Prevention of relapsing backache

The condition of non-specific back pain is characterized by high prevalence, non satisfactory therapeutic options and severe socioeconomic consequences. Therefore prevention seems an attractive option to downsize the problem. However, the construction of effective preventive measures is complicated by the obscure aetiology of the condition, the multidimensionality of risk and prognostic factors (bio psychosocial model!) and the variability of its natural as well as clinical course. This led to the development of a wide variety of preventive measures: e. g. exercise programs, educational measures (including back school), ergonomic modification of the work environment, mechanical supports (e. g. back belts) as well as multidisciplinary interventions. For two reasons the workplace seems to be a suitable setting for prevention. First, because a number of strong risk factors are associated with working conditions and second, because it allows addressing a large proportion of the adult population. Against this background the assessment at hand sets out to answer the following questions: What is the amount and methodological quality of the available scientific literature on the effectiveness of back pain prevention in the workplace environment? What are effective measures for the prevention of back pain and its consequences in the workplace environment and how effective are they? Is back pain prevention in the workplace environment cost-effective? Is there a need for more research? As primary outcomes for effectiveness the assessment will focus on time lost from work and the frequency and duration of episodes with back pain. The preventive measures assessed belong to the following categories: exercise programs, educational and information measures, multidimensional interventions, back belts, lifting teams and ergonomic interventions.

The assessment is based on a systematic review of the published literature according to the methodological requirements of DAHTA. Proceedings of the electronic literature searches are documented in the appendix. In addition references of review articles were searched. Methodological quality of publications (systematic reviews, HTA reports) was assessed using the checklists developed by the German Scientific Working Group for Technology Assessment in Health Care (GSWGTAHC) or with the Jadad-Score (controlled trials) respectively. Due to the large number of relevant publications the assessment is mainly based on data reported by systematic reviews and supplemented by the results of newer trials. A separate economic assessment was not performed because of the low amount of available data. An assessment of ethical, legal and social impact was omitted due to resource constraints.

For preventive interventions based on exercise programs most of the analysed trials demonstrate some effectiveness. Due to the heterogeneity of the programs it is not possible to conclude whether positive effects are associated with a special type, duration or intensity of exercise. For purely educational measures or information strategies applied in a workplace setting the available trials were not able to demonstrate effectiveness. Back school programs, which in addition to theoretical instructions offer intensive exercising may in the short term, be successful in reducing the incidence of new episodes of back pain. Some trials in high risk groups demonstrate effectiveness of multidimensional interventions on time lost from work. These programs include education and exercise as well as cognitive behavioural interventions to change pain perception. The assessment of the benefits of back belts for the prevention of back pain is based on results of high quality efficacy as well as effectiveness trials. Their results imply for the otherwise healthy working population no protective effect of back belts on time lost from work due to back pain, on the incidence of painful episodes or on days with impairment by back pain. So far there are no data from controlled trials that demonstrate the effectiveness of "lifting teams" in nursing care to prevent back pain or its consequences. However, results from uncontrolled pilot studies indicate a potential for effectiveness. Among "ergonomic interventions" three different approaches have to be distinguished: interventions addressing changes of the workplace setting, interventions addressing the individual's behaviour and combined interventions. Studies evaluating the effectiveness of setting interventions (modification of the physical workplace environment, changes of production processes, organisational changes) yield no dependable results. This conclusion is not based on indifferent trial results but rather on the lack of methodologically sound studies. Results from studies on ergonomic interventions addressing the individual confirm the conclusions drawn for exercise and educational measures. The most marked results are found in trials that examine the effectiveness of combined interventions in high risk groups and contain a strong participatory component. Hardly any of the trials studying the effects of ergonomic interventions satisfied methodological quality criteria that are accepted standard for clinical or public health intervention studies. There were no data allowing firm conclusions on the cost-effectiveness of interventions from any of the categories.

The significance of the results of the assessment at hand is strongly limited by the comprehensiveness of the questions addressed. Reviewing the literature on the basis of (even systematic) review articles impairs the differentiated examination of the role of target groups, program contents, application and duration, effect sizes and context factors. While the methodological quality of the review articles is quite high, the quality of individual trials (even those included in the review papers) is highly variable. While most trials examining preventive interventions addressed at individuals satisfy at least some methodological requirements many studies dealing with setting interventions do not.

In conclusion, sound scientific evidence for the effectiveness and cost-effectiveness of back pain prevention in the workplace environment is still quite scarce. Further research should include: The development of interventions guided by the bio psychosocial model of back pain aetiology that combines individual prevention as well as measures addressing the workplace environment.The integration of results from basic ergonomic research into prevention concepts and the conduct of trials focussing outcomes with relevance to health.at the workplace setting. The conduct of qualitative studies to identify factors that impair the effectiveness of prevention programs (e. g. motivation, compliance, people skills).The integration of cost-effectiveness evaluations into all interventional studies.