quebec Task Force classification for spinal disorders and the severity, treatment, and outcomes of sciatica and lumbar spinal stenosis. Spine 1996;21:2885-92. [PubMed]
28. Roland M, Morris R. A study of the natural history of back pain. Part I: development of a reliable and sensitive measure of disability in low-back pain. Spine 1983;8(2):141-4. [PubMed]
29. Waddell G. The back pain revolution. Edinburgh: Churchill Livingstone; 1998. p. 39-40.
30. Jensen MP, Strom SE, Turner JA, Romano JM. Validity of the Sickness Impact Profile Roland Scale as a measure of dysfunction in chronic pain patients. Pain 1992;50(2):157-62. [PubMed]
31. Beurskens AJ, de Vet HC, Koke AJ, van der Heijden GJ, Knipschild PG. Measuring the functional status of patients with low back pain: assessment of the quality of four disease-specific questionnaires. Spine 1995;20(9):1017-28. [PubMed]
32. Roland M, Fairbank J. The Roland–Morris Disability Questionnaire and the Oswestry Disability Questionnaire. Spine 2000;25(24):3115-24. [PubMed]
33. Riddle DL, Stratford PW, Binkley JM. Sensitivity to change of the Roland– Morris Back Pain Questionnaire: part 2. Phys Ther 1998;78(11):1197-207. [PubMed]
34. Bolton JE. Accuracy of recall of usual pain intensity in back pain patients. Pain 1999;83(3):533-9. [PubMed]
35. Strong J, Ashton R, Chant D. Pain intensity measurement in chronic low back pain. Clin J Pain 1991;7(3):209-18. [PubMed]
36. Waddell G, Newton M, Henderson I, Somerville D, Main C. 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. [PubMed]
37. Jacob T, Baras M, Zeev A, Epstein L. Low back pain: reliability of a set of pain measurement tools. Arch Phys Med Rehabil 2001;82(6):735-42. [PubMed]
38. Derogatis LR. Symptoms Checklist-90. Administration, scoring and procedures manual for the revised version. Baltimore: Clinical Psychometric Research; 1977.
39. Dionne CE, Bourbonnais R, Frémont P, et al. Predictors of return to work in good health among back pain patients in primary care settings: 3-month outcome in the RAMS-Prognosis Study [abstract]. International Forum V on Primary Care Research on Low Back Pain, 2002; Montreal.
40. Dozois DJA, Dobson KS, Wong M, Hughes D, Long A. Factors associated with rehabilitation outcome in patients with low back pain (LBP): prediction of employment outcome at 9-month follow-up. Rehabil Psychol 1995;40(4):243-59.
41. Severeijns R, Vlaeyen JW, van den Hout MA, Weber WE. Pain catastrophizing predicts pain intensity, disability, and psychological distress independent of the level of physical impairment. Clin J Pain 2001;17(2):165-72. [PubMed]
42. Nickel R, Egle UT, Eysel P, Rompe JD, Zollner J, Hoffmann SO. Health-related quality of life and somatization in patients with long-term low back pain: a prospective study with 109 patients. Spine 2001;26(20):2271-7. [PubMed]
43. Shutty MS Jr, DeGood DE, Schwartz DP. Psychological dimensions of distress in chronic pain patients: a factor analytic study of Symptom Checklist-90 responses. J Consult Clin Psychol 1986;54(6):836-42. [PubMed]
44. Karasek RA. Job Content Questionnaire and User's Guide. Los Angeles (CA): Department of Industrial and System Engineering, University of Southern California; 1985.
45. Brisson C, Blanchette C, Guimont C, Dionne G, Moisan J., Vézina M. Reliability and validity of the French version of the 18-item Karasek Job Content Questionnaire. Work Stress 1998;12(4):322-36.
46. Bigos SJ, Battie MC, Spengler DM, Fisher LD, Fordyce WE, Hansson TH, et al. A prospective study of work perceptions and psychosocial factors affecting the report of back injury [published erratum appears in Spine 1991;16(6):688]. Spine 1991;16(1):1-6. [PubMed]
47. Friedrich M, Cermak T, Heiller I. Spinal troubles in sewage workers: epidemiological data and work disability due to low back pain. Int Arch Occup Environ Health 2000;73(4):245-54. [PubMed]
48. Reigo T, Tropp H, Timpka T. Absence of back disorders in adults and work-related predictive factors in a 5-year perspective. Eur Spine J 2001;10(3):215-20; discussion 221. [PMC free article] [PubMed]
49. Williams DA, Feuerstein M, Durbin D, Pezzullo J. Health care and indemnity costs across the natural history of disability in occupational low back pain. Spine 1998;23(21):2329-36. [PubMed]
50. Williams RA, Pruitt SD, Doctor JN, Epping-Jordan JE, Wahlgren DR, Grant I, et al. The contribution of job satisfaction to the transition from acute to chronic low back pain. Arch Phys Med Rehabil 1998;79(4):366-74. [PubMed]
51. Loisel P, Durand P, Gosselin L, Simard R, Turcotte J, editors. La clinique des maux de dos — un modèle de prise en charge en prévention de la chronicité. Québec: Institut de recherche en santé et en sécurité du travail du Québec; mai 1996.
52. Siegrist J, Starke D, Chandola T, Godin I, Marmot M, Niedhammer I, et al. The measurement of effort–reward imbalance at work: European comparisons. Soc Sci Med 2004;58(8):1483-99. [PubMed]
53. Niedhammer I, Siegrist J, Landre MF, Goldberg M, Leclerc A. [Psychometric properties of the French version of the Effort–Reward Imbalance model]. Rev Epidemiol Sante Publique 2000;48(5):419-37. [PubMed]
54. Cherkin D, Deyo RA, Berg AO, Bergman JJ, Lishner DM. Evaluation of a physician education intervention to improve primary care for low-back pain. I. Impact on physicians. Spine 1991;16(10):1168-72. [PubMed]
55. Von Korff M, Barlow W, Cherkin D, Deyo RA. Effects of practice style in managing back pain. Ann Intern Med 1994;121(3):187-95. [PubMed]
56. Rosenstiel AK, Keefe FJ. The use of coping strategies in chronic low back pain patients: relationship to patient characteristics and current adjustment. Pain 1983;17:33-44. [PubMed]
57. Cedraschi C, Nordin M, Nachemson AL, Vischer TL. Health care providers should use a common language in relation to low back pain patients. Baillieres Clin Rheumatol 1998;12:1-15. [PubMed]
58. Bandura A. Self-efficacy: toward a unifying theory of behavioral change. Psychol Rev 1977;84(2):191-215. [PubMed]
59. Wallston KA, Wallston BS, DeVellis R. Development of the Multidimensional Health Locus of Control (MHLC) Scales. Health Educ Monogr 1978;6(2):160-70. [PubMed]
60. Talbot F, Nouwen A, Gauthier J. Is health locus of control a 3-factor or a 2-factor construct? J Clin Psychol 1996;52(5):559-68. [PubMed]
61. Vallerand RJ. Vers une méthode de validation trans-culturelle des questionnaires psychologiques: implications pour la recherche en langue française. Can Psychol 1989;30:662.
62. Kleinbaum DG, Kupper LL, Muller KE. Applied regression analysis and other multivariable methods. 2nd ed. Boston: PWS–Kent Publishing Company; 1988. p. 386-482.
63. Breiman L, Friedman JH, Olshen RA, Stone CJ. Classification and regression trees. Pacific Grove (CA): Wadsworth & Brooks; 1984. 368 p.
64. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982;143:29-36. [PubMed]
65. Dionne CE, Koepsell TD, Von Korff M, Deyo RA, Barlow WE, Checkoway H. Predicting long-term functional limitations among back pain patients in primary care settings. J Clin Epidemiol 1997;50:31-43. [PubMed]
66. McIntosh G, Frank J, Hogg-Johnson S, Hall H, Bombardier C. Low back pain prognosis: structured review of the literature. J Occup Rehabil 2000;10(2):101-15.
67. Thomas E, Silman AJ, Croft PR, Papageorgiou AC, Jayson MI, Macfarlane GJ. Predicting who develops chronic low back pain in primary care: a prospective study. BMJ 1999;318(7199):1662-7. [PMC free article] [PubMed]
68. Laupacis A, Sekar N, Stiell IG. Clinical prediction rules. A review and suggested modifications of methodological standards. JAMA 1997;277(6):488-94. [PubMed]

Articles from CMAJ : Canadian Medical Association Journal are provided here courtesy of Canadian Medical Association
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • MedGen
    MedGen
    Related information in MedGen
  • PubMed
    PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...