NOHSAC Report

Executive Summary

This review provides an overview of the management and governance of occupational safety and health (OSH) in five countries: the United Kingdom, the United States of America, Finland, Canada and Australia.

The workplace is known to contribute to diseases and injuries, and many causes and contributing factors have already been identified. Further actual or potential factors will be identified through on-going research, such as the use of surveillance systems.

It is generally acknowledged there are both moral and practical dimensions to safety and health systems. Employees should not risk injury or damage to their health when working, nor should others be adversely affected by their working.

The most common principle is one of hazard identification using some form of risk assessment based on an agreed rule or “standard”. This approach rests on a sequence of assumptions. First is that risks and hazards are known and understood. Second is that they can be accurately identified in practice. Third is that, once they have been identified, they can be eliminated, or at least reduced, and this will yield a subsequent reduction in cases of injury or illness. However, this does not always hold true, nor does it necessarily take account of multifactorial and complex causation.

There is some congruence in the philosophy underlying the OSH systems reviewed. The most important and distinct aspect of this consensus is the reasonable assumption that identifying risk factors for occupational disease and injury is the foundation for effective prevention strategies.

The most obvious divergence between the various systems involves the lack of agreement about what to do in order to manage identified risks.

There is a uniform lack of consistency in the type and manner of application of sanctions applied (usually to employers) if safety and health rules are broken or not followed. There is a universal reliance on punitive sanctions, without attempts to apply positive rewards, for example after rapid remedial action has been taken.

It seems likely that an approach embracing a mixture of methods will be the more successful overall. This should probably incorporate a mix of:

None of the systems reviewed appear to be actively questioning the possible limitations of the hazard identification model, or the assumption that a high level of avoidance of all risk will have only beneficial effects. This question has social, political and moral implications.

Identifying hitherto unknown risks and hazards, or potential contributing factors, remains a major challenge. The lack of consistency within systems, over time, and between the various systems has seriously hindered the ability of researchers to collect and aggregate data in a meaningful way. Significant efforts are currently underway to harmonise data collection, which will provide an important boost to statistical power in the quest to detect the relevance of suspected factors and perhaps their subtle interplay with other factors.

The relationship between health and injury lacks clarity. This is largely due to the ad hoc manner in which cases are currently classified and attributed. For example, 39% of cases in the UK are classified as injuries, whereas in the US, 94% are called work-related injuries. Poor definitions, together with incentives and disincentives to classify one way or another, provide arbitrary divisions that defy meaningful interpretation.

It can be argued that OSH may be fruitfully linked with public health initiatives and strategies. This is because there are often overlapping areas of interest and similar applicable methodologies. However, in practice, this rarely occurs, for reasons that are not always clear. Anecdotal information from those inside the respective systems suggests it is not just a matter of simple territorial or boundary issues. Rather, it seems there is a general perception that health and injury issues that involve work either are better funded or have a specific tagged funding stream. Hence, there is a general reluctance on the part of those involved in public health matters to use any of their scarce resources in areas that may attract better funding. Furthermore, those involved in work-related health and injury issues are usually constrained by some form of mandate to remain firmly in an arena where work relatedness is demonstrable.

When safety and health systems fail to prevent injury or illness, as they inevitably will do since they cannot be perfect, some form of support and compensation is generally made available to the worker. This varies widely in many details, however there are three common methods for delivery: insurance-model workers’ compensation; social welfare/security benefits; and recourse to appeal and/or litigation.

The impact a compensation system might have on OSH initiatives is not entirely clear. The most common method to inform prevention strategies is to feed back claims’ history in an attempt to identify problem areas. However, the effectiveness of this lacks an evidence base. There is consistent anecdotal evidence from those involved in direct management of such systems that a frequent outcome is merely behaviour modification, such as reclassification or recoding of cases by general practitioners (GPs) or others, rather than a reduction in total claims. However, this may only hold true for less severe diseases or injuries. Feedback on work fatalities, for example, seems to have a more robust effect. That is, the relationship between systems and prevention initiatives may be complex and vary across spectra, such as severity.

The effectiveness of OSH initiatives is hard to quantify for a number of reasons. These include changes within systems, over time, and a lack of comparability between systems. An important and valid question is whether OSH systems function better when they are run as a stand-alone department or embedded within another agency. There is no clear evidence on this matter. However, observation indicates that independent, or at least semi-autonomous, departments may well function more effectively with greater focus and the ability to evolve more rapidly in response to changing needs.

This question leads naturally to a further important issue, namely, how good the available evaluation research is. Despite substantial and well intentioned efforts, the evidence base on the effectiveness of prevention strategies remains weak and equivocal. All the systems reviewed do place a strong emphasis on research, and this is perfectly appropriate and understandable. However, the best method for efficiently delivering research is not immediately clear. Many countries have given the task to a single, large research organisation. However, all seek external and independent research providers. Perhaps it can be argued that the most flexible and effective approach is to have a semi-autonomous research organisation that is required to drive a research agenda based on expert and stakeholder consultation, and that manages and coordinates a number of specialised groups that conduct the actual research. This should be augmented by overall independent evaluation of the research outputs.

The near future may be an exciting time for development within OSH systems, since there is a rapidly maturing approach based on more comprehensive data systems that are harmonised so as to allow more powerful comparisons, and there is a growing recognition that more sophisticated methods targeted at key areas identified by the stronger data sets will yield more effective prevention strategies.