A review of five studies found the prescribing of opioids to treat acute musculoskeletal disorders (MSDs) is linked to longer duration of time on disability. However, the review authors advise caution in drawing conclusions about cause and effect from these findings, given the studies’ methodological or design limitations.
The review, conducted at the Institute for Work & Health (IWH), set out to examine what the research to date says about the link between opioids prescribed soon after onset of an MSD and work outcomes such as length of work absence and return to work. It focused on opioid prescriptions written within 12 weeks of onset of a musculoskeletal injury or disorder—i.e. prescriptions likely meant to treat acute or subacute pain, rather than chronic MSDs.
The team found a link between early prescription of opioids and prolonged work disability in four of the five studies included in the review. “The review found a consistency across the studies suggesting opioids are associated with prolonged disability,” says Dr. Nancy Carnide, an IWH post-doctoral fellow who led the systematic review as part of her doctoral dissertation. “It’s certainly worth noting that we didn’t see positive work outcomes associated with opioids. But despite that consistency, we need to be cautious in drawing the conclusion that opioids cause disability.” The review was published in the July 2017 issue of the Clinical Journal of Pain (doi:10.1097/AJP.0000000000000452).
Five studies found
In light of the overwhelming evidence on the risks of opioid use, workers’ compensation agencies in many jurisdictions have taken steps to improve the prescribing of opioids among claimants with MSDs, says Carnide. To support optimal opioid prescribing management, the team set out to conduct a systematic review, to see what the existing research says about opioid prescriptions and work outcomes.
“We set out to do a systematic review that synthesizes only high quality evidence. In the end, that was not possible due to the high risk of bias in the studies we found,” says Carnide. “But given the importance of this issue, we ultimately decided to present the study findings, along with a more detailed discussion of the methodological limitations found in each of these studies.”
The team found only five studies in the research literature up to July 2014 that met review criteria. “We excluded studies that had no controls, such as case series studies. We also set out to specifically look for studies that attempted to demonstrate the opioids preceded the work outcome, so cross-sectional studies that provided a snapshot in time were not included,” says Carnide. None were randomized controlled studies, the gold standard in study design.
Most of the studies included in the review had been conducted using administrative data. Four were based on workers’ compensation claims in the U.S. and Canada for work-related back injuries; one was based on motor vehicle insurance claims in Australia. All studies focused on work disability outcomes derived from wage replacement benefit data, namely time on benefits or benefit status.
Four out of the five found workers with early opioid prescriptions were at higher risk of experiencing a longer disability duration. A fifth study focused on the length of time between prescriptions—an indicator for the authors of how closely the prescriptions were monitored. This latter study found prescriptions that were written in shorter intervals were linked to shorter disability duration; those written further apart were linked with longer disability duration.
Despite the general consistency in findings, the team found the included studies were at high risk of bias. The reliance on administrative data had some advantages (such as accurate recording of time on benefits as an outcome).
However, this data source also came with challenges. The review found uncertainty as to whether the studies accurately measured participants’ use of opioids. Participants may not have taken all the medication they were prescribed, or they may have had access to prescription opioids outside the insurance or compensation claim that was not recorded in the studies.
Another inherent challenge faced by researchers, particularly when using administrative data to investigate this topic, is what researchers call “confounding by indication,” says Carnide.
“A big challenge is how to tease apart the relationship between the underlying reasons for receiving the prescription—such as significant pain, poor function or distress—and the outcome. Is it the opioid itself causing disability, or is it the underlying reason leading to the opioid prescription that’s behind the prolonged disability?”
While a strictly causal interpretation of the findings in these studies is not yet warranted, “there is nothing in the literature that suggests opioids are linked with positive work outcomes,” she adds.
“In fact, there is little evidence to suggest they are effective even for pain and function in musculoskeletal disorders, such as back pain. Given what we know around the risks of opioid use, considerable caution is needed before considering opioids as treatment for workers with MSDs.”