Practice change intentions after academic detailing align with subsequent opioid prescribing

Abstract

Objective: Self-reported behavior change is used to evaluate the effectiveness of educational outreach interventions delivered to clinicians, such as academic detailing (AD). However, self-reported changes in behavior are often not corroborated with data on actual behavior change. To assess alignment between self-reported practice change intentions and actual opioid prescribing behavior among primary care clinicians after an AD intervention.

Methods: We used a difference-in-differences approach to compare pre-post changes in opioid prescribing using data from the Illinois Prescription Monitoring Program. An opioid-focused AD intervention was delivered to primary care clinicians from a large health system in the Chicago metropolitan area from June 2018 to August 2018. Immediately after the AD intervention, clinicians were administered a single-item self-reported practice change measure. Clinicians were categorized into 2 groups on the basis of their responses: (1) intention to change and (2) no-to-moderate intention to change. Outcomes were mean total opioid prescriptions and high-dose opioid prescriptions (≥ 90 morphine milligram equivalents) per clinician per month. Repeated measures linear regression models were used to compare changes in opioid prescribing outcomes between the 2 groups in the 6 months before and after the AD intervention.

Results: A total of 149 clinicians were included for analysis. An intention to change was reported by 72 clinicians and no-to-moderate intention to change was reported by 77 clinicians. In the 6 months after the AD intervention, there were 1.48 (95% CI -2.48 to -0.47) fewer total opioid prescriptions and 0.50 (-0.69 to -0.31) fewer high-dose opioid prescriptions per clinician per month in the intention to change group than in the no-to-moderate intention to change group.

Conclusion: This study showed considerable alignment between self-reported practice change intentions and actual changes in opioid prescribing behavior. Future opioid-focused educational outreach interventions should consider using standardized single-item practice change measures as an immediate indicator of future behavior change.