State-to-State Variation in Opioid Dispensing Changes Following the Release of the 2016 CDC Guideline for Prescribing Opioids for Chronic Pain
Xiru Lyu,Gery P Guy,Grant T Baldwin,Jan L Losby,Amy S B Bohnert,Jason E Goldstick
DOI: https://doi.org/10.1001/jamanetworkopen.2023.32507
2023-09-05
Abstract:Importance: Evidence suggests that opioid prescribing was reduced nationally following the 2016 release of the Guideline for Prescribing Opioids for Chronic Pain by the US Centers for Diseases Control and Prevention (CDC). State-to-state variability in postguideline changes has not been quantified and could point to further avenues for reducing opioid-related harms. Objective: To estimate state-level changes in opioid dispensing following the 2016 CDC Guideline release and explore state-to-state heterogeneity in those changes. Design, setting, and participants: This cross-sectional study included information on opioid prescriptions for US individuals between 2012 and 2018 from an administrative database. Serial cross-sections of monthly opioid dispensing trajectories in each US state and the District of Columbia were analyzed using segmented regression to characterize preguideline dispensing trajectories and to estimate how those trajectories changed following the 2016 guideline release. Data were analyzed January to March 2023. Exposure: The March 2016 CDC Guideline for Prescribing Opioids for Chronic Pain. Main outcomes and measures: Four measures of opioid dispensing: opioid dispensing rate per 100 000 persons, long-acting opioid dispensing rate per 100 000 persons, high-dose (90 or more morphine milligram equivalents [MME] per day) dispensing rate per 100 000 persons, and average per capita MME. All measures were calculated monthly, from January 2012 through December 2018. Results: Data from approximately 58 900 retail pharmacies were included in analysis, representing approximately 92% of US retail prescriptions. The overall monthly dispensing rate in the US in early 2012 was approximately 7000 per 100 000 population. Following the 2016 guideline release, the already-decreasing slope accelerated nationally for the overall dispensing rate (preguideline slope, -23.19; postguideline slope, -48.97; change in slope, 25.97 [95% CI, 18.67-32.95]), long-acting dispensing rate (preguideline slope, -1.03; postguideline slope, -5.94; change in slope, 4.90 [95% CI, 4.26-5.55]), high-dose dispensing (preguideline slope, -3.52; postguideline slope, -7.63; change in slope, 4.11 [95% CI, 3.49-4.73]), and per-capita MME (preguideline slope, -0.22; postguideline slope, -0.58; change in slope, 0.36 [95% CI, 0.30-0.42]). For all outcomes, nearly all states showed analogous acceleration of an already-decreasing slope, but there was substantial state-to-state heterogeneity. Slope changes (preguideline - postguideline slope) ranged from 9.15 (Massachusetts) to 74.75 (Mississippi) for overall dispensing, 1.88 (Rhode Island) to 13.41 (Maine) for long-acting dispensing, 0.71 (District of Columbia) to 13.68 (Maine) for high-dose dispensing, and 0.06 (Hawaii) to 0.91 (Arkansas) for per capita MME. Conclusions and relevance: The 2016 CDC Guideline release was associated with broad reductions in prescription opioid dispensing, and those changes showed substantial geographic variability. Determining the factors associated with these state-level differences may inform further improvements to ensure safe prescribing practices.