The coronavirus disease 2019 (COVID-19) pandemic has profoundly disrupted health care delivery in the US.1 The Centers for Disease Control and Prevention noted a 9.1% increase in reported 12-month counts of drug overdose deaths from March 2019 to March 2020, from 67 726 to 73 860.2 On March 13, 2020, a COVID-19 national emergency was declared. To diminish potential barriers to treatment access, 3 days later, federal guidelines on telemedicine use were released, providing authorized practitioners increased flexibility to prescribe buprenorphine to patients with opioid use disorder (OUD) during this public health emergency.3 Other local, state, and federal policy initiatives have also attempted to preserve access to medication treatment for OUD, yet the cumulative outcome of these undertakings is not clear.
Using retail pharmacy claims from Symphony Health, US database that includes 92% of retail pharmacy claims,4 we examined secular trends of total US retail pharmacy sales and sales of buprenorphine products with a US Food and Drug Administration–approved indication for treatment of OUD.5 We counted the number of individuals who filled prescriptions in each week between May 1, 2019, and June 28, 2020, except for the week of March 8 to March 15, 2020, which was excluded because this was the week before the transitioning week (March 16). The results were not sensitive to analysis including this week.
We plotted the raw and fitted number of individuals filling prescriptions during the prepandemic and pandemic periods. We defined the transition between these periods as occurring on March 16, 2020, when the Substance Abuse and Mental Health Services Administration and Drug Enforcement Administration approved telemedicine use for buprenorphine prescribing. We further stratified national changes by payer, including Medicaid, Medicare Part D, commercial plans, and prescriptions paid in cash. Using quasi-experimental interrupted time series analysis, we examined whether there was a change in the level or rate of growth (slope) of filled prescriptions between the 2 periods. Data analysis was conducted with Stata, version 16.1 (StataCorp LLC), using 2-sided t tests for regressions, with significance set at P < .05. The study was exempted from review by Indiana University’s Institutional Review Board. Data were deidentified. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
The weekly number of individuals filling prescriptions of buprenorphine indicated for OUD increased steadily from 172 165 in May 2019 to 216 939 in March 2020—a 26.01% increase. The weekly number of individuals filling prescriptions of any type also increased steadily during this period, albeit at a lower rate of 8.78% (Figure, A).
The weekly growth rate in the level of prescription fills for buprenorphine significantly declined by 0.50% (95% CI, −0.84% to −0.15%) between the prepandemic and pandemic periods (Table). We found no statistically significant or sudden declines in the level of these prescriptions in March (intercepts). In contrast to buprenorphine prescriptions, the pandemic period was associated with an abrupt and statistically significant decline in both the level (−5.03 million, 95% CI, −6.85 million to −3.20 million) and growth rate (−0.57%, 95% CI, −1.01% to −0.14%) when considering all filled prescriptions collectively.
Filled prescriptions for buprenorphine paid in cash significantly declined in both the levels and growth rate in the pandemic period; no significant changes were found in filled prescriptions paid by Medicaid (Figure, B and Table). In general, the changes in prescriptions paid by Medicare and commercial plans followed the patterns described above for all buprenorphine prescriptions.
Our findings suggest that, since the COVID-19 national emergency declaration, the number of individuals filling buprenorphine prescriptions has plateaued but has not decreased; however, filled prescriptions for all medications collectively have decreased considerably. Study limitations include our inability to identify off-label prescribing, illicit diversion, or dispensing of buprenorphine via opioid treatment programs. In addition, we cannot precisely identify treatment initiation vs continuation. Efforts to engage patients with OUD via telehealth may have helped retain patients who otherwise would have discontinued treatment. It is also plausible that continued telehealth service provision plays an important role in addressing the ongoing opioid crisis. Given buprenorphine’s role in treating OUD, our findings may contribute to a better understanding of the pandemic’s association with the treatment of OUD.
Accepted for Publication: October 24, 2020.
Published Online: December 21, 2020. doi:10.1001/jamainternmed.2020.7497
Corresponding Author: Thuy D. Nguyen, PhD, Department of Health Management and Policy, School of Public Health, University of Michigan, 1415 Washington Heights, M3234 SPH II, Ann Arbor, MI 48109 (thuydn@umich.edu).
Author Contributions: Dr Nguyen had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Nguyen, Gupta, Ziedan, Simon.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Nguyen, Gupta, Ziedan, Simon, Stein.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Nguyen, Gupta, Ziedan.
Administrative, technical, or material support: Simon.
Supervision: Simon.
Conflict of Interest Disclosures: Dr Alexander reported being a past chair of the US Food and Drug Administration’s Peripheral and Central Nervous System Advisory Committee; serving as a paid adviser to IQVIA; being a cofounding principal and equity holder in Monument Analytics, a health care consultancy whose clients include the life sciences industry as well as plaintiffs in opioid litigation; and being a member of OptumRx's National P&T Committee. No other disclosures were reported.
Funding/Support: Dr Stein was supported by grant 1R01DA045800-01 from the National Institute on Drug Abuse during the conduct of the study.
Role of the Funder/Sponsor: The data sponsor was not involved in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.
Disclaimer: The statements, findings, conclusions, views, and opinions contained and expressed herein are not necessarily those of Symphony Health or any of its affiliated or subsidiary entities.
Additional Information: The data, technology, and services used in the generation of these research findings were supplied pro bono by the COVID-19 Research Database partners, who are acknowledged at https://covid19researchdatabase.org.