Skip to main content
NCBI home page
Drug and Alcohol Dependence Reports logoLink to Drug and Alcohol Dependence Reports
. 2023 Jan 20;6:100135. doi: 10.1016/j.dadr.2023.100135

Impact of COVID-19-related regulatory changes on nationwide access to buprenorphine: An interrupted time series design

Payel Jhoom Roy a,⁎, Katherine Callaway Kim b, Katie Suda a,b,c, Jing Luo a, Xiaoming Wang d, Donna Olejniczak a, Jane M Liebschutz a
PMCID: PMC9851915  PMID: 36694665

Highlights

  • •

    Patients received longer prescriptions of buprenorphine.

  • •

    Overall total milligrams buprenorphine increased post-initial pandemic period.

  • •

    Stably-treated patients experienced fewer treatment disruptions.

  • •

    Medicaid patients had improved access to buprenorphine.

  • •

    Regulatory changes appear to have helped patients maintain access to buprenorphine.

Keywords: Buprenorphine, COVID-19, Buprenorphine-naloxone, Access, Pandemic

Abstract

Background

The impact of COVID-19-related healthcare changes on access to buprenorphine (BUP) nationwide in the US is unknown.

Methods

We conducted an interrupted time series with the IQVIA LRx database. The study timeline included BUP prescriptions from 52 weeks before (2/23/19–2/21/20) to 52 weeks after (4/4/20–4/2/21) the initial pandemic period (2/22/20–4/3/20). Segmented regression estimated relative changes in total milligrams (MG) of BUP available per week nationwide at 1, 26, and 52 weeks post-initial-pandemic. We evaluated treatment disruptions in previously stable patients, defined as ≥6 months of BUP prescriptions.

Results

A total of 31 617 849 prescriptions were included. Total MG BUP dispensed increased at 1 and 26 weeks and then returned to baseline trends at 52 weeks post-initial pandemic period (4.1% [95% CI: 3.7,4.5], 2.1% [1.5,2.6], 0.1% [-0.6,0.9]). Stably-treated patients saw a decrease in 7-, 14-, and 28-day treatment disruptions at 52 weeks post-initial-pandemic period (-21.6% [-25.6,-17.7]; -10.8% [-16.3,-5.3]; -27.3% [-33.0,-21.6]). Men retained an increase in MG BUP compared to women at 52 weeks (0.7% [0.01,1.4] versus -0.6% [-1.5,0.2]). Younger age groups (18–29 years and 30–39 years) had a decrease in MG BUP at 52 weeks compared to expected baseline trend (-16.6 [-24.2, -9.0]; -1.6 [-3.0, -0.1). Patients with Medicaid demonstrated an increase in MG BUP at 52 weeks (8.3% [6.3,10.3]). MG BUP prescribed by APP prescribing increased by over 140 000 mg per week prior to the pandemic and continued to increase.

Conclusions

Regulatory changes around buprenorphine prescribing facilitated patient access to buprenorphine during the pandemic.

1. Introduction

COVID-19-related social distancing orders disrupted many aspects of health care delivery, including treatment for substance use disorders, concurrent with the national emergency declaration on March 13, 2020, backdated to March 1, 2020 (Trump, 2020). In health care settings, physician offices closed (Ellison, 2020) or converted to telemedicine practices (Mehrotra et al., 2020) and elective surgeries were limited (Sarac et al., 2020). These changes resulted in an overall decrease in health care utilization (Cox et al., 2020).

Experts have warned about the dangers of an overlap between the COVID-19 pandemic and the addiction epidemic (Volkow, 2020). In the 12-month period ending April 2021, over 100,000 Americans died from a drug-related overdose, an estimated 75% of which involved an opioid (Centers for Disease Control and Prevention, 2021). A recent study in a large, national emergency medical services database demonstrated consistently elevated rates of all overdose incidents and overdose-related cardiac arrests in the year 2020 compared to 2018–2019 (Friedman et al., 2020). The opioid epidemic and COVID-19 pandemic appear to have converged, with uncertain but potentially dire consequences.

FDA-approved medications for the treatment of opioid use disorder (including methadone, buprenorphine, and naltrexone; MOUD) significantly reduce opioid-related morbidity and mortality (Larochelle et al., 2018; Mancher and Leshner, 2019). Despite these life-saving benefits, access to MOUD remains limited due to multiple patient, provider, and structural barriers (Sharma et al., 2017; Weintraub et al., 2018). Federal regulations require that providers obtain a specialized waiver from the Drug Enforcement Agency (DEA) to prescribe buprenorphine (Fiscella et al., 2018; Joudrey et al., 2019; Richert and Johnson, 2015; Wakeman and Rich, 2018). The Ryan Haight Online Pharmacy Consumer Protection Act of 2008 targeting rogue internet pharmacies resulted in the uniform practice of face-to-face encounters to prescribe controlled substances (110TH, 2008; Lacktman, 2017; Shore, n.d.). This law prohibited initial prescriptions of controlled substances without an in person visit. While this Act historically allowed for some narrow exceptions for telemedicine (e.g. Indian Health Service or by a VA provider during a medical emergency) (Lacktman, 2017), uptake of telemedicine prescribing has been low (Harris et al., 2020).

The COVID-19 pandemic resulted in substantial changes to the regulatory environment for MOUD. On March 20, 2020, the Justice Department issued a press release stating waivered providers could prescribe buprenorphine for patients they have never met if the communication was conducted via a real-time, two-way, audiovisual telemedicine system (Drug Enforcement Agency, 2020a), On March 31, 2020, the Justice Department and DEA further amended their recommendations, stating new patient intakes could be conducted via telephone encounters only, without a video or in-person requirement (Drug Enforcement Agency, 2020b). This further expanded access to buprenorphine for patients who did not have audiovisual capabilities, such as access to internet connections, laptops, and smartphones. Early studies found these changes may have had a positive impact on ensuring continued access for patients receiving buprenorphine up to 6 months after the pandemic declaration (Cremer et al., 2022; Currie et al., 2021; Nguyen et al., 2021).

The current study builds on previous analyses by evaluating national changes in buprenorphine access as a result of COVID-19-related prescribing guideline changes up to one-year post-initial-pandemic period. We hypothesize that during the pandemic, many patients may have received fewer prescriptions but for longer duration. Unlike previous studies which only assessed the number of patients receiving a buprenorphine prescription or the number of prescriptions dispensed, we constructed a novel measure of access which accounts for total exposure to medication, including length of prescription, proportion of days covered, daily dose, and total milligrams dispensed. We additionally quantified differences in buprenorphine access by patient and prescriber characteristics, given concerns that the COVID-19 pandemic may have worsened access to MOUD (Feder et al., 2019; Foti et al., 2021; Hadland et al., 2017; Hawk et al., 2021; Hicks et al., 2015; Madgavkar et al., 2020; Samples et al., 2018; Stein et al., 2021).

2. Material and methods

2.1. Data source

We conducted an observational study of national buprenorphine dispensing from 2/23/2019 to 4/2/2021 in the IQVIA Longitudinal Prescription database (LRx) (Olfson et al., 2020). This all-payer dataset comprised 90% of prescriptions filled in U.S. retail pharmacies, 60–85% of mail-order service, and 75–80% of long-term care. LRx does not contain information on diagnoses or procedures. Patient race, ethnicity, and residential location were unavailable to protect privacy. The University of Pittsburgh Institutional Review Board approved the study.  We followed STROBE reporting guidelines for cross-sectional studies (von Elm et al., 2014).

2.2. Study cohort

The study cohort comprised individuals who filled any buprenorphine prescription from 2/23/19 to 4/2/21.  As shown in Appendix Figure 1 in the Supplement, we excluded prescriptions to individuals <18 years old, with missing or international prescriber identifiers, and with dosages outside typical clinical ranges (<1 or >90 days supplied, daily dose <0.35 or >32 MG). Included products were all sublingual films and tablets, subcutaneous suspension, and implant formulations of buprenorphine and buprenorphine-naloxone FDA-approved to treat opioid use disorder (Appendix Figure 1 in the Supplement). For long-acting injectable formulations, we divided the prescribed dose by the days supplied (e.g. 300 mg / 28 days = 10.7 mg/day). We excluded buprenorphine formulations FDA-approved for the treatment of chronic pain (e.g. transdermal buprenorphine).

2.3. Outcomes

The primary outcome was the total number of milligrams (MG) of buprenorphine available each week in the one year before and the one year after the pandemic declaration. This outcome allowed us to capture changes in both the quantity (i.e.: number of prescriptions) and length (i.e.: days supplied) of buprenorphine prescriptions. Using the discrete days supplied and dosage variables in LRx, we created a weekly record for each patient which captured the number of covered days and average daily dose. We defined total MG across all patients as the sum of days covered multiplied by average daily dose. We tallied the total MG BUP available across all formulations indicated for opioid use disorder. We assumed that formulation did not change over time. Because injectable buprenorphine use may have decreased due to the requirement for an in-person visit, we ran similar analyses on buprenorphine films only to evaluate trends (see Statistical Analyses below).

We had several secondary outcomes to understand drivers of observed changes in total MG, including (1) mean MG per prescription, (2) mean days supplied per prescription, and the number of (3) new and (4) existing patients with active prescriptions in each week. New patients were defined as having received no active fill in the previous 365 days while existing patients were defined as having received at least one active fill in the previous 365 days. “New” patients were considered part of the “existing” group in any weeks with active medication following their first prescription. Finally, we calculated the percentage of patients stably-treated with buprenorphine (≥180 days of continuous treatment) who failed to fill an expected prescription within (a) 7, (b) 14, or (c) 28 days from the last day supplied of their most recent fill.

2.4. Sub-group analyses

To understand differential impacts of the pandemic on buprenorphine access, we performed exploratory sub-group analyses for our primary outcome (total MG) by (a) patient gender (Male/Female), (b) patient age group (18–29, 30–39, 40–49, 50–64, ≥65 years), (c) payment type (Medicaid, Medicare, 3rd party insurance, cash), and (d) prescriber specialty (advanced practice provider [APP, including physician assistants and nurse practitioners] or physician [General Medicine, Addiction Medicine, Psychiatry and Neurology, Emergency Medicine, Pain Medicine, and Other]). The APP group was of particular interest given the relatively recent passage of the Comprehensive Addiction and Recovery Act of 2016 that broadened the scope of buprenorphine prescribing to APPs (Fornili and Fogger, 2017; Stein et al., 2021). Each of these variables was sourced from LRx, except for prescriber specialty, which was linked from the National Plan and Provider Enumeration System (NPPES, see Appendix Table 2 in the Supplement).

2.5 Statistical Analysis

We used segmented regression interrupted time series (ITS) models to quantify relative changes in total MG of buprenorphine available during the first year of the COVID-19 pandemic (4/4/20–4/2/21), relative to the year prior (2/23/19–2/21/20) (Beard et al., 2019). A priori, we excluded the three weeks before and three weeks after the initial national emergency declaration (2/22/20–4/3/20) when defining the initial-pandemic period. This wash-out period accounted for potential practice changes in anticipation of the pandemic. Because we were interested in national changes in total buprenorphine utilization pre- versus post-pandemic, the ITS models used aggregated (rather than individual-level) data of the entire population of interest (all available data; not a sample). Model terms included immediate changes in the first week post-initial-pandemic-period (level change) and changes in slope thereafter (trend change). We included quadratic trend change terms to allow for nonlinear effects over time. For each outcome, we selected for only level, trend, and quadratic trend change terms which were significant at an entry and exit cutoff of α=0.2. We accounted for autocorrelation by selecting for significant (p-value <0.05) auto-regressive terms up to a lag of 3. These selections were performed separately for each model to allow for differences between the outcomes and by stratified group. We calculated overall effects (capturing both changes in levels and trends) by comparing observed versus expected values of each outcome at 1 week (4/10/20), 6 months (26 weeks, 10/2/20), and 1 year (52 weeks, 4/2/21) post-initial-pandemic period. These overall relative changes were defined as: (the observed weekly value – the expected weekly value) / (the expected value). Established methods were used to calculate 95% confidence intervals using the standard errors from the final selected autoregressive ITS models (Wagner et al., 2002; Zhang et al., 2009). All data pulls and statistical analyses were conducted in SAS version 9.4 (Cary, NC). Figures were created in RStudio, version 1.3.1093 (Boston, MA).

2.5. Sensitivity analyses

We performed several sensitivity analyses to test the robustness of our results. First, we repeated our main ITS analyses without level and trend change term variable selection (while still accounting for auto-correlation). Second, we tested shorter interruption, or washout, periods of 4 weeks (3/1/20–3/27/20), 2 weeks (3/7/20–3/20/20), and 1 week (3/7/20–3/13/20).  Finally, we repeated our analysis of treatment disruptions using 90 instead of 180 days to define stable treatment.

3. Results

A total of 31 617 849 prescriptions were included. Table 1 displays cohort characteristics in three time periods: pre-initial pandemic, initial pandemic, and post-initial pandemic.

Table 1.

Cohort Characteristics.

Pre-Initial Pandemic (2/23/19 −2/21/20) Initial Pandemic Perioda (2/22/20–4/3/20) Post-Initial Pandemic (4/4/20 - 4/2/21)
No. BUP Prescriptions 15 071 820 2 383 853 15 505 091
No. Patients 1 269 651 814 013 1 329 502
Female Gender, No. (%) 553 121 (43.6) 357 945 (44.0) 576 823 (43.4)
Mean (Std) Age, years 41.4 (12.1) 42.2 (12.5) 41.8 (12.1)
Age Group, No. (%)
 18–29 years 217 606 (17.1) 97 589 (12.0) 199 411 (15)
 30–39 years 498 458 (39.3) 301 760 (37.1) 519 123 (39)
 40–49 years 307 634 (24.2) 201 904 (24.8) 341 885 (25.7)
 50–64 years 251 186 (19.8) 169 922 (20.9) 270 419 (20.3)
 ≥65 years 62 063 (4.9) 42 874 (5.3) 73 221 (5.5)
Stably Treated Patients,b No. (%)
 Without disruption ≥ 7 days 537 201 (42.3) 358 826 (44.1) 621 931 (46.8)
 Without disruption ≥ 14 days 598 802 (47.2) 424 286 (52.1) 686 963 (51.2)
 Without disruption ≥ 28 days 650 054 (51.2) 485 633 (59.7) 741 411 (55.8)
Payment Typec, No. Rx. (%)
  Medicaidd 5 986 968 (39.7) 964 943 (40.5) 6 805 580 (43.9)
 3rd Party Insurance 6 325 100 (42.0) 996 853 (41.8) 6 027 437 (38.9)
 Medicaree 1 435 990 (9.5) 252 029 (10.6) 1 605 862 (10.4)
 Cash 1 323 104 (8.8) 169 801 (7.1) 1 065 257 (6.9)
Prescriber Specialtyc, No. Rx. (%)
  General Medicine Physician 5 745 972 (38.1) 867 944 (36.4) 5 224 681 (33.7)
 Advanced Practice Provider (APP) 3 055 853 (20.3) 610 455 (25.6) 4 862 473 (31.4)
 Addiction Medicine Physician 2 116 989 (14.0) 299 048 (12.5) 1 802 189 (11.6)
 Psychiatry & Neurology Physician 1 795 617 (11.9) 265 996 (11.2) 1 541 268 (9.9)
 Emergency Medicine Physician 518 300 (3.4) 75 290 (3.2) 480 515 (3.1)
 Pain Medicine Physician 412 374 (2.7) 65 883 (2.8) 384 789 (2.5)
 Other Physicianf 1 426 715 (9.5) 199 237 (8.4) 1 209 176 (7.8)
Open in a new tab

Abbreviations: No., number, Rx., prescription.

Notes: a. Prescriptions within the 6 weeks surrounding the pandemic declaration (2/23/19–4/3/20, midpoint: 3/13/20) were not included in the interrupted time series analyses. b. Stable treatment defined as ≥ 180 days supplied of buprenorphine without a treatment disruption (i.e. 7, 14, or 28 days without buprenorphine prescription). c. Variable reported on prescription level. d. Included fee-for-service and managed Medicaid. e. Included traditional and Medicare Advantage plans. f. Other physicians included surgery, anesthesiology, obstetrics/gynecology, specialists, pediatrics, physical medicine & rehabilitation, preventative medicine, and all other specialties (see Appendix Table 2 in Supplement).

3.1. Nationwide trends

The results of the interrupted time series analysis are displayed in Table 2. The total MG BUP dispensed increased at 1 and 26 weeks and then returned to baseline trends at 52 weeks post-initial pandemic period (4.1% increase [95% CI: 3.7,4.5], 2.1% [1.5,2.6], 0.1% [−0.6,0.9], respectively; Fig. 1A). The mean MG per prescription increased at 1, 26, and 52 weeks (9.3% [95% CI: 8.6, 10], 4.6% [3.9, 5.3], 5.5% [4.4, 6.6], respectively; Fig. 1B), as did the mean days supplied (8.8% increase [95% CI: 8.2,9.5]; 4.4% [3.7,5.1]; 5.9% [4.9,6.9], respectively; Fig. 1B). Fig. 1C shows the trends for number of active patients per week stratified by new and existing categories. The number of existing patients increased at 1 (2.3% [95% CI: 2.0,2.6]) and 26 (1.0% [95% CI: 0.6,1.3]) weeks compared to expected trends but decreased at 52 weeks (−1.3% [95% CI: −1.8,−0.8]; the number of new patients initially did not change at 1 week (0% [95% CI: 0,0]), decreased at 26 weeks (−5.4% [95% CI: −9.6,−1.2]), and then increased compared to expected baseline trend at 52 weeks (6.4% [95% CI: 0.04,12.7]). Appendix Table 3 provides the number of patients as denominators at each time point.

Table 2.

Interrupted Time Series Analysis of National Changes in Buprenorphine Prescribing, Overall and by Selected Characteristics, 2/23/2019 – 4/2/2021.

Variable Pre-Initial Pandemic Period (2/23/19–2/21/20) Post-Initial Pandemic Period (4/4/20–4/2/21) Relative Change from Baseline Trendb (95% CI)
Intercept Weekly Trend Δ Levela Δ Trenda Δ Trend2,a 4/10/20 (1 wk.) 10/2/20 (26 wks.) 4/2/21 (52 wks.)
No. Patients 625.8 k 1.2 k 18.1 k** −448.2** 2.6 (2.3,2.9) 0.9 (0.6,1.3) −0.6 (−1.1,−0.1)
 Existing Patients 608.9 k 1.3 k 15.8 k** −218.7 −5.5* 2.3 (2.0,2.6) 1.0 (0.6,1.3) −1.3 (−1.8,−0.8)
 New Patientsc 32.8 k −179.3 4.3* 0 (0,0) −5.4 (−9.6,−1.2) 6.4 (0.04,12.7)
Mean MG per Rx. 227.4 0.23 22.4** −0.7** 0.01** 9.3 (8.6,10.0) 4.6 (3.9,5.3) 5.5 (4.4,6.6)
Mean Days Supplied per Rx. 15.9 0.01 1.5** −0.05** 0.0008** 8.8 (8.2,9.5) 4.4 (3.7,5.1) 5.9 (4.9,6.9)
Total MG Available 62.0 M 132.9 k 2.9 M** −54.1 k** 4.1 (3.7,4.5) 2.1 (1.5,2.6) 0.1 (−0.6,0.9)
Gender
 Male 34.6 M 70.4 k 1.6 M** −26 k** 4.2 (3.8,4.6) 2.4 (1.9,2.9) 0.7 (0.01,1.4)
 Female 27.4 M 62.6 k 1.2 M** −28.4 k** 4.0 (3.5,4.4) 1.6 (1.0,2.2) −0.6 (−1.5,0.2)
Payment Type
 Medicaidd 22.5 M 58.7 k 2.1 M** 45.7 k* −764.3* 7.9 (7.0,8.9) 9.9 (8.4,11.5) 8.3 (6.3,10.3)
 3rd Party Insurance 27.8 M 53 k 335.7 k* −90.6 k** 940.6** 1.1 (0.4,1.8) −4.2 (−5.1,−3.3) −5.5 (−6.7,−4.3)
 Medicaree 7.2 M 25.1 k 343.1 k** −12.1 k** 3.9 (3.3,4.6) 0.4 (−0.3,1.2) −2.8 (−3.8,−1.7)
 Cash 4.3 M 94.1 k −10.5 k** 2.2 (0.7,3.6) −3.9 (−5.3,−2.5) −10.2 (−12.6,−7.7)
Age Group
 18–29 years 8.1 M 534.1 k** −720.9** 6.6 (5.3,7.9) 1.0 (−1.1,3.2) −16.6 (−24.2,−9.0)
 30–39 years 23.4 M 42.4 k 1 M** −555.6** 3.9 (3.2,4.6) 2.5 (1.6,3.3) −1.6 (−3.0,−0.1)
 40–49 years 14.8 M 46.2 k 717.7 k** −35.2 k 762.9 4.1 (2.9,5.3) 1.7 (−1.6,5.0) 4.6 (0.3,8.8)
 50–64 years 12.9 M 38.1 k 656.4 k** −39.6 k* 626.6* 4.3 (3.3,5.4) 0.4 (−1.4,2.1) 1.6 (−0.7,3.9)
 ≥65 years 2.5 M 14.2 k 95.9 k* −13.7 k 296.7 2.8 (1.1,4.6) −1.6 (−8.0,4.7) 4.1 (−3.9,12.1)
Prescriber Specialty
 General Medicine 25.4 M 16.3 k 1.1 M** −46.7 k** 336.2 4.2 (3.4,4.9) 0.5 (−0.5,1.6) −1.5 (−2.9,−0.1)
 APP 7.9 M 141.6 k 688 k** −346.4* 4.2 (3.1,5.4) 2.4 (1.1,3.7) −0.9 (−3.5,1.7)
 Addiction Medicine 9.3 M −10.1 k 344.1 k** 4 (3.1,4.9) 4.1 (3.1,5.0) 4.2 (3.2,5.2)
 Psych. & Neuro. 8.7 M −8.7 k 332.5 k** −4.5 k* 4.1 (3.4,4.7) 2.8 (1.7,3.8) 1.4 (−0.3,3.0)
 Emergency Med. 2.1 M 2.1 k 23.8 k −26.7* 1.1 (0.1,2.1) 0.3 (−0.8,1.5) −2 (−4.2,0.2)
 Pain Med. 2.2 M 62.4 k** −2.8 k** 2.9 (1.9,3.8) −0.4 (−1.3,0.6) −3.7 (−5.4,−2.1)
 Other Physicianf 6.5 M −6.3 k 135.6 k** 2.4 k 2.2 (1.4,3.0) 3.3 (2.1,4.5) 4.5 (2.5,6.5)
% Stable Pat. with Disruption
 ≥7 days 1.3 −0.2** −0.002 −12.6 (−16.6,−8.5) −17.0 (−19.4,−14.6) −21.6 (−25.6,−17.7)
 ≥14 days 1.0 −0.002 −0.09* −9.7 (−15.1,−4.3) −10.2 (−15.7,−4.8) −10.8 (−16.3,−5.3)
 ≥28 days 0.6 −0.001 −0.07** 0.00003** −11.6 (−14.7,−8.5) −15.5 (−18.9,−12.1) −27.3 (−33.0,−21.6)
Open in a new tab

Abbreviations: Pat., patient; Num., number; Wk., week; Gen. Med., General Medicine; APP, Advanced Practice Provider; Psych & Neuro., Psychiatry & Neurology.

Notes: a. From segmented regression analyses with an interruption at the initial pandemic period (2/23/19–4/3/20). Non-statistically significant trend and level change terms were removed via forward-backward selection with an entry/exit cut-off of p = 0.2; removed terms are not shown in this table. * denotes p-values <0.05, ** denotes p-values<0.001. b. Relative changes were defined as: (the observed weekly value – the expected weekly value) / (the expected value). 95% confidence intervals were calculated based on the standard errors from the ITS model; bold denotes confidence intervals which do not contain zero. c. New patients were defined as those with no active buprenorphine prescription in the previous 365 days. d. Included fee-for-service and managed Medicaid. e. Included traditional and Medicare Advantage plans. f. Other physicians included surgery, anesthesiology, obstetrics/gynecology, specialists, pediatrics, physical medicine & rehabilitation, preventative medicine, and all other specialties (see Appendix Table 2 in Supplement).

Fig. 1.

Fig. 1

Open in a new tab

National Changes in Buprenorphine Prescribing, 2/23/2019–4/2/2021

Abbreviations: MG, milligrams; Rx., prescription.

Notes: a. Week identifiers run from Saturday to Friday; dates shown are for the last day in the week (Friday). Blue bar represents the initial pandemic period (2/22/20–4/3/20). Dashed lines represent expected values from the baseline trend. Dotted lines represent observed values. b. Weekly outcomes defined based on active prescriptions whose days supplied overlapped with that week; see Appendix Table 3 in Supplement for relevant denominators. c. Weekly outcomes defined based on prescriptions filled in that week; see Appendix Table 3 in Supplement for relevant denominators. d. New patients were those with no active prescription in the previous year (365 days); existing patients were those with at least one active prescription in the previous year.

3.2. Treatment disruptions

Stably-treated patients saw a decrease in 7-, 14-, and 28-day treatment disruptions at 52 weeks post-initial-pandemic period (−21.6% [95% CI: −25.6,−17.7]; −10.8% [−16.3,−5.3]; −27.3% [−33.0,−21.6]) compared to the expected baseline trend. Fig. 2 shows changes in rates of treatment disruptions.

Fig. 2.

Fig. 2

Open in a new tab

Percentage of Stably Treated Patients with a New Treatment Disruptiona, 2/23/19–4/2/2021.

Notes: a. Week identifiers run from Saturday to Friday; dates shown are for the last day in the week (Friday). Blue bar represents the initial pandemic period (2/22/20–4/3/20). Dashed lines represent expected values from the baseline trend and dotted lines represent observed values. New treatment disruptions for each week were defined based on the number of previously-stable patients (≥180 days of continuous treatment with no disruption) who failed to refill within 7, 14, or 28 days from the end date of their last active prescription.

3.3. Trends by stratified group

Men retained a greater pre- versus post-pandemic increase in MG BUP compared to women at 52 weeks (0.7% [95% CI: 0.01,1.4] versus −0.6% [−1.5,0.2]; Fig. 3A). All age groups initially saw an increase in MG BUP prescribed at 1 week, however younger age groups (18–29 years and 30–39 years) had a decrease in MG BUP at 52 weeks compared to pre-pandemic trends (−16.6% [95% CI: −24.2, −9.0]; −1.6% [−3.0,−0.1]; Fig. 3B).

Fig. 3.

Fig. 3

Open in a new tab

National Changes in Total MG of Available Buprenorphine, 2/23/2019–4/2/2021, by Selected Characteristics

Abbreviations: Pat., patient; Med., medicine; APP, Advanced Practice Provider; Psych/Neuro; Psychiatry & Neurology.

Notes: a. Week identifiers run from Saturday to Friday; dates shown are for the last day in the week (Friday). Blue bar represents the initial pandemic period (2/22/20–4/3/20). Dashed lines represent expected values from the baseline trend. Dotted lines represent observed data. b. Weekly outcomes defined based on active prescriptions whose days supplied overlapped with that week; see Appendix Table 3 in Supplement for relevant denominators. Patients with overlapping prescriptions from different payment types or provider categories were counted in both groups.

While all patients, regardless of payor type, initially saw an increase in MG BUP at 1 week, individuals with Medicaid had an increase in MG BUP at 52 weeks (8.3% [95% CI: 6.3,10.3]). Individuals paying with cash (−10.2% [95% CI: −12.6, −7.7]), commercial insurance (−5.5% [−6.7, −4.3]), and Medicare (−2.8% [−3.8, −1.7]) had a decrease in MG BUP compared to the expected baseline trend at 52 weeks (Fig. 3C).

General Medicine providers prescribed the highest MG BUP compared to other specialties; General Medicine demonstrated an initial increase in MG BUP at 1 week (4.2% [95% CI: 3.4,4.9]) and then a downward trend from 26 (0.5% [−0.5,1.6]) to 52 (−1.5% [−2.9,−0.1]) weeks. Addiction Medicine providers, compared to other physician specialties, had an increase in MG BUP dispensed at all time points (1 week: 4% [95%CI: 3.1,4.9]); 26 weeks: 4.1% [3.1,5.0]; 52 weeks: 4.2% [3.2,5.2]; Fig. 3D). MG BUP prescribed by APPs was increasing rapidly prior to the pandemic (trend: over 140 000 MG per week; Table 2) and in the first week post-initial pandemic, total MG prescribed by APPs increased by 688,000 (level change). However, this trend slowed as the pandemic continued (quadratic trend change = −346.4 MG/week) so that by 52 weeks, APP MG dispensed was not different than pre-pandemic trends.

3.4. Sensitivity analyses

The results of our sensitivity analyses are available in the supplemental appendix. ITS estimates shown in Appendix Table 4 report all level and trend changes, regardless of significance. These results were similar to our main analyses, which only included change terms with p-values < 0.2. When we evaluated changes in select outcomes (total MG available per week, total number of patients, mean MG per prescription, and mean days supplied per prescription) at 1, 26, and 52 weeks using shorter interruption periods (1, 2, and 4 weeks), we found similar trends as our primary analysis (Appendix Table 5). Appendix Figure 2 and Appendix Table 6 demonstrate treatment disruptions using a shorter time period (90 days vs 180 days). Similar to our main findings using 180 days, we found decreases in 7-, 14-, and 28-day treatment disruptions post-initial-pandemic.

4. Discussion

This study provides a robust profile of BUP available for the treatment of opioid use disorder in the first year of the COVID-19 pandemic. Our results demonstrate changes in 1) a broad range of prescribing practices, 2) new and existing patient numbers, and 3) treatment disruptions, using a national database that is inclusive of all payor types. Total MG BUP available increased nationwide in the initial pandemic period, likely driven by an increase in days supplied and MG BUP per claim, with a commensurate decrease in treatment disruptions.

We found that access to BUP, as represented by MG BUP per week, was higher than would be expected per the baseline trend early in the pandemic. However, by one year, MG BUP available returned to baseline trends. This can be seen in the fact that days supplied, and thus MG per prescription, was initially higher than expected; these trends also remained higher throughout the post-initial pandemic period. We also found that every patient group, regardless of variable studied, had an increase in MG BUP in the first week post-initial pandemic period.

While Nguyen et al. (Nguyen et al., 2021) and Jones et al. (Jones et al., 2021) found no significant changes in the number of individuals receiving BUP, we did find a small increase in the number of existing patients with active prescriptions at 1 and 26 weeks during the post-initial pandemic period. Unlike previous studies, we identified each week in which a patient had an active prescription, rather than each week that a patient refilled. Using refills instead of active prescriptions may underestimate total access, since prescriptions with longer days supplied will appear as fewer fills overall, even if number of weeks with BUP available remain the same. In terms of new patients, we found the same trend as Currie et al. (Currie et al., 2021) in the initial months after the pandemic declaration (an initial decrease followed by rebound). The initial increase in buprenorphine coverage among existing patients despite a decrease in new patients is likely explained by longer prescription lengths (Fig. 1B) and fewer treatment disruptions (Fig. 2). At 52 weeks however, we found an increase in the number of new patients receiving a buprenorphine prescription compared to the expected baseline trend.

There is mixed evidence as to whether gender influences initiation and discontinuation of buprenorphine in smaller samples (Hadland et al., 2017; Samples et al., 2018). There has been evidence that women have been disproportionately affected by the COVID-19 pandemic in the workforce. For instance, as of December 2019 women accounted for just over 50% of the US workforce (Law, 2020) but 2.5 million more women than men lost their jobs between February and May 2020 (Kochhar, 2020). Our study found a small, differential impact of gender on decreased MG BUP available.

This study also found decreased access to BUP with younger cohorts, particularly age 18–29, by 52 weeks, despite all age groups initially having increased medication available. Our findings support a known trend that young adults do not receive medication treatment for their opioid use disorder at the same rates as older adults (Borodovsky et al., 2018), possibly related to stigma, decreased health care utilization, and limited workforce capacity (Adams et al., 2022).

Regarding payor type, patients with Medicaid retained access and saw an increase in MG BUP throughout the study period, while patients with all other payor types had a decrease in MG BUP. A study by Cantor et al. (Cantor et al., 2021) found that BUP access remained steady among a cohort of patients with commercial insurance, however this study followed patients until August 2020 and our analysis found a decrease in total MG available at 6 months (October 2020). This trend can perhaps be explained by job losses among commercially-insured patients at a population level. Nguyen et al. (Nguyen et al., 2021) found that prescriptions paid for by cash decreased while Medicaid prescriptions were unchanged, however the time period for this study ended June 2020. Our study did find similar changes over the one-year post-pandemic period by insurance type and, in particular, retained and improved access for patients receiving Medicaid.

Our study also found a marked decrease in treatment disruptions among people who were stable in treatment for at least 6 months, regardless of the definition of treatment disruption (a gap of 7, 14, and 28 days). We also found that even prior to the pandemic, rates of treatment disruption were low (<1.5%) among this cohort of stably-treated patients. We do note some non-linearity in treatment disruptions, particularly around the holiday season and new year. There is no definition of minimum treatment time for BUP, however the National Quality Forum suggests at least 6 months (Samples et al., 2020). Our findings related to treatment disruptions support 6 months as a possible benchmark for demonstrating clinical stability.

Prior research has demonstrated variability in prescribing based on provider specialty (Hicks et al., 2015) and primary care physicians account for the highest percentage of active BUP prescribers (Foti et al., 2021; Hawk et al., 2021; Stein et al., 2021; Wen et al., 2019). This study found that Addiction Medicine physicians consistently maintained higher rates of BUP in the one-year post-initial pandemic, while total MG prescribed by General Medicine physicians decreased by 52 weeks compared to the expected baseline trend. We also found a near 3-fold increase in total available buprenorphine prescribed by APPs during our study interval, consistent with trends previously seen starting in 2017 since the passage of the Comprehensive Addiction and Recovery Act of 2016 (Stein et al., 2021).

Strengths of this study include the long follow-up period in which we were able to track patterns in filled prescriptions. The nationwide dataset allows for an understanding of overall trends at a higher level and the impact of federal policies. In addition, this dataset has millions of prescriptions, resulting in a more precise estimate of differences over time. We used a unique outcome, MG of BUP available per week, to account for differences in prescribing practices like longer prescription lengths leading to fewer prescriptions overall.

4.1. Limitations

Our dataset lacked the granular, patient-level characteristics needed to better understand differences by race, ethnicity, or income. Given this is a prescription database, we are unable to demonstrate any association of BUP prescription and patient outcomes. Thus, we cannot report changes in opioid-related morbidity and mortality as a result of prescribing practices. We also do not have a control group, which would not have been feasible given the pandemic (Chan et al., 2022). In addition, this was an exploratory analysis and therefore, some of our findings may be due to chance from running multiple models (Noble, 2009). We also did not control for time-varying covariates. Lastly, we were unable to accurately reflect access to methadone and naltrexone for the treatment of opioid use disorder, as this database does not contain dispensing information from opioid treatment programs, nor does it contain information regarding indication for naltrexone. This also meant we could not accurately reflect buprenorphine that may have been dispensed from an opioid treatment program, as opposed to prescribed (Stevens et al., 2022).

5. Conclusion

COVID-19-related disruptions in health care delivery, changes to regulations, and interruptions in insurance coverage have created a natural experiment to study the impact of policy on prescription fills. This study found that regulatory changes around buprenorphine prescribing facilitated patient access to buprenorphine during the pandemic through a combination of increasing MG of buprenorphine per prescription and days supplied, and that stably-treated patients saw a marked decrease in treatment disruptions. Efforts to ensure access to buprenorphine during the pandemic, including expansion of telemedicine, appear to overall have stabilized maintenance on effective, evidence-based treatment.

Contributors

The authors have no financial conflicts of interest to disclose and all contributed in preparing, drafting, and revising the manuscript.

Declaration of Competing Interest

The authors declare no financial conflicts of interest.

Acknowledgments

Role of Funding Source

Nothing declared

Funding

This work was supported by the National Institutes of Health 3UG1DA049436–03S1.

Footnotes

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.dadr.2023.100135.

Appendix. Supplementary materials

mmc1.docx (247.9KB, docx)

References

  1. Adams S.H., Schaub J.P., Nagata J., Park M.J., Brindis C.D., Irwin C.E. Young Adult Anxiety or Depressive Symptoms and Mental Health Service Utilization During the COVID-19 Pandemic. J. Adolesc. Heal. 2022 doi: 10.1016/j.jadohealth.2022.02.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Beard E., Marsden J., Brown J., Tombor I., Stapleton J., Michie S., West R. Understanding and using time series analyses in addiction research. Addiction. 2019;114:1866–1884. doi: 10.1111/add.14643. [DOI] [PubMed] [Google Scholar]
  3. Borodovsky J.T., Levy S., Fishman M., Marsch L.A. Buprenorphine Treatment for Adolescents and Young Adults With Opioid Use Disorders: A Narrative Review. J. Addict. Med. 2018;12:170–183. doi: 10.1097/ADM.0000000000000388. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Cantor J., Dick A.W., Haffajee R., Pera M.F., Bravata D.M., Stein B.D., Whaley C.M. Use of buprenorphine for those with employer-sponsored insurance during the initial phase of the COVID-19 pandemic. J. Subst. Abuse Treat. 2021;129 doi: 10.1016/j.jsat.2021.108384. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Centers for Disease Control and Prevention Drug Overdose Deaths in the U.S. Top 100,000 Annually [WWW Document] Natl. Cent. Heal. Stat. 2021 https://www.cdc.gov/nchs/pressroom/nchs_press_releases/2021/20211117.htm URL. accessed 3.18.22. [Google Scholar]
  6. Chan G.C.K., Lim C., Sun T., Stjepanovic D., Connor J., Hall W., Leung J. Causal inference with observational data in addiction research. Addiction. 2022;117:2736–2744. doi: 10.1111/add.15972. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cox C., Kamal R., McDermott D. How have healthcare utilization and spending changed so far during the coronavirus pandemic? [WWW Document] Peterson-KFF Heal. Syst. Tracker. 2020 https://www.healthsystemtracker.org/chart-collection/how-have-healthcare-utilization-and-spending-changed-so-far-during-the-coronavirus-pandemic/#item-costs-use-covid_estimated-monthly-percent-change-in-average-per-facility-revenues-from-2019-to-2020-by-f URL. (accessed 7.12.20) [Google Scholar]
  8. Cremer L.J., Board A., Guy G.P., Schieber L., Asher A., Parker E.M. Trends in pharmacy-based dispensing of buprenorphine, extended-release naltrexone, and naloxone during the COVID-19 pandemic by age and sex – United States, March 2019 – December 2020. Drug Alcohol Depend. 2022;232 doi: 10.1016/j.drugalcdep.2021.109192. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Currie J.M., Schnell M.K., Schwandt H., Zhang J. Prescribing of Opioid Analgesics and Buprenorphine for Opioid Use Disorder During the COVID-19 Pandemic. JAMA Netw. Open. 2021;4 doi: 10.1001/jamanetworkopen.2021.6147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. 110TH CONGRESS, 2008. Ryan Haight Online 3 Pharmacy Consumer Protection Act of 2008. [PubMed]
  11. Drug Enforcement Agency, 2020a. DEA's response to COVID-19 [WWW Document]. URL https://www.dea.gov/press-releases/2020/03/20/deas-response-covid-19 (accessed 6.18.20).
  12. Drug Enforcement Agency DEA SAMHSA Buprenorphine Telemedicine [WWW Document] US Dep. Justice. 2020 https://www.samhsa.gov/sites/default/files/dea-samhsa-buprenorphine-telemedicine.pdf URL. accessed 4.20.20. [Google Scholar]
  13. Ellison A. COVID-19 threatens to close physician offices nationwide [WWW Document]. Becker's. Hosp. Rev. 2020 https://www.beckershospitalreview.com/hospital-physician-relationships/covid-19-threatens-to-close-physician-offices-nationwide.html URL. accessed 7.12.20. [Google Scholar]
  14. Feder K.A., Krawczyk N., Mojtabai R., Crum R.M., Kirk G., Mehta S.H. Health insurance coverage is associated with access to substance use treatment among individuals with injection drug use: Evidence from a 12-year prospective study. J. Subst. Abuse Treat. 2019;96:75–81. doi: 10.1016/j.jsat.2018.08.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fiscella K., Wakeman S.E., Beletsky L. Buprenorphine Deregulation and Mainstreaming Treatment for Opioid Use Disorder: X the X Waiver. JAMA Psychiatry. 2018 doi: 10.1001/jamapsychiatry.2018.3685. [DOI] [PubMed] [Google Scholar]
  16. Fornili K.S., Fogger S.A. Nurse Practitioner Prescriptive Authority for Buprenorphine. J. Addict. Nurs. 2017;28:43–48. doi: 10.1097/JAN.0000000000000160. [DOI] [PubMed] [Google Scholar]
  17. Foti K., Heyward J., Tajanlangit M., Meek K., Jones C., Kolodny A., Alexander G.C. Primary care physicians’ preparedness to treat opioid use disorder in the United States: A cross-sectional survey. Drug Alcohol Depend. 2021;225 doi: 10.1016/j.drugalcdep.2021.108811. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Friedman J., Beletsky L., Schriger D.L. Overdose-Related Cardiac Arrests Observed by Emergency Medical Services During the US COVID-19 Epidemic. JAMA Psychiatry. 2020 doi: 10.1001/jamapsychiatry.2020.4218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hadland S.E., Wharam J.F., Schuster M.A., Zhang F., Samet J.H., Larochelle M.R. Trends in Receipt of Buprenorphine and Naltrexone for Opioid Use Disorder Among Adolescents and Young Adults, 2001-2014. JAMA Pediatr. 2017;171:747. doi: 10.1001/jamapediatrics.2017.0745. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Harris M., Johnson S., Mackin S., Saitz R., Walley A.Y., Taylor J.L. Low Barrier Tele-Buprenorphine in the Time of COVID-19. J. Addict. Med. Publish Ah. 2020 doi: 10.1097/ADM.0000000000000682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hawk K., Hoppe J., Ketcham E., LaPietra A., Moulin A., Nelson L., Schwarz E., Shahid S., Stader D., Wilson M.P., D'Onofrio G. Consensus Recommendations on the Treatment of Opioid Use Disorder in the Emergency Department. Ann. Emerg. Med. 2021 doi: 10.1016/j.annemergmed.2021.04.023. [DOI] [PubMed] [Google Scholar]
  22. Hicks L.A., Bartoces M.G., Roberts R.M., Suda K.J., Hunkler R.J., Taylor T.H., Schrag S.J. US Outpatient Antibiotic Prescribing Variation According to Geography, Patient Population, and Provider Specialty in 2011. Clin. Infect. Dis. 2015 doi: 10.1093/cid/civ076. [DOI] [PubMed] [Google Scholar]
  23. Jones C.M., Guy G.P., Board A. Comparing actual and forecasted numbers of unique patients dispensed select medications for opioid use disorder, opioid overdose reversal, and mental health, during the COVID-19 pandemic, United States, January 2019 to May 2020. Drug Alcohol Depend. 2021;219 doi: 10.1016/j.drugalcdep.2020.108486. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Joudrey P.J., Edelman E.J., Wang E.A. Drive Times to Opioid Treatment Programs in Urban and Rural Counties in 5 US States. JAMA. 2019;322:1310–1312. doi: 10.1001/jama.2019.12562. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kochhar R. Hispanic women, immigrants, young adults, those with less education hit hardest by COVID-19 job losses [WWW Document] Pew Res. Cent. 2020 https://www.pewresearch.org/fact-tank/2020/06/09/hispanic-women-immigrants-young-adults-those-with-less-education-hit-hardest-by-covid-19-job-losses/ URL. accessed 5.5.22. [Google Scholar]
  26. Lacktman N.M. Prescribing controlled substances without an in-person exam: The practice of telemedicine under the Ryan Haight Act [WWW Document] Becker's Heal. IT. 2017 https://www.beckershospitalreview.com/healthcare-information-technology/prescribing-controlled-substances-without-an-in-person-exam-the-practice-of-telemedicine-under-the-ryan-haight-act.html URL. accessed 6.21.20. [Google Scholar]
  27. Larochelle M.R., Bernson D., Land T., Stopka T.J., Wang N., Xuan Z., Bagley S.M., Liebschutz J.M., Walley A.Y. Medication for opioid use disorder after nonfatal opioid overdose and association with mortality: A cohort study. Ann. Intern. Med. 2018;169:137–145. doi: 10.7326/M17-3107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Law T. Women Are Now the Majority of the U.S. Workforce — But Working Women Still Face Serious Challenges [WWW Document] Time. 2020 [Google Scholar]
  29. Madgavkar, A., White, O., Krishnan, M., Mahajan, D., Azcue, X., 2020. COVID-19 and gender equality: Countering the regressive effects [WWW Document]. McKinsey Glob. Inst. URL https://www.mckinsey.com/featured-insights/future-of-work/covid-19-and-gender-equality-countering-the-regressive-effects#
  30. Mancher, M., Leshner, A.I. (Eds.), 2019. Medications for Opioid Use Disorder Save Lives, in: Medications For Opioid Use Disorder Save Lives. Washington (DC). https://doi.org/10.17226/25310
  31. Mehrotra A., Ray K., Brockmeyer D.M., Barnett M.L., Bender J.A. Rapidly Converting to “Virtual Practices”: Outpatient Care in the Era of Covid-19. NEJM Catal. Innov. Care Deliv. 2020 [Google Scholar]
  32. Nguyen T.D., Gupta S., Ziedan E., Simon K.I., Alexander G.C., Saloner B., Stein B.D. Assessment of Filled Buprenorphine Prescriptions for Opioid Use Disorder During the Coronavirus Disease 2019 Pandemic. JAMA Intern. Med. 2021;181:562. doi: 10.1001/jamainternmed.2020.7497. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Noble W.S. How does multiple testing correction work? Nat. Biotechnol. 2009;27:1135–1137. doi: 10.1038/nbt1209-1135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Olfson M., Zhang V, Schoenbaum M., King M. Trends in Buprenorphine Treatment in the United States, 2009-2018. JAMA. 2020;323:276. doi: 10.1001/jama.2019.18913. (Shu) [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Richert T., Johnson B. Long-term self-treatment with methadone or buprenorphine as a response to barriers to opioid substitution treatment: the case of Sweden. Harm. Reduct. J. 2015;12:1. doi: 10.1186/s12954-015-0037-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Samples H., Williams A.R., Crystal S., Olfson M. Impact Of Long-Term Buprenorphine Treatment On Adverse Health Care Outcomes In Medicaid. Health Aff. 2020;39:747–755. doi: 10.1377/hlthaff.2019.01085. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Samples H., Williams A.R., Olfson M., Crystal S. Risk factors for discontinuation of buprenorphine treatment for opioid use disorders in a multi-state sample of Medicaid enrollees. J. Subst. Abuse Treat. 2018;95:9–17. doi: 10.1016/j.jsat.2018.09.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sarac B.A., Schoenbrunner A.R., Wilson S.C., Chiu E.S., Janis J.E. Coronavirus Disease 2019 State Guidelines on Elective Surgery. Plast. Reconstr. Surg. - Glob. Open Publish Ah. 2020 doi: 10.1097/GOX.0000000000002904. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sharma A., Kelly S.M., Mitchell S.G., Gryczynski J., O'Grady K.E., Schwartz R.P. Update on Barriers to Pharmacotherapy for Opioid Use Disorders. Curr. Psychiatry Rep. 2017;19:35. doi: 10.1007/s11920-017-0783-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Shore, J., n.d. Ryan Haight Online Pharmacy Consumer Protection Act of 2008 [WWW Document]. Am. Psychiatr. Assoc.. URL https://www.psychiatry.org/psychiatrists/practice/telepsychiatry/toolkit/ryan-haight-act (accessed 6.21.20).
  41. Stein B.D., Saloner B., Schuler M.S., Gurvey J., Sorbero M., Gordon A.J. Concentration of Patient Care Among Buprenorphine-Prescribing Clinicians in the US. JAMA. 2021;325:2206. doi: 10.1001/jama.2021.4469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Stevens M.A., Tsai J., Savitz S.T., Nath B., Melnick E.R., D'Onofrio G., Jeffery M.M. Trends and Disparities in Access to Buprenorphine Treatment Following an Opioid-Related Emergency Department Visit Among an Insured Cohort, 2014-2020. JAMA Netw. Open. 2022;5 doi: 10.1001/jamanetworkopen.2022.15287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Trump, D.J., 2020. Proclamation on Declaring a National Emergency Concerning the Novel Coronavirus Disease (COVID-19) Outbreak [WWW Document]. URL http://www.newamsterdampgh.com/#home-section (accessed 6.28.20).
  44. Volkow N.D. Collision of the COVID-19 and Addiction Epidemics. Ann. Intern. Med. 2020;173:61–62. doi: 10.7326/M20-1212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. von Elm E., Altman D.G., Egger M., Pocock S.J., Gøtzsche P.C., Vandenbroucke J.P. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for reporting observational studies. Int. J. Surg. 2014;12:1495–1499. doi: 10.1016/j.ijsu.2014.07.013. [DOI] [PubMed] [Google Scholar]
  46. Wagner A.K., Soumerai S.B., Zhang F., Ross-Degnan D. Segmented regression analysis of interrupted time series studies in medication use research. J. Clin. Pharm. Ther. 2002;27:299–309. doi: 10.1046/j.1365-2710.2002.00430.x. [DOI] [PubMed] [Google Scholar]
  47. Wakeman S.E., Rich J.D. Barriers to Medications for Addiction Treatment: How Stigma Kills. Subst. Use Misuse. 2018;53:330–333. doi: 10.1080/10826084.2017.1363238. [DOI] [PubMed] [Google Scholar]
  48. Weintraub E., Greenblatt A.D., Chang J., Himelhoch S., Welsh C. Expanding access to buprenorphine treatment in rural areas with the use of telemedicine. Am. J. Addict. 2018;27:612–617. doi: 10.1111/ajad.12805. [DOI] [PubMed] [Google Scholar]
  49. Wen H., Borders T.F., Cummings J.R. Trends In Buprenorphine Prescribing By Physician Specialty. Health Aff. 2019;38:24–28. doi: 10.1377/hlthaff.2018.05145. [DOI] [PubMed] [Google Scholar]
  50. Zhang F., Wagner A.K., Soumerai S.B., Ross-Degnan D. Methods for estimating confidence intervals in interrupted time series analyses of health interventions. J. Clin. Epidemiol. 2009;62:143–148. doi: 10.1016/j.jclinepi.2008.08.007. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

mmc1.docx (247.9KB, docx)

Articles from Drug and Alcohol Dependence Reports are provided here courtesy of Elsevier

RESOURCES