Literature Collection

IMPORTANCE: Guidelines recommend that adult patients receive screening for alcohol and drug use during primary care visits, but the adoption of screening in routine practice remains low. Clinics frequently struggle to choose a screening approach that is best suited to their resources, workflows, and patient populations. OBJECTIVE: To evaluate how to best implement electronic health record (EHR)-integrated screening for substance use by comparing commonly used screening methods and examining their association with implementation outcomes. DESIGN, SETTING, AND PARTICIPANTS: This article presents the outcomes of phases 3 and 4 of a 4-phase quality improvement, implementation feasibility study in which researchers worked with stakeholders at 6 primary care clinics in 2 large urban academic health care systems to define and implement their optimal screening approach. Site A was located in New York City and comprised 2 clinics, and site B was located in Boston, Massachusetts, and comprised 4 clinics. Clinics initiated screening between January 2017 and October 2018, and 93 114 patients were eligible for screening for alcohol and drug use. Data used in the analysis were collected between January 2017 and October 2019, and analysis was performed from July 13, 2018, to March 23, 2021. INTERVENTIONS: Clinics integrated validated screening questions and a brief counseling script into the EHR, with implementation supported by the use of clinical champions (ie, clinicians who advocate for change, motivate others, and use their expertise to facilitate the adoption of an intervention) and the training of clinic staff. Clinics varied in their screening approaches, including the type of visit targeted for screening (any visit vs annual examinations only), the mode of administration (staff-administered vs self-administered by the patient), and the extent to which they used practice facilitation and EHR usability testing. MAIN OUTCOMES AND MEASURES: Data from the EHRs were extracted quarterly for 12 months to measure implementation outcomes. The primary outcome was screening rate for alcohol and drug use. Secondary outcomes were the prevalence of unhealthy alcohol and drug use detected via screening, and clinician adoption of a brief counseling script. RESULTS: Patients of the 6 clinics had a mean (SD) age ranging from 48.9 (17.3) years at clinic B2 to 59.1 (16.7) years at clinic B3, were predominantly female (52.4% at clinic A1 to 64.6% at clinic A2), and were English speaking. Racial diversity varied by location. Of the 93,114 patients with primary care visits, 71.8% received screening for alcohol use, and 70.5% received screening for drug use. Screening at any visit (implemented at site A) in comparison with screening at annual examinations only (implemented at site B) was associated with higher screening rates for alcohol use (90.3%-94.7% vs 24.2%-72.0%, respectively) and drug use (89.6%-93.9% vs 24.6%-69.8%). The 5 clinics that used a self-administered screening approach had a higher detection rate for moderate- to high-risk alcohol use (14.7%-36.6%) compared with the 1 clinic that used a staff-administered screening approach (1.6%). The detection of moderate- to high-risk drug use was low across all clinics (0.5%-1.0%). Clinics with more robust practice facilitation and EHR usability testing had somewhat greater adoption of the counseling script for patients with moderate-high risk alcohol or drug use (1.4%-12.5% vs 0.1%-1.1%). CONCLUSIONS AND RELEVANCE: In this quality improvement study, EHR-integrated screening was feasible to implement in all clinics and unhealthy alcohol use was detected more frequently when self-administered screening was used at any primary care visit. The detection of drug use was low at all clinics, as was clinician adoption of counseling. These findings can be used to inform the decision-making of health care systems that are seeking to implement screening for substance use. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT02963948.

OBJECTIVE: Buprenorphine is a highly effective treatment for opioid use disorders, but its continuation in the perioperative setting remains controversial, unlike the accepted practice of perioperative methadone continuation. METHODS: We conducted a retrospective cohort study from 2006 to 2014 comparing post-cesarean section opioid analgesic requirements of women with opioid use disorders treated with methadone or buprenorphine. Preoperative, intraoperative, and postoperative opioid requirements (morphine equivalent dose [MED]), postoperative complications, and length of stay were compared between the methadone and buprenorphine groups. RESULTS: During the 9-year study period, there were 185 women treated with methadone (mean dose 93.7 mg, SD 2.6) and 88 women treated with buprenorphine (mean dose 16.1 mg, SD 7.8). There were no statistically significant differences in MED requirements in the methadone versus buprenorphine groups: preoperative MED (11.4 mg [SD 31.5] vs 20.0 mg [SD 15.1]; mean difference [MD] 8.6, 95% confidence interval [CI] -1.9, 19.1), intraoperative MED (3.5 mg [SD 6.6] vs 5.2 mg [SD 13.7]; MD 1.8, 95% CI -1.1, 4.6), and postoperative MED during hospitalization (97.7 mg [SD 65.6] vs 85.1 mg [SD 73.0]; MD -12.6, 95% CI -31.1, 5.8). There were no statistically significant differences in postoperative complications or length of stay. CONCLUSIONS: Our study suggests that buprenorphine treatment will not interfere more than methadone with pain management after a cesarean section with no significant differences in opioid analgesic requirements, postoperative complications, or length of hospital stay. Future studies should investigate the generalizability to other surgeries.

INTRODUCTION: Academic detailing, an educational outreach that promotes evidence-based practices to improve the quality of care for patients, has primarily been delivered using one-on-one in-person interactions. In 2018, the U.S. Department of Veterans Affairs (VA) Pharmacy Benefits Management implemented a pilot virtual academic detailing program to increase naloxone prescribing among veterans at risk for opioid overdose or death. The aim of this evaluation was to compare virtual and in-person academic detailing on naloxone prescribing rates at VA. METHODS: A retrospective quasi-experimental pretest-posttest non-equivalent groups design was used to compare virtual academic detailing and in-person academic detailing on naloxone prescribing rates 12 months before and after providers received a naloxone-specific encounter at three VA regional networks between January 1, 2018 to May 31, 2020. Subgroup analysis was performed on rural providers. Generalized estimating equation models were constructed to compare the difference in naloxone prescribing rates before and after receiving virtual or in-person academic detailing controlling for provider-level characteristics. RESULTS: Providers who received virtual (N = 67) or in-person (N = 186) academic detailing had significant increases in naloxone prescribing, but the differences in the naloxone rates between the groups were not statistically significant (difference in changes in naloxone rates=+0.63; 95% CI: -2.23, 3.48). Similar findings were reported for rural providers. DISCUSSION: Providers who received naloxone-related in-person or virtual academic detailing had increased naloxone prescribing rates; however, there were no differences between the two types of modalities. Virtual academic detailing is a viable alternative for delivering academic detailing and allows academic detailers to expand their reach to rural providers.