Objectives:

Chronic pain is common, and opioid therapy is frequently prescribed for this condition. This report updates and expands on a prior Comparative Effectiveness Review on long-term (≥1 year) effectiveness and harms of opioid therapy for chronic pain, including evidence on shorter term (1 to 12 months) outcomes.

Data sources:

A prior systematic review (searches through January 2014), electronic databases (Ovid MEDLINE^®, Embase^®, PsycINFO^®, Cochrane CENTRAL, and Cochrane Database of Systematic Reviews through August 2019), reference lists, and clinical trials registries.

Review methods:

Predefined criteria were used to select studies of patients with chronic pain prescribed opioids that addressed effectiveness or harms versus placebo, no opioid use, or nonopioid pharmacological therapies; different opioid dosing methods; or risk mitigation strategies. Effects were analyzed at short-term (1 to <6 months), intermediate-term (≥6 to <12 months), and long-term (≥12 months) followup. Studies on the accuracy of risk prediction instruments for predicting opioid use disorder or misuse were also included. Random effects meta-analysis was conducted on short-term trials of opioids versus placebo, opioids versus nonopioids, and opioids plus nonopioids versus an opioid or nonopioid alone. Magnitude of effects was classified as small, moderate, or large using predefined criteria, and strength of evidence was assessed.

Results:

We included 115 randomized controlled trials (RCTs), 40 observational studies, and 7 studies of predictive accuracy; 134 were new to this update. Opioids were associated with small benefits versus placebo in short-term pain, function, and sleep quality. There was a small dose-dependent effect on pain, and effects were attenuated at longer (3 to 6 month) versus shorter (1 to 3 month) followup. Opioids were associated with increased risk of discontinuation due to adverse events, gastrointestinal adverse events, somnolence, dizziness, and pruritus versus placebo. In observational studies, opioids were associated with increased risk of an opioid abuse or dependence diagnosis, overdose, all-cause mortality, fractures, falls, and myocardial infarction versus no opioid use; there was evidence of a dose-dependent risk for all outcomes except fracture and falls.

There were no differences between opioids and nonopioid medications in pain, function, or other short-term outcomes. Opioid plus nonopioid combination therapy was associated with little improvement in pain at short-term followup versus an opioid alone. Co-prescription of benzodiazepines or gabapentinoids was associated with increased risk of overdose versus an opioid alone. No RCT evaluated intermediate- or long-term benefits of opioids versus placebo. One trial found stepped therapy starting with opioids to be associated with higher pain intensity and no difference in function or other outcomes versus stepped therapy starting with nonopioid therapy.

Limited evidence indicated no differences between long- and short-acting opioids in effectiveness, but long-acting opioids were associated with increased risk of overdose. One RCT found a taper support intervention associated with greater improvement in function but no difference in pain versus usual care.

Estimates of diagnostic accuracy for various risk prediction instruments were highly inconsistent, and there was no evidence on the effectiveness of risk mitigation strategies for improving clinical outcomes, with the exception of one study that found provision of naloxone associated with decreased emergency department visits.

Trials of patients with prescription opioid dependence found buprenorphine maintenance associated with better outcomes than buprenorphine taper and similar effects of methadone versus buprenorphine. Evidence was insufficient to evaluate benefits and harms of opioid therapy in patients at higher risk for opioid use disorder.

Conclusions:

At short-term followup, for patients with chronic pain, opioids are associated with small beneficial effects versus placebo but are associated with increased risk of short-term harms and do not appear to be superior to nonopioid therapy. Evidence on intermediate-term and long-term benefits remains very limited, and additional evidence confirms an association between opioids and increased risk of serious harms that appears to be dose-dependent. Research is needed to develop accurate risk prediction instruments, determine effective risk mitigation strategies, clarify risks associated with co-prescribed medications, and identify optimal opioid tapering strategies.