Background The goal of this investigation is to discover whether or not patients with psychiatric diagnoses are less likely to be prescribed opioids for pain in emergency departments compared with other patients.

Methods Pain-related visits to US emergency departments were identified using reason-for-visit and physician diagnosis codes for 13 years (1993–2005) of the National Hospital Ambulatory Medical Care Survey. The outcome measure was the prescription or administration of an opioid analgesic.

Results Roughly 10 million pain-related visits were made by persons with psychiatric diagnoses in the USA between 1993 and 2005. Across all years, only 18% (95% CI 16 to 20) of pain-related visits by patients with psychiatric diagnoses resulted in an opioid prescription, whereas 33% (95% CI 32 to 34) of visits by other patients did. Lower prescription rates for patients with psychiatric diagnoses were seen for every year of the survey and this difference occurred at every level of pain severity. Controlling for confounding factors did not attenuate this difference. In a multivariate model, patients with psychiatric diagnoses were about half as likely as other patients to be prescribed opiates (adjusted OR 0.49; 95% CI 0.44 to 0.56). Major limitations of the study include the uncertain precision of psychiatric and drug/alcohol diagnoses and the lack of detail about each patient visit.

Conclusion Having a psychiatric diagnosis was associated with a lower likelihood of receiving an opioid among persons presenting with pain to the ED.