Back pain. Shoulder pain. Elbow pain. These types of problems are often sustained by workers on the job, and can account for many days off work, years of reduced income and high workers’ compensation costs. The burden of these musculoskeletal disorders (MSDs) is significant for workers, employees and society at large.
What do we know about what works in reducing this burden? Dr. Barbara Silverstein tackled this question during the Institute for Work & Health’s (IWH’s) 2006 Alf Nachemson Lecture.
For years, researchers have studied interventions designed to reduce the burden of workplace MSDs. Silverstein – who is Research Director of the Safety & Health Assessment and Research for Prevention (SHARP) Program with the Washington State Department of Labor and Industries – described a variety of these studies.
Her descriptions highlighted the challenges of researching in workplaces, of dealing with uncertain findings, and of translating evidence into practice, policy and regulations.
Research and response
Clinical guidelines say that bed rest and inactivity rarely relieve low-back pain, says Silverstein. Yet what do practitioners actually advise their patients?
A survey published in the Journal of Environmental Medicine in 2006 shows that emergency room physicians tend to advise inactivity. Occupational medicine physicians, in line with the guidelines, were more likely to advise activity.
How many more emergency room doctors do you think there are than occupational medicine doctors?
asks Silverstein. The longer that the physician has been in practice, irrespective of discipline, the less likely they are going to advise activity.
So we know what works with regards to the clinical management of low-back pain, but these findings raise questions about putting knowledge into practice. In this regard, the IWH is leading the way in how we transfer knowledge and have an exchange about what really works and doesn’t,
she says.
Compared to a clinical setting, a workplace poses additional challenges in studying how interventions lead to change. Early in her studies, Silverstein says, I believed that intervention studies could be done purely, particularly in workplaces. If you were able to see an exposure [to a task that increased the risk of developing an MSD], and you decreased the exposure, you would decrease the effect. If you increased the exposure you would increase the effect, and nothing else explained the change.
The reality is that besides the task, there are other factors, such as the workplace environment, which also affect change. In addition, there are basic issues inherent to intervention studies that pose a challenge in finding what works.
How many workplaces are excited about having us come in and do research? Are they representative of those who do not want to be bothered?
asks Silverstein. Another problem is that many workplace interventions involve so few workers that they don’t have the strength to detect when a change, as a result of an intervention, is significant.
Workplaces are dynamic, forever changing and they change irrespective of our study design or the intervention that we want to see implemented or evaluate,
says Silverstein.
Recently, the U.S. media has pronounced that ergonomics doesn’t work in office settings, based on the findings of a systematic review in the Cochrane Collaboration Database. The review found limited evidence for exercise therapy, for rest breaks, for or against changes to keyboard design, and conflicting evidence about ergonomics programs.
The conclusion in press was that the benefit of expensive ergonomic interventions in the workplace is not clearly demonstrated,
she says.
But if you look more closely at the studies in the review, Silverstein noted a number of problems. Among these were the fact that the studies didn’t always identify if the health conditions were related to work – so an intervention at work might not have been a solution anyway. Many of the studies were short-term, poorly designed and did not have enough power to show a difference.
The findings were not, as pronounced by the media, definite proof that ergonomics doesn’t work.
A series of randomized controlled trials between 1999 and 2005 have shown that some interventions do have an effect. Among the findings:
- participatory ergonomics and workstation modifications reduce symptoms in younger workers, but not in older ones
- intensive ergonomics training decreases symptoms in the short term only
- both ergonomics training and an adjustable chair reduced symptoms in office workers in the short term and increased productivity
- workstation changes and breaks prompted by software did not reduce symptoms, according to various studies
- a 2005 systematic review from the Institute for Work & Health showed that almost all participatory ergonomics studies of medium- or high-quality reported some evidence of positive effects on reducing injuries and workers’ compensation claims, and lost workdays
When a large, well-designed study doesn’t seem to show results, it pays to look more closely at what’s happening, Silverstein says. A multi-site Norweigan study of participatory ergonomics in thousands of workers and supervisors failed to show any improvements in MSD symptoms.
At the start of the study, 94 per cent of participants had symptoms. The changes that were implemented included job redesign, increased job variety and reduced repetition. But these did not reduce symptoms in the three groups being studied.
However, during the study, there was restructuring in the mills involved, and employees changed locations, creating confusion in study groups. Another important point was that the follow-up survey occurred shortly after the intervention, so there was little time for the symptoms to improve.
It is incredibly important to document both planned and unplanned changes to understand what’s actually happening,
says Silverstein.
Regulating MSD prevention
When practices to prevent MSDs are mandated by legislation, another layer of complexity is added, as experiences in Washington State have shown.
We know that manual patient or resident handling in hospitals and nursing homes increases the risk of MSDs, primarily in the back, but also in the shoulder,
says Silverstein. A systematic review and several studies show that lift equipment and safe patient handling prevent MSDs.
Washington state legislature passed safe patient handling legislation, but the law only applies to hospitals. The incidence rate for MSDs is actually highest in nursing homes, which do not fall under the legislation, Silverstein points out.
Politically, the state has four nurse legislators and all hospitals are unionized, giving them a strong voice, while very few nursing homes have unionized staff, she notes. The legislation would be important in improving back injury rates in nursing homes.
However, on the positive side, the act did make Washington the first state to have mandatory legislation on safe patient handling, and hospitals and unions supported the move.
I would say that so far it’s successful,
says Silverstein. My hope is that nursing homes will be next. Did they need the research in the legislature? I think we informed [the process], but it’s naïve to think that researchers are dominant influences on the legislative process.
Washington state also had ergonomics regulations – called the Ergonomics Rule – but this legislation was repealed in 2003. The rule did appear to have an impact on reducing MSDs, according to surveys of thousands of employers conducted by SHARP researchers before, during and after the law was struck down.
After the rule passed in 2001, requirements for companies were phased in over seven years. Industries were grouped based on their size and type.
The first group consisted of workplaces with more than 50 employees in industries such as trucking, nursing and carpentry. By 2003, this group had to identify hazards and educate workers. By 2005, these workplaces would have had to reduce exposure to these hazards to an acceptable level.
SHARP researchers conducted the workplace survey at four different points in time. The first survey was in 1998, and second took place in 2001 and the third survey was done in mid-2003, when the first group had identified hazards. Responses ranged from 5,900 to 7,500 employers for each survey.
The survey asked employers about MSD rates, number of employees exposed to different risk factors, if any prevention measures were taken, and other questions.
Between 2001 and 2003, there was a decrease in the percentage of employees exposed to hazardous tasks, such as awkward lifting, lifting above the shoulder and others. In addition, the percentage of workplaces taking steps to reduce the risk of MSDs increased.
The reduction in exposure between 2001 and 2003 suggests there was some impact of there being a rule,
says Silverstein. However, in November 2003 the rule was revoked. There was extended controversy about the rule over the phase-in period.
The 2005 survey revealed an increase in the percentage of employees who were exposed to MSD hazards. At this point, says Silverstein, there was no rule and no possibility of workplace inspection related to ergonomics issues.
The costs of the repeal are high, according to estimates from SHARP researchers. The economic effect of not having the ergonomics rule fully in place is about $182 million per year,
says Silverstein.
Where do we go from here?
Despite the challenges in finding what works to reduce MSDs – and in implementing these changes – past experiences have led to progress.
I think we have had some improvements in designing MSD intervention studies,
says Silverstein. However, we will never have control over workplaces and we need to take that into account. All study designs that take place in dynamic workplaces face the possibility of being overwhelmed by external circumstances, such as plants shutting down or being restructured.
Researchers need to have contingency plans and to document everything that happens, she says. They also need to use information beyond randomized controlled trials, such as case studies and natural experiments occurring in companies.
Researchers need to hook up with these workplaces to try to get better designs, where possible. This is the only way to get enough studies,
she says. I think where we can do the randomization, we absolutely need to do it. But I don’t think we can afford to let the ‘perfect’ be the enemy of the ‘good’.
The Alf Nachemson Lectureship was established in 2002 to honour Dr. Nachemson’s significant contribution to research evidence in clinical decision-making. Nachemson is a founding member of the IWH Scientific Advisory Committee.