Ergonomic changes made little difference to work-related pain

Clinical takeaway: Do not count on ergonomic changes alone to relieve pain or cut sick leave in workers with work-related arm, neck or shoulder complaints within three months. Pain appeared to ease slightly at longer follow-up. 

Work-related arm, neck and shoulder complaints account for about a third of worker compensation costs in the US. Ergonomic fixes, from adjusted workstations to specialized keyboards and posture training, play a major role in treating them. 

The 2013 Cochrane review of conservative treatments for these complaints found too little evidence to judge the widely used strategies. This update adds randomized trials published since then and tests ergonomic changes on their own. 

Ergonomic changes did not outperform no treatment, usual care or placebo for pain, function, sick leave or recovery up to three months after treatment. Sick leave is the firmest result: 14% of workers who implemented ergonomic changes took sick leave versus 16% of controls, a gap within chance across three trials. Pain scores ran slightly lower with ergonomic changes, but the range of plausible effects included no effect on pain. Function and recovery each rest on a single trial of about 380 workers and showed no clear difference. All of these short-term results come from trials of workers with non-specific complaints. 

Beyond three months after treatment, pain fell slightly more with ergonomic changes than with controls across eight trials of 873 workers, a small effect whose range of plausible values stayed clear of no effect. Up to three months after treatment, healthcare use split: one trial of 210 workers found about half as many sought medical help with ergonomic changes (18% versus 38% with usual care), and a second trial of 68 found no difference. 

Relative to exercise or other active treatment, short-term pain and function results were too uncertain to call, and pain at longer follow-up did not differ. No trial reported recurrence. 

The review included 20 trials of 2,478 workers, 13 of them office or computer workers. Ergonomic changes were compared with no treatment, usual care, placebo, exercise or other active treatment. All 20 trials carried high risk of bias, so the planned sensitivity analysis restricted to low-risk trials could not be run. 

The authors call for adequately powered trials with a low risk of bias, including concealed allocation and independent blinded assessors. Trial design has also been a bottleneck: the authors excluded trials in which it was unclear whether complaints were work-related or whether participants had pain or were only at risk for it. They also want a clear definition of work-related complaints of the arm, neck or shoulder. 

Poonam Mehta, lead author at the University of Technology Sydney, says: “Our findings suggest that, on current evidence, these measures alone shouldn’t be expected to reduce pain or sick leave in the short term.” 

Source: Mehta P, et al. (2026 Oct 27) Cochrane Database Syst Rev. Ergonomic interventions for treating work-related complaints of the arm, neck or shoulder in adults