Background: Chronic back pain, defined as pain lasting longer than three months, is a common clinical problem that is often resistant to treatment. Contraindications or lack of response to nonsteroidal anti-inflammatory drugs sometimes lead physicians to prescribe opioids to relieve chronic back pain. Although opioid medications have been linked to drug-seeking behaviors and addiction, the strength of this association is uncertain. Martell and colleagues reviewed the prevalence and effectiveness of opioid use for chronic back pain and the risk of substance abuse disorders in these patients.

The Study: This systematic review and meta-analysis searched for relevant English language studies in multiple databases, reference lists from reviews, and recommendations from leading pain experts. Eligible studies included those in which researchers administered oral, topical, or transdermal opioids to nonpregnant adults with no history of opioid dependence. The initial literature search identified 1,626 abstracts; 38 studies met all of the inclusion criteria. Two reviewers independently assigned quality scores to each study using standardized criteria. Studies of opioid effectiveness that lasted less than three months or that were rated as less-than-excellent in quality were excluded from the meta-analysis.

Results: The prevalence of opioid prescribing for chronic back pain varied, with a higher proportion of patients receiving opioids in specialty treatment centers (range, 11 to 66 percent) than in primary care settings (range, 3 to 31 percent). Patients who reported greater disability and poorer function were more likel y to receive opioids than patients who reported more severe pain. In patients who were prescribed opioids, the prevalence of substance abuse disorders varied considerably across studies. Most studies were rated as poor quality because of their infrequent use of validated diagnostic instruments and control groups.

The effectiveness of opioids for back pain was evaluated in two groups of studies: those that compared different types of opioids and those that compared opioids with nonopioids or placebo. The longest period of follow-up was 16 weeks; most trials lasted 30 days or less. Meta-analysis demonstrated that reductions in pain scores with opioids were not statistically significant.

Conclusions: Although opioids are commonly prescribed for chronic back pain, the authors conclude that the available evidence does not support long-term benefits from their use. They also conclude that patients with chronic back pain who use opioids appear to have a higher prevalence of substance abuse. However, the authors caution that the low quality of evidence precludes drawing definite conclusions. For this reason, they call for more well-designed research studies to guide decisions about the use of opioids for this common chronic condition.