Opioid and Benzodiazepine Use Among a Population-Based Cohort of Adults With Chronic Spinal Cord Injury: Correspondence Between Self-report and State Prescription Monitoring Data.

OBJECTIVE

To compare self-reported opioid and benzodiazepine use to rates of prescriptions dispensed among persons with chronic spinal cord injury (SCI), to evaluate the extent of potential underreporting.

DESIGN

Cross-sectional self-report assessment (SRA) and retrospective review of prescription monitoring program (PMP) data.

SETTING

Community dwelling adults in the Southeastern United States.

PARTICIPANTS

In total, 345 adults (N=345) (>18y) with chronic (>1y) SCI who were identified through a state population-based SCI registry and who responded to SRA opioid use questions in a parent study.

INTERVENTIONS: Not applicable.

MAIN OUTCOME MEASURES: Self-reported prescription pain reliever, sedative, and tranquilizer use, and prescriptions dispensed based on PMP data. We compared self-reported past year utilization with records of dispensed prescriptions to assess potential underreporting, defined as reporting "never" using either opioids (pain relievers) or benzodiazepines (sedatives/tranquilizers) in the past year, but having had 2 or more respective prescriptions dispensed in the year before the SRA.

RESULTS

Among the 345 participants who responded to the opioid use questions, we were able to match 252 to the PMP; reasons for not matching include not filling a controlled substance prescription in-state, either because of lack of reportable prescriptions or the prescription was dispensed out of state. Evaluating each prescription medication, rates of underreporting for each drug ranged from 0.4% to 4% for hydrocodone, oxycodone, tramadol, codeine, buprenorphine, and hydromorphone. There was 100% agreement in the self-reported use and dispensed records for fentanyl, oxymorphone, meperidine, and methadone. The rates of potential underreporting of benzodiazepines were <2%. Overall, of the 252 matched individuals, 11% underreported opioids (2 or more prescriptions dispensed) and 1% underreported benzodiazepines.

CONCLUSIONS

The findings suggest considerable agreement between self-reported and dispensed prescription medication use, and only minimal potential underreporting of select opioids and benzodiazepines, confirming the appropriateness of self-report of these prescription medications with people with SCI.