METHODS

Guidelines on the occupational health management of LBP were retrieved from personal files of the authors. Retrieval was checked by a Medline search using the keywords “low back pain”, “guidelines”, and “occupational” up to October 2001, and personal communication with experts in the field. Guidelines had to meet the following inclusion criteria:

• Guidelines aimed at the management of workers with LBP (in occupational health care settings or addressing occupational issues), or separate sections of guidelines that dealt with these topics.

• Guidelines available in English or Dutch (or translated into these languages).

The exclusion criteria were:

• Guidelines on primary prevention (that is, prevention before the onset of the symptoms) of work related LBP (for example, lifting instructions for workers).

• Clinical guidelines for the management of LBP in primary care.²

The quality of the included guidelines was appraised using the AGREE instrument, which is a generic tool designed primarily to help guideline developers and users assess the methodological quality of clinical practice guidelines.³

The AGREE instrument provides a framework for the assessment of quality on 24 items (table 1), each rated on a four point scale. The full operationalisation is available on www.agreecollaboration.org.

Two reviewers (BS and HH) independently rated the quality of the guidelines, and then met to discuss disagreements and to reach consensus on the ratings. When they could not reach consensus, a third reviewer (MvT) reconciled remaining differences and made a final decision on the ratings. To facilitate analysis in this review, ratings were transformed into dichotomous variables of whether each quality item was or was not met.

The selected guidelines were further characterised and compared regarding the guideline committee, the presentation of the guideline, the target group, and the extent to which the recommendations were based on available scientific evidence. The assessment recommendations were also summarised and compared, as were recommendations on advice, treatment, and return to work strategies. All of this information was extracted directly from the published guidelines.

RESULTS

Selection of studies

Our search found 10 guidelines,^4–18 but four were excluded because they dealt with the management of LBP in primary care,¹⁵ were aimed at the guidance of sick listed employees in general (not specifically LBP),¹⁶ were intended for the primary prevention of LBP at work,¹⁷ or were not available in English or Dutch.¹⁸ The final selection therefore consisted of the following six guidelines, listed by date of issue:

1. Canada (Quebec). 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. Quebec Canada (1987).⁴

2. Australia (Victoria). Guidelines for the management of employees with compensable low back pain. Victorian WorkCover Authority, Australia (1996).⁵ (This guideline is a revised version of guidelines developed by the South Australian WorkCover Corporation in October 1993.)

3. USA. Occupational Medicine Practice Guidelines. American College of Occupational and Environmental Medicine. USA (1997).⁶

4. New Zealand

   1. Active and working! Managing acute low back pain in the workplace. Accident Compensation Corporation and National Health Committee. New Zealand (2000).⁷

   2. Patient guide to acute low back pain management. Accident Compensation Corporation and National Health Committee. New Zealand (1998).⁸

   3. Guide to assessing psychosocial yellow flags in acute low back pain. Accident Compensation Corporation and National Health Committee. New Zealand (1997).⁹

5. Netherlands. Dutch guideline for the management of occupational physicians of employees with low back pain. Dutch Association of Occupational Medicine (NVAB). Netherlands (1999).¹⁰

6. UK

   1. Occupational health guidelines for the management of low back pain at work—principal recommendations. Faculty of Occupational Medicine. UK (2000).¹¹

   2. Occupational health guidelines for the management of low back pain at work—leaflet for practitioners. Faculty of Occupational Medicine. UK (2000).¹²

   3. Occupational health guidelines for the management of low back pain at work—evidence review. Faculty of Occupational Medicine. UK (2000).¹³

   4. The Back Book, The Stationery Office. UK (1996).¹⁴

Two guidelines (4 and 6) could not be evaluated independently from additional documents to which they refer (4b–c, 6b–d) so these documents were also included in the review.

Appraisal of the quality of the guidelines

Initially, there was agreement between the two reviewers regarding 106 (77%) of the 138 item ratings. After two meetings, consensus was reached for all but four items, which required adjudication by the third reviewer. Table 1 presents the final ratings.

All included guidelines clearly presented the different options for the management of LBP in occupational health. In five of the six guidelines the overall objectives of the guideline were described specifically,^{4–6,10–14} the target users of the guideline were clearly defined,^5–14 easily identifiable key recommendations were included,^4,6–14 or key review criteria were presented for monitoring and/or audit purposes.^{4–9,11–14}

The results of the AGREE appraisal showed that none of the guidelines paid sufficient attention to potential organisational barriers and cost implications in implementing the recommendations. It was also unclear for all included guidelines whether or not they were editorially independent from the funding body, and whether or not there were conflicts of interest for the members of the guideline development committees. Furthermore, it was unclear for all guidelines whether experts had externally reviewed the guidelines prior to publication. Only the UK guideline clearly described the method used for the formulation of the recommendations, and provided for updating the guideline.¹¹

Development of the guidelines

Table 2 presents background information on the development process of the guidelines.

The target users for the guidelines were physicians and other health care providers in the field of occupational health care. Several guidelines were also directed at informing employers, workers,^6–8,11,14 or members of organisations interested in occupational health.⁴ The Dutch guideline was only targeted at the occupational health physician.¹⁰

The guideline committees responsible for the development of the guidelines were generally multidisciplinary, including disciplines like epidemiology, ergonomics, physiotherapy, general practice, occupational medicine, occupational therapy, orthopaedics, and representatives of employers’ associations and trade unions. Chiropractic and osteopathic representatives were in the guideline committee of the New Zealand guidelines.^7–9 The Quebec task force (Canada) also included representatives of rehabilitation medicine, rheumatology, health economics, law, neurosurgery, biomechanical engineering, and library sciences. In contrast, the guideline committee of the Dutch guideline consisted only of occupational physicians.¹⁰

The guidelines were issued as a separate document,^4,5,10 as a chapter in a textbook,⁶ or as several interrelated documents.^{7–9,11–14}

The UK,¹³ USA,⁶ and Canadian⁴ guidelines provided information on the search strategy applied to the identification of relevant literature and the weighing of the evidence. On the other hand, the Dutch¹⁰ and the Australian⁵ guidelines supported their recommendations only by references. In the New Zealand guidelines there were no direct links between recommendations and references,^7–9 and the reader was referred to other literature for background information.

Patient population and diagnostic recommendations

Despite the fact that all guidelines focused on workers with LBP, it was often not clear whether they dealt with acute or chronic LBP or both. Acute and chronic LBP were often not defined, and when cut off points were given (for example, <3 months) it was usually not clear whether these referred to the onset of symptoms or to absence from work. However, the Canadian guideline introduced a classification system (acute/subacute/chronic) based on the distribution of claims of spinal disorders by time since absence from work.⁴

All guidelines distinguished specific and non-specific LBP. Specific LBP concerns the potentially serious “red flag” conditions like fractures, tumours, or infections, and the Dutch and UK guidelines also distinguished the radicular syndrome or nerve root pain.^10–13 All guidelines were consistent in their recommendations to take a clinical history and to carry out a physical examination including neurological screening. In cases of suspected specific pathology (“red flags”), x ray examinations were recommended by most guidelines. In addition, the New Zealand and the US guideline also recommended x ray examination when symptoms did not improve after four weeks.^6,9 The UK guideline stated that x ray examinations are not indicated and do not assist occupational health management of the patient with LBP (as distinct from any clinical indications).^11–13

Most of the guidelines considered psychosocial factors—“yellow flags”—as obstacles to recovery that should be addressed by health care providers. The New Zealand⁹ and UK guideline^11,12 explicitly listed factors and suggested questions in order to identify those psychosocial “yellow flags”.

All guidelines addressed the importance of the clinical history identifying physical and psychosocial workplace factors relevant to LBP, including physical demands of work (manual handling, lifting, bending, twisting, and exposure to whole body vibration), accidents or injuries, and perceived difficulties in returning to work or relationships at work. The Dutch and the Canadian guidelines contained recommendations to carry out a workplace investigation¹⁰ or an assessment of occupational skills when necessary.⁴

Recommendations regarding information and advice, treatment, and return to work strategies

Most of the guidelines recommended reassuring the employee and providing information about the self limiting nature and good prognosis of LBP. Encouragement of return to ordinary activity as normally as possible was frequently advised.

In line with the recommendation to return to normal activity, all guidelines also stressed the importance of returning to work as rapidly as possible, even if there is still some LBP and if necessary starting with modified duties in more severe cases. Work duties could then be increased gradually (hours and/or tasks), until full return to work was reached. The US and Dutch guidelines provided explicit time schedules for return to work. The Dutch guideline proposed return to work within two weeks with adaptation of duties when necessary.¹⁰ The Dutch guideline also stressed the importance of time contingent management with regard to return to work.¹⁰ The US guideline proposed every attempt to maintain the patient at maximal levels of activity, including work activities; targets for disability duration in terms of return to work were given as 0–2 days with modified duties, and 7–14 days if modified duties are not used/available.⁶ In contrast to the others, the Canadian guideline advised return to work only when symptoms and functional restrictions had improved.⁴

In general, the most frequently recommended treatment options in all the included guidelines were: medication for pain relief,^5,7,8 gradually progressive exercise programmes,^6,10 and multidisciplinary rehabilitation.^10–13 The US guideline recommended referral within two weeks to an exercise programme consisting of aerobic exercises, conditioning exercises for trunk muscles, and exercise quota.⁶ The Dutch guideline recommended that if there is no progress within two weeks of work absence, workers should be referred to a graded activity programme (gradually increasing exercises) and if no progress by four weeks, then to a multidisciplinary rehabilitation programme.¹⁰ The UK guideline recommended that workers who have difficulty returning to normal occupational duties by 4–12 weeks, should be referred to an active rehabilitation programme. This rehabilitation programme should include education, reassurance and advice, a progressive active exercise and fitness programme, and pain management according to behavioural principles; it should be embedded in an occupational setting and directed strongly towards return to work.^11–13 Extensive lists of possible treatment options were presented in the guidelines of Canada and Australia,^4,5 although most of these were not based on scientific evidence.

DISCUSSION

The management of LBP in an occupational health setting must address the relation between low back complaints and work, and develop strategies aimed at a “safe” return to work. This review compared available occupational health guidelines from various countries. Guidelines are rarely indexed in Medline, so when searching for guidelines we had to rely primarily on personal files and personal communication.

Quality aspects and development process of the guidelines

The assessment by the AGREE instrument³ showed some differences in the quality of the guidelines reviewed, which may partly reflect the variation in the dates of development and publication of the guidelines. The Canadian guideline, for example, was published in 1987 and the Australian guideline in 1996.^4,5 The other guidelines were more recent and incorporated a more extensive evidence base and more up to date guideline methodology.

Several common flaws related to the development process of the guidelines were shown by the assessment by the AGREE instrument. Firstly, it is important to make clear whether a guideline is editorially independent from the funding body, and whether there are conflicts of interest for the members of the guideline committee. None of the included guidelines clearly reported these issues. Further, reported external review of the guideline by clinical and methodological experts prior to publication was also lacking in all guidelines included in this review.

Several guidelines provided comprehensive information on the way relevant literature was searched and translated into recommendations.^4,6,11,13 Other guidelines supported their recommendations by references,^5,7,9,10 but this does not permit assessment of the robustness of the guidelines or their recommendations.

Guidelines depend on the scientific evidence, which changes over time, and it is striking that only one guideline provided for future update.^11,12 Possibly there are updates planned for the other guidelines but they are not explicitly stated (and conversely stating there will be future update does not mean it will actually occur). This lack of reporting may also hold true for other AGREE criteria that we rated negatively. The use of the AGREE framework as a guide for both the development and the reporting of guidelines should help to improve the quality of future guidelines.

Assessment and management of LBP

The diagnostic procedures recommended in the occupational health guidelines were largely similar to the recommendations of clinical guidelines,² and, logically, the main difference was the emphasis on addressing occupational issues. The reported methods for addressing workplace factors in the assessment of LBP of the individual worker concerned the identification of difficult tasks, risk factors, and obstacles for return to work by occupational histories. Obviously, these obstacles for return to work not only concern physical load factors, but also work related psychosocial problems regarding responsibilities, cooperation with co-workers, and the social atmosphere at the workplace.¹⁰ Screening for work related psychosocial “yellow flags” may help to identify those workers who are at risk for chronic pain and disability.^11–13

A potentially important feature of the guidelines is that they were consistent regarding their recommendations to reassure the employee with LBP, and to encourage and support return to work even with some persisting symptoms. There is general consensus that most workers do not have to wait until they are completely free of pain before returning to work. The lists of treatment options provided by the Canadian and Australian guidelines may reflect the lack of evidence at that time,^4,5 leaving users of the guidelines to choose for themselves. It is, however, questionable whether such lists really contribute to improved care, and in our view guideline recommendations should be based on sound scientific evidence.

The US, Dutch, and UK occupational guidelines^6,10–13 recommend that active multidisciplinary treatment is the most promising intervention for return to work, and this is supported by strong evidence from RCTs.^19,20 However, more research is still needed to identify the optimum content and intensity of those treatment packages.^13,21

Despite some evidence for a contribution of workplace factors in the aetiology of LBP,²² systematic approaches for workplace adaptations are lacking, and are not offered as recommendations in the guidelines. Perhaps this represents a lack of confidence in the evidence on the overall impact of workplace factors, a difficulty of translation into practical guidance, or because these issues are confounded with local legislation (which was hinted at in the UK guideline¹¹). It may be that the “participatory ergonomics” intervention, which proposes consultations with the worker, the employer, and an ergonomist, will turn out to be a useful return to work intervention.^23,24 The potential value of “getting all the players onside”²⁵ was stressed in the Dutch and the UK guidelines,^11–13 but further evaluation of this approach and its implementation is required.

Development of future guidelines in occupational health care

The purpose of this review was to give both an overview and a critical appraisal of occupational guidelines for the management of LBP. The critical appraisal of the guidelines is meant to help direct future development and planned updates of guidelines. In the still emerging field of guideline methodology we consider all past initiatives as laudable; we recognise the need for clinical guidance, and appreciate that guidelines developers cannot wait for research to provide all the methodology and evidence required. However, there is room for improvement and future guidelines and updates should consider the criteria for proper development, implementation, and evaluation of guidelines as suggested by the AGREE collaboration.

The implementation of the guidelines is beyond the scope of this review, but it was noted that none of the guideline documents specifically described implementation strategies, so it is uncertain to what extent the target groups may have been reached, and what effects that may have had. This may be a fruitful area for further research.

The very existence of these occupational health guidelines shows that existing primary care clinical guidelines for LBP² are considered inappropriate or insufficient for occupational health care. There is a clear perception internationally that the needs of the worker experiencing back pain are intrinsically linked to a variety of occupational issues not covered by usual primary care guidance and, consequently, practice. What emerges is that, despite the methodological flaws, considerable agreement is evident on a range of fundamental occupational health strategies for managing the worker with back pain, some of which are innovative and challenge previously held views. There is agreement on the fundamental message that prolonged work loss is detrimental, and that early work return should be encouraged and facilitated; there is no need to wait for complete symptom resolution. Although the recommended strategies vary somewhat, there is considerable agreement on the value of positive reassurance and advice, availability of (temporary) modified work, addressing workplace factors (“getting all the players onside”), and rehabilitation for workers having difficulty returning to work.