Poster Thumbnail
Students Show Increased Preparedness without Improved Attitudes Toward People who Use Drugs

Jasmine Landry – Medical Student, Albany Medical College

Leah Miller-Lloyd – Medical Student, Albany Medical College

Sara Silberstein – Student, Albany Medical College

Kevin Flatley – Medical Student, Albany Medical College

Lisa Campo-Engelstein, PhD – Associate Professor, Albany Medical College

Introduction: People who use drugs (PWUD) experience many barriers to receiving adequate healthcare, including judgment by clinicians (Miller-Lloyd 2018). Clinicians have a lower regard for patients with substance use disorder (SUD) compared to patients with other chronic diseases such as diabetes. Physicians in training tend to develop more stigmatizing attitudes toward PWUD later in their training, and their satisfaction in treating this population decreases (Avery 2017). However, physicians who feel more prepared to treat SUD are more likely to have favorable attitudes toward PWUD and provide superior clinical care (Wakeman 2016). Thus, we seek to evaluate perceived preparedness and attitudes among medical students. We hypothesize that medical students will demonstrate increased preparedness later in their medical curriculum but that they will have more negative attitudes toward people who use drugs. Methods: All medical students at Albany Medical College (n=560) were recruited to take the Drug Problems Perceptions Questionnaire (DDPPQ), a survey designed to evaluate providers’ attitudes towards working with patients who use drugs (Watson 2007). The 20-item survey is scored on a 7-point Likert scale and was administered via an anonymous Qualtrics web link. Preclinical (years 1 and 2) and clinical (years 3 and 4) student responses were compared for each survey item using a two-sample T test. Lower response scores represent a more positive change (toward “strongly agree”). Results: 354 students completed the survey in a 2-month data collection period. Response means for individual survey items were aggregated into 5 previously validated subscales (Watson 2007). Students in their clinical rotations (years 3 and 4) demonstrated significant increases in role adequacy (knowledge of drugs and drug-related problems, ability to advise patients about drugs and their effects, etc.) compared to students in the preclinical years (p<0.001). They also had significant, though smaller, increases in role support (ability to find mentors to answer questions about PWUD, etc.) compared to preclinical students (p<0.001). Role adequacy and role support are proxies for students’ perceived preparedness to work with patients who use drugs. There were no significant changes in attitudes toward patients who use drugs following clinical experience as assessed by these questions: “In general, one can get satisfaction from working with drug users” and “in general, it is rewarding to work with drug users” (p=0.17 and 0.38, respectively). There was a significant, but small, shift toward more “strongly agree” in response to “In general, I feel I can understand drug users” and “I feel I am able to work with drug users as well as other client groups” (p=0.02 for each). Conclusions: Medical students feel more prepared to work with people who use drugs following their clinical experiences. However, students show no improvements in attitudes towards this population as they progress in their medical education. As previous research demonstrates that medical trainees develop more stigmatizing attitudes through residency, interventions are needed to establish and reinforce positive attitudes in students prior to and throughout residency.


  View Poster
Poster Thumbnail
Naltrexone implants improve therapeutic adherence with multiple substance use disorder

Guillermo Arechiga, MD – anestesiologist, Hospital General de Occidente

Jorge Bonilla, MD – Anesthesiologist, Hospital General de Occidente

Irene Mendoza, PhD – Quemist, Instituto Politecnico Nacional

Jose Correa, PhD – Researcher, Instituto Politecnico Nacional

Yamileth Ruiz, MD, MNSc, MPH, MEd – Clinical research director, Mexico Compounding

Background. According to the NSDUH, 20.1 million American adults (12 and older) struggled with a substance use disorder (SUD) in 2017. About 1 in 9 people who had SUDs in the past year had both an alcohol use disorder and an illicit drug use disorder – representing approximately 2.3 million people.1 In community settings, multiple substance abusers seeking treatment has become the rule and far from the exception2. Oral Naltrexone is a front-line treatment for alcohol and opioid use disorder, but its efficacy is limited by poor adherence and relapse3,4,5. The clinical interest in extended-release formulations of opioid antagonists, such as naltrexone (NTX) implants, has the potential to increase abstinence success rates once medication assisted therapy is discontinued. Objective: To examine the therapeutic adherence of multiple-substance users with and without NTX implants in combination with conventional treatment. Material and methods: Randomized controlled trial with a total of 54 subjects diagnosed with addiction to opioids, alcohol and methamphetamines. Subjects were divided into 2 groups: group A patients received 1,200 mg of NTX implants plus conventional therapy; group B patients received only conventional therapy, each for 90 days. Cravings were assessed with a modified Penn and Alcohol Craving Scale (PACS); depression and anxiety were evaluated with the Hamilton rating scale. Results: All patients in group A completed the 90-day treatment period, while 62.90% of the patients in group B left the study after 30 days. NTX and 6-β-naltrexol median concentrations detected after 90 days of treatment were 58.12 ng/ml and 70 ng/ml, respectively. Methamphetamines were only detected in 1 of 13 patients who initially tested positive for methamphetamine use in group A, and in 1 of the remaining subjects in group B at the end of the trial. Opioids were not detected in 5/5 patients who mentioned consuming them in group A at the end of the trial. Of the patients who mentioned consuming opioids in group B (n=10), 2 patients who remained in the trail tested negative at 4 weeks, and 1 patient tested negative at 8 weeks, and all subjects remained abstinent from alcohol in group A. Only one patient in group A had elevated levels of liver enzymes at the end of the study; although, this represented a decrease of 42.56% from their baseline levels. PACS values were significantly decreased in group A patients following treatment (p < 0.05). HAM-D and HAM-A scores also improved for both groups. Conclusion: NTX significantly decreased cravings in patients diagnosed with multiple substance use disorder and promotes abstinence without elevated liver enzymes and has a beneficial effect on depression, thereby supporting its use as a complementary tool for cognitive behavioral therapy.


  View Poster
Poster Thumbnail
Prenatal Opioid Use: Characterization of Patients Presenting to Two Urban Prenatal Clinics

Victoria H. Coleman-Cowger, PhD – Senior Clinical Research Manager, The Emmes Corporation

Erica Peters, PhD – Senior Research Scientist, Battelle

Emmanuel Oga, MD, MPH – Epidemiologist, RTI International

Caroline M. Mulatya, PhD – Biostatistician, The Emmes Corporation

Katrina Mark, MD – Assistant Professor, University of Maryland School of Medicine

Background and Purpose: Opioid use during pregnancy has risen dramatically in recent years, mirroring increasing rates observed in the general population. As a result, there has been a sharp increase in neonatal abstinence syndrome (NAS). The rate of NAS in Maryland rose from 6.2 in 2004 to 11.4 per 1,000 births in 2012. Nationally, rates of opioid use disorder at delivery hospitalization more than quadrupled during 1999–2014, to 6.5 per 1000 births in 2014. The purpose of the current analysis is to characterize pregnant women who use opioids and are at risk of giving birth to babies with NAS and other negative birth outcomes. Method: A convenience sample of 500 pregnant women presenting to two urban Maryland prenatal clinical settings was enrolled into a study examining prenatal substance use screening tools between January 2017 and January 2018. Participants consented to urine drug testing and completed demographic and substance use questionnaires. We provide a descriptive characterization of women who tested positive for opioids, including type of opioids used, other substances used, demographic characteristics, and birth outcomes. Results: Of the 500 study participants, 5.6% (n=28) screened positive for opioids from urine testing (n=20) and/or self-report (n=13). The majority of the subsample screening positive for opioids was African-American, in the first trimester of pregnancy, and identified from the prenatal clinical setting serving a Medicaid-eligible population rather than the one serving a privately-insured population. Average age of this participant subset was 27 years. Specific opioids identified in urine were: methadone, oxycodone, buprenorphine, codeine, and propoxyphene. Polysubstance use was detected in approximately 21% of the subsample, primarily marijuana. Negative birth outcomes, including preterm birth, low birthweight, and birth defects, were present in 39% of the subsample. NAS was noted in the electronic medical records of 18% of the subsample. Conclusions: Opioid use in pregnancy is higher in our sample than in the general population (5.6% vs. 1.3% - National Survey on Drug Use and Health, 2017), and negative birth outcomes occurred with high frequency. While it is important to distinguish between opioids often utilized as part of medication-assisted treatment (e.g., methadone and buprenorphine), prescribed opioids for pain, and illicit opioids in order to conduct an individualized risk/benefit analysis of continued opioid use during pregnancy, it is equally important to understand the implications of opioid use in any form. Universal screening during prenatal care is recommended to identify prenatal opioid use and provide appropriate maternal supports.


  View Poster
Poster Thumbnail
Public libraries as partners in confronting the opioid crisis

Margaret Lowenstein, MD, MPhil – Fellow, National Clinician Scholars Program, University of Pennsylvania Perelman School of Medicine

Rachel Feuerstein-Simon, MPA, MPH – Research Program Manager, University of Pennsylvania Perelman School of Medicine

Risha Sheni, BS – Research Assistant, University of Pennsylvania Perelman School of Medicine

Roxanne Dupuis, MSPH – Graduate Student, Harvard University

Carolyn Cannuscio, ScD – Associate Professor, University of Pennsylvania Perelman School of Medicine

Background: Prior work has demonstrated that public libraries are well-positioned to address health and social issues. More recently, public libraries have been directly impacted by the opioid overdose crisis, including acting as first responders when patrons engage in substance use or even overdose on-site. In October 2018, one of the manufacturers of naloxone announced an initiative to make the drug available to each of the 9000 library systems in the U.S., but little is known about librarians’ attitudes or capacity to recognize and respond to overdoses. Our aim was to document variation in public library staffs perpsectives towards overdose reversal and opioid crisis response in public libraries. Methods: We conducted semi-structured interviews with public library staff who attended a national professional society meeting in March 2018. Interviews domains included attitudes, experiences, and needs related to substance use. We also inquired specifically about experiences with drug overdose and overdose response. Interview transcripts were analyzed in NVivo 11 using modified grounded theory to identify emergent themes. 10% of transcripts were doubled-coded, with high inter-rater reliability (mean kappa = 0.83). Results: We completed 45 interviews with library staff from 26 states and a range of urban, suburban, and rural locations. Participants were primarily female (89%), white (93%), between the ages of 51-60 (41%) and had completed a professional degree (85%). Most participants reported that their libraries had been directly affected by the opioid epidemic. Six participants (13%) reported that their libraries had experienced on-site drug overdoses, and many others had heard about overdoses in other libraries in their community or region. Participants were also aware of increasing calls for libraries to stock naloxone ad reverse overdoses, but only 3 (9%) had naloxone on-site at the time of the study. Participants expressed a range of attitudes regarding overdose response. Concerns included library staffs’ ability or training to respond to overdose, issues of legal and personal responsibility, and personal safety in overdose response. Library staffs also reported an actively evolving discussion about how overdose response fit into the broader mission of public libraries and the need to balance the commitment to serving vulnerable populations in active addiction with other library constituencies. Library staffs acknowledged the role that stigma towards individuals with substance use disorders played in these discussions. Despite barriers, many participants were open to engaging patrons with substance use disorders and desired greater support and partnerships with first responders, health and social service organizations to support these efforts. Conclusions: Public libraries are facing many of the consequences of the opioid epidemic, including overdose. Most library staff felt they had a role to play in addressing the opioid crisis in their library but expressed varying levels of comfort and preparedness to do so. Our findings suggest that there are opportunities for public health policymakers engage libraries as part of comprehensive, community-based strategies to address the opioid epidemic.

  View Poster
Poster Thumbnail
Buprenorphine Maintenance Is Associated With Increased Engagement With Prenatal Care

David Hartman, MD – Physician, Virginia Tech Carilion School of Medicine

Kimberly Simcox, D.O. – Physician, Carilion Clinic

Jennifer Wells, M.D. – Physician, Carilion Clinic

William Rea, M.D. – Associate Professor, Carilion Clinic

Objectives: Opioid replacement therapy (ORT) using either methadone or buprenorphine is a common treatment strategy in the management of opioid-dependent pregnant patients. While the comparative effects of each medication on neonatal outcomes are well studied, the effects on prenatal care engagement are less clear. The aim of this study was to quantify engagement with prenatal and other outpatient care for pregnant patients maintained on buprenorphine and methadone during pregnancy. Methods: Pregnant patients that were maintained on either buprenorphine or methadone were retrospectively identified and included. The number of prenatal care and other outpatient visits were quantified, and multivariate linear regression was used to control for length of gestation. Results: 230 eligible patients were identified, 172 maintained on buprenorphine and 58 patients maintained on methadone. There were no significant difference in length of gestation (p=.062) or maternal age at delivery (p=.200) between groups. Patients prescribed buprenorphine attended significantly more prenatal care visits (p=.010) than pregnant women prescribed methadone (7.2 vs 8.8), and attended significantly more (p=<.001) outpatient care visits in general (19.1 vs 25.7). In a multivariate model, length of gestation was the strongest predictor of number of prenatal care visits (p=<.001), but drug class remains a significant predictor of routine prenatal care visits (p=.036) Conclusions: Pregnant patients maintained on buprenorphine attended more prenatal care visits than those maintained on methadone. Differences in the way patients engage with ORT during pregnancy, dictated by the regulations around each medication, might have effects on compliance with prenatal and other medical care.


  View Poster
Poster Thumbnail
Implementation of a Mobile Buprenorphine Clinic

Ryan Dono, MD – Family Medicine Physician / Family Medicine Residency Core Faculty, Medical Director of Health Care for the Homeless Program, Greater Lawrence Family Health Center, Lawrence Family Medicine Residency

Morgan Younkin, MD, MPH – Family Medicine Resident, Greater Lawrence Family Health Center/Lawrence Family Medicine Residency

Rebecca Lee, MD – Family Medicine Resident, Greater Lawrence Family Health Center/Lawrence Family Medicine Residency

Amy Bositis, RN MSN – Director of Clinical Programs, Greater Lawrence Family Health Center

Christopher Bositis, MD – Clinical Program Director, HIV and Viral Hepatitis; Faculty in Family Medicine Residency Program, Greater Lawrence Family Health Center/Lawrence Family Medicine Residency

Introduction: In response to an outbreak of 129 new HIV infections from 2015 – 2018 in persons who inject drugs (PWIDs) in Lawrence and Lowell, Massachusetts [1], and a related need for expanded community access to Opioid Use Disorder (OUD) treatment, the Greater Lawrence Family Health Center (GLFHC) initiated a mobile buprenorphine clinic through an existing Health Care for the Homeless Program (HCHP). Mobile Medical Units (MMUs) have been previously demonstrated to more effectively reach patients from racial and ethnic minorities, those experiencing homelessness, current PWIDs, uninsured patients, and those with less treatment experience than fixed site treatment centers [2]. Methods: GLFHC’s HCHP, equipped with a MMU, pharmacy support, and a multidisciplinary team, initiated an innovative buprenorphine clinic offering on-demand and same-day buprenorphine induction 5 days per week. The MMU is equipped with phlebotomy, restroom, nursing capacity for medication and vaccine administration, a driver trained in syringe exchange and naloxone distribution, a medical assistant to assist in insurance application, and a family physician or nurse practitioner. Patients are seen without regard to insurance status or ability to pay, with medication fees waived for uninsured patients. Following an induction interview, buprenorphine is prescribed by a clinician, if appropriate, after which a pharmacy courier transports the medication from a fixed site pharmacy to the MMU, and an addiction-trained RN administers induction doses and ensures follow up for buprenorphine refills. Additional services provided include: harm reduction counseling, HIV pre-exposure prophylaxis (PrEP), and general primary care including Hepatitis A, Hepatitis B, Tdap and influenza vaccinations. A peer support group was also initiated. We analyzed patient descriptors, retention in treatment, and primary care services during the period from program initiation on July 3, 2019 to January 20, 2019. Results: 46 patients were induced on buprenorphine during the study period. 37/46 were experiencing homelessness, 3 of whom obtained housing during the study period. 2/46 (3%) patients were HIV positive, 25/46 (54%) had Hepatitis C infection, and 40/46 (87%) had a documented psychiatric diagnosis. 25/46 (54%) patients maintained an active buprenorphine prescription in our program, 28/46 (60%) continued buprenorphine at any area program including our own, and 3/46 (7%) transitioned to methadone treatment, for an overall OUD treatment retention rate of 67%. 11/25 (44%) active patients had no opioids aside from buprenorphine present on their most recent toxicology screen, and 18/25 (72%) demonstrated reduction in use by negative drug testing for opioids. 28/46 (61%) patients had documentation of PrEP eligibility during buprenorphine induction and 5/28 (18%) initiated PrEP, 1 of whom continued for longer than 1 month. 18/46 (39%) individuals received at least one vaccine, 12 receiving Hepatitis A vaccine. Conclusion: GLFHC’s mobile clinic has given a highly vulnerable population access to OUD treatment, harm reduction and primary care services, and retention rates thus far appear comparable to those reported in traditional primary care settings [3]. While continued efforts are needed to expand PrEP uptake and vaccination coverage, this model shows great promise for lowering the overall burden of addiction and its infectious comorbidities in our community.

  View Poster
Poster Thumbnail
Opioid prescribing trends in postpartum women: A multicenter study

Karissa Sanchez Traun, MD – Resident, Gundersen Health System

Charles W. Schauberger, MD, MS – Obstetrician/Gynecologist, Gundersen Health System

Luis D. Ramirez, MPH – Biostatistician, Gundersen Health System

Cresta W. Jones, MD, FACOG – Assistant Professor, University of Minnesota Medical School

Alisha Lindberg, MD – OB/GYN Resident, University Of Minnesota

Opioid prescribing trends in postpartum women: A multicenter study Karissa Sanchez Traun, MD1, Charles Schauberger, MD2, Luis Ramirez, MPH3, Cresta W. Jones, MD4, Alisha Lindberg, MD4, Ricardo Molero Bravo, MD5, Tricia Wright, MD MS5, Benjamin D. Traun, MD6, Suzanne E. Peterson, MD7, Vania Rudolph, MD7 1Department of Family Medicine, Gundersen Health System, La Crosse, Wisconsin 2Department of Obstetrics & Gynecology, Gundersen Health System, La Crosse, Wisconsin 3Department of Biostatistics, Gundersen Health System, La Crosse, Wisconsin 4Department of Obstetrics & Gynecology, University of Minnesota, Minneapolis, Minnesota 5Department of Obstetrics and Gynecology, University of Hawaii, Honolulu, Hawaii 6Department of Family Medicine and Community Health, University of Wisconsin, Madison, Wisconsin 7Department of Obstetrics and Gynecology, Swedish Medical Center, Seattle, Washington Objective: To evaluate inter-hospital variability in opioid prescribing patterns in postpartum women and to better understand the role of clinical variables in prescribing. Methods: Data was extracted from electronic medical records for 4248 patients who delivered at six hospitals across the United States from January to March 2016. The primary outcome of the study was postpartum opioid prescription at the time of hospital discharge. Additional variables included the specific opioid prescribed, dosage, number of tablets, age, parity, tobacco use, route of delivery, and lacerations. Statistical analysis was performed with SAS software. Statistical methods included chi-square to analyze frequency of opioid prescription by hospital, parity, tobacco use, delivery method, and laceration type. ANOVA was used to analyze morphine equivalent dose by hospital. Results: The percentage of women prescribed postpartum opioids varied significantly by hospital, ranging from 27.6% at Hospital A to 70.9% at Hospital E (p < 0.001). Median dose per tablet was 7.5 morphine milligram equivalents at five of the six hospitals. Median number of tablets prescribed ranged from 20 at Hospital E to 40 at Hospital F (p <0.0001). Primiparous women were more likely to receive opioids than multiparous women when broken down by a parity of 1, 2, 3, 4 and ≥5 (52.8% vs. 48.0% vs. 47.6% vs. 40.1%, vs. 45.8%, p 0.0005). Women who underwent cesarean sections were more likely to receive opioids than those who had vaginal deliveries (97.7% vs. 30.7%, p < 0.0001). Among women who had vaginal deliveries, opioid prescription rates were higher in women who experienced either a 2nd degree laceration (35.5%, p = 0.0002) or a 3rd/4th degree laceration (59.3%, p < 0.001). Conclusions: Postpartum opioid prescription rates vary widely among hospitals, but providers within the same hospital tend to follow similar prescribing trends. Parity, route of delivery, and lacerations seem to account for much of the differences in prescribing rates, but even when controlling for these factors, different hospitals prescribe varying dosages and quantities of opioids.


  View Poster
Poster Thumbnail
The Association of Loneliness and Relapse in Patients with Opioid Use Disorder

John McDonagh, MD, FASAM – Director of Addiction Medicine, Fair Haven Community Health Care

Benjamin Oldfield, MD, MHS – Medical Director of Population Health, Fair Haven Community Health Care

Cory Williams – Research Assistant, Fair Haven Community Health Care

Maya Spell – Substance Abuse Coordinator, Fair Haven Community Health Care

Douglas P. Olson, MD – Vice President of Clinical Affairs, Fair Haven Community Health Care

Background: Loneliness is defined as a perceived lack of social connection and has been associated with adverse physical and mental health outcomes. While loneliness has been associated with poorer outcomes in people with tobacco and alcohol use disorders, its role in those with opioid use disorder (OUD) is unclear This study sought to identify a potential association between loneliness and relapse in patients with OUD. Methods: Patients receiving buprenorphine for treatment of OUD at Fair Haven Community Health Care (FHCHC), an urban federally qualified health center in New Haven, Connecticut, were eligible for inclusion in the study. A convenience sample of patients were asked to complete the UCLA (Version 3) Loneliness Scale, which calculates a score from 20 to 80 based on the response of 20 items. All patients being treated with buprenorphine were eligible for inclusion into the study. Relapse was defined as any toxicology test positive for a non-prescribed opioid over 36 weeks following enrollment. Covariates included demographics, comorbid illness, and utilization variables. Multivariable logistic regression was performed to determine independent predictors of relapse. All study procedures and materials were approved by the FHCHC institutional review board. Results: From December 2017 through August 2018, 51 patients were enrolled in the study. Their mean age was 40. Among those enrolled, 17 (33%) were female, 27 (53%) were white, and 18 (35%) were Latino. Patients were prescribed a mean daily buprenorphine dose of 12 mg (SD 8) and 14 (27%) were concurrently prescribed antidepressants. Patients had a mean loneliness score of 39 (range 20 to 65, SD 13). During the study period 23 (45%) patients had inappropriate toxicology results. In unadjusted analyses, those who relapsed had a higher mean loneliness score (46, SD 11) than those who did not relapse (34, SD 10, p<0.001). In the multivariable regression model, a higher calculated loneliness score was associated with an increased odds of relapse (adjusted odds ratio 1.10, 95% confidence interval 1.01-1.20). Conclusion: Among those being treated with buprenorphine for OUD, loneliness may be associated with increased odds of relapse or treatment failure. Further observational data is needed to confirm this relationship.


  View Poster
Poster Thumbnail
18-month Safety and Efficacy of RBP-6000 Extended-release Buprenorphine for OUD

Anne Andorn, MD – Head Late Stage Development, Indivior Inc.

Brent Boyett, DMD – Chief Medical Officer, Pathway Healthcare

Sunita Shinde, MD – Clinical Research Physician, Indivior Inc.

Norma-Lynn Fox, PhD – Senior Clinical Scientist, Indivior Inc

Yue Zhao, DrPH – Director, Indivior Inc.

Introduction RBP-6000 is a once-monthly, extended-release, subcutaneous depot formulation of buprenorphine, approved by the FDA in 2017 for treatment of moderate-to-severe opioid use disorder (OUD).1,2 Retention in treatment is key to abstinence and other aspects of successful treatment of OUD; this report considers treatment with RBP-6000 for up to 18 months.3,4 Methods Data on long-term efficacy and safety come from 3 studies of adults with moderate-to-severe OUD: (1) Study 1 (NCT02357901): a 6-month, randomized, double-blind study of RBP-6000 (n=404) vs. placebo (n=100). There were 2 dosage groups of RBP-6000: 300 mg SC monthly, and 300 mg SC monthly for the first 2 injections followed by 100 mg SC monthly. (2) Study 2 (NCT02510014): an open-label study including 257 “rollover” subjects who completed Study 1 (6 months) and 412 “de novo” subjects (12 months). All subjects received a 300 mg dose of RBP-6000 followed by flexible dosing of either 300 mg or 100 mg monthly.. (3) Study 3 (NCT02896296): a 6-month, open-label, flexible dose study including 208 subjects who completed Study 2. All subjects received RBP-6000 monthly, as in Study 2. Eligible subjects entered an open-label phase with buprenorphine/naloxone sublingual film prior to initiating treatment in Studies 1 and 2 (including rollovers) with RBP-6000. Retention was based on time to discontinuation. Efficacy was based on urine samples negative for opioids, and additionally, by self-reports in Studies 1 and 2. Injection-site tolerability and pain was assessed by clinician grading and subject report on a 0 to 100 mm VAS scale. Safety assessments included treatment-emergent adverse events (TEAEs) and clinical safety assessments. Results Retention at 6 and 12 months was similar for subjects initially treated with RBP-6000 in double-blind Study 1 who then rolled over to open-label Study 2 (66% at 6 months, 51.4% at 12 months) and de novo subjects who initiated treatment in Study 2 (65.5% at 6 months, 51% at 12 months). Of 208 subjects enrolled, 166 (79.8%) completed Study 3. Among subjects treated for 12 months in Studies 1 and 2 (rollover), 107/174 (61.5%) were abstinent based on UDS and self-report at 12 months; among subjects treated for 12 months in Study 2 (de novo), 157/207 (75.8%) were abstinent at 12 months. In Study 3, 142/157 (90.4%) were abstinent based on UDS at 18 months. The maximum reported intensity at any time point for each injection-site assessment was none or mild and worst mean VAS pain score at any post-injection time point in Studies 1 and 2 was 64.0. The mean VAS pain scores decreased over time from the 1-minute to 2-hour time points. No TEAE was reported in at least 5% of subjects who rolled over from Study 1 to Study 2. Overall no new safety signals were observed. Conclusions Subjects who continued to receive monthly RBP-6000 up to 18 months had high levels of retention in treatment and abstinence; no new safety signals emerged over time. These data support the use of RBP-6000 as a long-term treatment for OUD.

  View Poster
Poster Thumbnail
Effect of Concurrent Drug Use on Neonatal Outcomes Among Buprenorphine Exposed Pregnancies

Misty L. McDowell, MD – Maternal Fetal Medicine Fellow, Indiana University

James E. Slaven, Jr., MS, MA – Biostatistician III, Indiana University

Sara K. Quinney, PharmD, PhD – Assistant Professor, Indiana University

Anthony L. Shanks, MD – Maternal Fetal Medicine Fellowship Program Director, Indiana Universtiy

Tara D. Benjamin, MD, MS – Assistant Professor, Indiana University

Introduction: Buprenorphine (BUP) is the preferred medication for opioid maintenance therapy in pregnancy. It has been shown that a dose-response relationship exists between maternal BUP dose and neonatal abstinence syndrome (NAS) incidence. We sought to determine the impact of concurrent illicit drug use at delivery on these same neonatal outcomes in BUP exposed pregnancies. Methods: This is a retrospective cohort study of 257 singleton pregnancies with antenatal exposure to BUP from 2015-2018 at a single center. No major congenital anomalies occurred. BUP dose at delivery was divided into low (0-10mg), medium (11-20mg), and high (>20mg) dose groups. Each group was then subdivided into positive versus negative urine drug screen (UDS) at delivery. Correlations between delivery dose and neonatal outcomes of interest were analyzed. Student's t-tests were used when comparing continuous variables and Chi-square tests for categorical variables. Kruskal-Wallis tests were performed when appropriate. Results: A positive UDS at delivery was noted in 42 mothers: 6 (17%) in the low, 20 (13%) in the medium, and 16 (22%) in the high dose group. Opiates were most common (52%) followed by marijuana (33%). A statistically significant increase in NAS duration in the high dose group (p=0.06) and decrease in birthweight in the medium dose group (p=0.002) were noted in neonates of mothers with a positive UDS at delivery. No statistically significant difference was seen among other neonatal outcomes such as APGAR scores and gestational age at delivery by BUP dose and UDS status. Conclusion: Concurrent drug use among pregnant women receiving BUP for opioid maintenance therapy is common. Opioids were most frequently concurrently used. In neonates of mothers on BUP therapy, a positive UDS at delivery was associated with a lower birthweight in the medium dose group and longer NAS duration in the high dose group. Similar trends in birthweight and length of NAS treatment were noted in the other dose groups, but these did not reach statistical significance. Larger studies are needed to clarify these relationships, but these findings demonstrate the profound impact that concurrent drug use can have on neonatal outcomes in mothers on opiate maintenance therapy.


  View Poster
Poster Thumbnail
Nurse Practitioner and Physician Assistant Provision of MAT for Opioid Use Disorder

Cory Page, MPH, MPP – Project Coordinator, Behavioral Health Research Workforce Center, University of Michigan School of Public Health

Barbara Andraka-Christou, J.D., PhD – Assistant Professor, University of Central Florida

Jessica Buche, MPH, MA – Program Manager, Behavioral Health Workforce Research Center, University of Michigan School of Public Health

Dana Foney, PhD – Director, Data and Evaluation, The National Council for Behavioral Health

Angela Beck, PhD, MPH – Clinical Assistant Professor of Human Behavior and Health Education, University of Michigan School of Public Health

Background: The 2016 Comprehensive Addiction and Recovery Act (CARA) expanded addiction treatment for opioid use disorder (OUD) and authorized nurse practitioners and physician assistants to prescribe medication-assisted treatment (MAT) for OUD until 2021. Because nurse practitioners are more likely to serve rural and Medicaid-eligible populations than physicians, and physician assistants specialize in expanding physicians’ practice, authorizing both providers to prescribe MAT may increase access to OUD treatment. Limited information was known about nurse practitioners and physician assistant experiences in providing MAT. Methods: We disseminated an online Qualtrics™ survey from July to August 2018 to a random sample of 3,711 nurse practitioners and physician assistants. The sample included high-frequency providers, or those who practiced addiction medicine or addiction psychiatry, and low-frequency providers who were less likely to engage in MAT. We targeted nurse practitioners and physician assistants who received the federal buprenorphine prescribing waiver (n=653) those who prescibe extended-release naltrexone (i.e., Vivitrol®) according to a comprehensive list maintained by the drug manufacturer, Alkermes, Inc. (n=426). Survey themes included demographics, professional characteristics, practice settings, screening for substance use disorder, substance use disorder maintenance, MAT drug knowledge and usage, and treatment barriers. Results: We received 240 responses to the survey. Most providers screened 76-100% of their new patients for substance use disorders, but provided treatment for only 0-25% of patients. Nurse practitioners and physician assistants were more likely to rely on physical observation, patient self-reporting of drug use/cravings, and urinalysis when screening for OUD, rather than on standardized mental health and addiction screening tools. Respondents were confident in their ability to detect OUD in a patient, but less confident in their ability to treat it. Those surveyed were most familiar with oral buprenorphine, followed by methadone and oral naltrexone, and agreed that all three medications decrease cravings, rates of relapse, and rates of overdose. They also agreed that MAT should be combined with counseling and peer support, and that these supports increase MAT’s efficacy. Respondents were largely unfamiliar with newer formulations of MAT medications, such as injected or implanted buprenorphine. Conclusions: Nurse practitioners and physician assistants may increase patient access to MAT through implementation of policies aimed at strengthening education and training. Standardizing graduate program and professional training curricula to feature MAT and OUD treatment as a core requirement, incorporating CARA’s training requirements to equip graduating practitioners with a buprenorphine waiver, and implementing coursework to help identify and reduce stigmas are strategies for ensuring that workforces have the tools and training to effectively treat OUD. Expanding patient panels and reducing the amount of training required for a waiver could also incentivize nurse practitioners and physician assistants to apply for a buprenorphine waiver. Integrating behavioral health and medical provider settings through co-location or telehealth, as well as increasing students’ access to active peer MAT providers through telecommunication technology, would also improve access to services and providers’ competency in treating OUD.


  View Poster
Poster Thumbnail
Recovery from OUD Post Monthly Buprenorphine-XR Treatment: 12-Month Longitudinal Outcomes

Walter Ling, MD – Professor, UCLA David Geffen School of Medicine

Vijay R. Nadipelli, BPharm, MS – Head, Global Health Economics & Outcomes Research, Indivior Inc.

Naoko A. Ronquest, PhD – Director, Indivior Inc.

Arnie P. Aldridge, PhD – Research Economist, RTI International

Caitlyn T. Solem, PhD – Executive Director, Pharmerit International

Susan Learned, MD, PharmD, PhD – Senior Vice President, Indivior Inc.

Christian Heidbreder, PhD – Chief Scientific Officer, Indivior Inc.

Nicholas C. Peiper, PhD

Introduction: The RECOVER (Remission from Chronic Opioid Use: Studying Environmental and SocioEconomic Factors on Recovery, NCT03604861) study explores outcomes in people with opioid use disorder (OUD) who initiated their recovery journey through participation in a Phase III study evaluating buprenorphine extended-release monthly injection (BUP-XR). The objectives are to characterize opioid abstinence over a 24-month observational window; examine clinical, environmental, and socioeconomic factors associated with opioid abstinence and relapse; and understand the long-term effects of BUP-XR treatment on clinical, humanistic, and economic outcomes in OUD. This analysis focuses on cumulative and past-week opioid abstinence during the first 12 months of the observational window. Methods: Participants could enroll in RECOVER 28 days after completing or discontinuing participation in a BUP-XR Phase III study, comprising a double-blind, randomized, placebo-controlled trial (NCT02357901) and an open-label safety study (NCT02510014). Three data sources inform the RECOVER analyses: a self-administered assessment completed every 3 months; urine drug screen (UDS) completed every 3 months; and publicly available information obtained from multiple sources at various times. Self-assessments were used to collect data on past week, past month, and past 6-month opioid use. Cumulative opioid abstinence over 12 months and past-week point-prevalent abstinence by BUP-XR treatment duration groups were evaluated descriptively. Multivariate models were used to control for treatment duration, demographics, drug use history and treatment, time since last BUP-XR dose, state-level opioid treatment policy indexes, and pre-Phase III study BUP-XR treatment differences. Results: The RECOVER study includes 533 participants. Approximately half (50.8%) self-reported abstinence for 12-months following their last study injection; 20.3% reported receiving pharmacologic therapy for a substance use disorder during RECOVER. Past-week, self-reported opioid abstinence was noted on 68.0% of assessments. Longer BUP-XR treatment duration was associated with higher complete opioid abstinence over the 12-month period. Continuous 12-month abstinence was reported by 62.7% of participants receiving BUP-XR for 12 months compared with 32.1% of participants receiving BUP-XR for =2 months. After adjustment, 12-month compared with =2-month BUP-XR treatment duration remained a significant predictor of opioid abstinence over the 12-month period (75.3% vs 24.1%; P=0.001). Models of past week self-reported abstinence suggested that the adjusted likelihood of abstinence at 12 months was 82.1% (95%CI:72.7%-88.7%) for those who stayed on BUP-XR for 12 months, 70.8% (59.9%-79.7%) for 6-11 months, 48.8% (41.5%-56.2%) for 3-5 months, and 58.6% (44.5%-71.3%) for =2 months. When combining self-reported past-week abstinence with UDS outcomes, the adjusted likelihood of past week abstinence at 12 months for those who stayed on BUP-XR for 12 months was 62.4% (47.0%-75.7%), 6-11 months 49.8% (38.2%-61.4%), 3-5 months 36.9% (29.9%-44.5%), and =2 months 35.6% (24.6%-48.5%). Conclusion: Within the 12 months following RECOVER enrollment, half of participants self-reported complete abstinence, despite the low prevalence of additional pharmacologic OUD treatments. Longer BUP-XR treatment durations were associated with higher rates of opioid abstinence over 12-months and past-week opioid use, as measured by self-report and self-report plus UDS results. These findings provide important insights into understanding the role of novel pharmacologic treatments in supporting a patient’s recovery.

  View Poster
Poster Thumbnail
Computable phenotype for opioid misuse using EHR data among hospitalized patients

Majid Afshar, MD, MSCR – Assistant Professor of Medicine, Loyola University Medical Center

Meng Xie, BE – Research Assistant, Loyola University Chicago

Kristin Swope, BS – Medical Student, Loyola University Chicago

Niranjan Karnik, MD, PhD – Associate Dean for Community Behavioral Health, Rush University

Elizabeth M. Salisbury-Afshar, MD, MPH, FAAFP, DFASAM, FACPM – Director of the Center for Multi-System Solutions to the Opioid Epidemic, American Institutes for Research (AIR)

Background and Aims: A computable phenotype represents a clinical condition that may be determined solely from electronic health record (EHR) data. We aimed to examine the test characteristics of a computable phenotype for opioid misuse and test its face validity using data from a health system’s EHR. Design: An observational retrospective study between January 1, 2007, and September 30, 2017 Setting: A tertiary urban health system Participants: All adult hospital inpatient encounters (n=229,884) and a smaller stratified random sample of patient encounters (n=615) for manual review. Interventions: A rule-based computable phenotype was developed a priori using EMR data elements to produce an operational definition for opioid misuse. Measurements: Sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV) of the computable phenotype was measured in the random sample. To examine face validity, the computable phenotype was scaled and applied to all inpatient encounters and comparisons of patient- and census-level characteristics between groups with and without opioid misuse were made. A final multivariable model was derived from 53 candidate variables to identify risk factors associated with opioid misuse. Findings: The sensitivity, specificity, NPV, and PPV of the computable phenotype were 89.8%, 83.7%, 91.2%, and 79.2%, respectively. Laboratory and medication data identified an additional 1,417 encounters with opioid misuse beyond claims data and comprised 20% of the cohort with opioid misuse (n=7,086). The multivariable model identified the following 14 variables to be associated with opioid misuse: age, sex, race/ethnicity, insurance status, neurological disease, liver disease, no obesity, alcohol use disorder, psychoses, depression, chronic pain, census-derived food stamp usage, and prior 30- and 365-day healthcare utilization. Conclusion: We demonstrate good test characteristics and face validity of a computable phenotype for opioid misuse. The operational definition may be applied to better inform hospitals about their patients with opioid misuse and target interventions.


  View Poster
Poster Thumbnail
Peer Engagement Specialists in the Emergency Department: Impact on Readmission

Keshab Subedi, MS, MS – Statistician, Christiana Care Health System

Claudine Jurkovitz, MD, MPH – Senior Physician Scientist, Christiana Care Health System

Beverly Wilson, MS, MS – Program Administrator, Project Engage, Christiana Care Health System

Terry Horton, MD – Chief, Division of Addiction Medicine, Christiana Care Health System

Background: Project ‘Engage ED’ is a program using embedded Peer Engagement Specialists (PES) to connect Emergency Department (ED) patients with substance use disorder (SUD) to community-based treatment. To ensure the earliest possible transition from the ED, the PES facilitate patients admission to SUD treatment or provide education on treatment options if patients refuse SUD treatment. This study aims to assess the effect of project ‘Engage ED’ on length of stay (LOS) and one year ED readmission. Methods: The analysis included patients from a large hospital ED who were seen by the PES in 2013-2017 and were discharged directly from the ED. The control group consisted of patients who visited the ED in the same time period as the Engage patients, had a history of SUD and a primary complaint similar to that of the Engage patients. Two subgroups were created for the controls, one matched with the patients admitted to SUD treatment, the other matched with those who were only provided education using a one-to-one propensity match model, based on demographic and comorbidity covariates. We used two separate Generalized Linear Models, with log link function and gamma distribution to evaluate the PES intervention’s impact on index visit LOS. Similarly, two separate negative binomial models were employed to evaluate the PES impact on the number of ED visits within one year of the first Engage intervention. Results: The Intervention group included 1594 patients discharged from the ED. Of those, 66 % were male, 72 % white, average age was 38 years, 414 (26%) were admitted to SUD treatment, 1180 (74%) were informed of SUD treatment options. Only 368 patients in ‘admitted to SUD treatment’ group and 979 in the ‘education’ group were successfully propensity-matched with controls. The LOS of the patients in the Engage intervention was highly right skewed with a median of 7.4 and 5.9 hours for ‘admitted to SUD treatment’ and ‘education’ patients groups respectively. For patients arriving in the evening shift, the LOS increased by 39% (p= 0.015) for those connected to SUD treatment compared to controls and decreased by 11% (p=0.003) for those who only received information. There was no change in LOS for the patients arriving during the day shift. The number of ED visits within one year after the index visits ranged from 0 to 61 (median=1) and 0 to 98 (median=1) for those admitted to SUD treatment and those who received information respectively. Admission to SUD treatment reduced the number of ED visits within one year post Index visit by 59 % (p <0.001) compared to controls. For patients who only received SUD treatment information the number of ED visits remained unchanged (p=0.097). Conclusion: Compared to controls, patients who were successfully connected to SUD treatment experienced an increase in the index visit LOS but a significant reduction of ED visits whereas the number of ED visits did not decrease for patients who refused SUD treatment and were only provided information. Further research exploring factors governing the patients’ decision to start SUD treatment is warranted.


  View Poster
Poster Thumbnail
Characterizing age-based differences in fentanyl risk communication

Miriam Harris, MD, FRCPC, MS.c – Addiction Medicine Fellow, Boston Medical Center, Boston University School of Medicine

Sarah Bagley, MD – Assistant Professor of Medicine and Pediatrics, Boston Medical Center, Boston University School of Medicine

Alexander Y. Walley, MD, MSc, FASAM – Director, Addiction Medicine Fellowship, Boston Medical Center, Boston University School of Medicine

Spoorthi Sampath, BSc – Medical student, Boston University School of Medicine

Christine Gunn, MA, PhD – Assistant Professor, Boston University School of Medicine

Introduction: In many states, deaths due to fentanyl overdose are increasing amid reports that people do not understand the risks associated with fentanyl use.1,2 Overdose risk communication, a two-way process of information exchange to negotiate better risk-related outcomes, has not been examined. This study aimed to explore experiences with and preferences for fentanyl-related overdose risk communication, and identify how these vary by age. Methods: We purposively sampled two groups of people with past year fentanyl use in Boston for qualitative interviews: Those < 25 years and those >35 years. Equal numbers of men and women were recruited in each group. We conducted a grounded content analysis of professionally transcribed interviews. Codes were analyzed to build themes around risk communication preferences and behaviors, and compare strategies currently employed vs. those desired by each age group. Results: Twenty participants were enrolled, equally sampled across age categories. All but one reported experiencing overdose, and many attributed this to fentanyl. Participants wanted to talk about overdose risk with others; the most credible sources being people who had experienced addiction. While participants were open to discussing risks with health care providers, fourteen participants had experienced stigma in health care settings that undermined risk conversations. They sought communication that was non-judgmental, respectful, expressed compassion, and did “not sugar coat it”. Younger participants wanted interactive, practical training, such as how to recognize an overdose. Older participants wanted facilitated referrals or direct handoffs to treatment in post-overdose settings as a means to reduce risk, which were lacking in such settings. Conclusions: People using fentanyl wanted to talk about risks with others, emphasizing the need for compassion and recognizing the difficulties of making behavior change or entering treatment. Harm reduction and treatment messages desired by each age group differed, and offer opportunities for outreach efforts to tailor topics and services.


  View Poster
Poster Thumbnail
Medications for Opioid Use Disorder following Injection Drug Associated Endocarditis

Simeon Kimmel, MD, MA – Research Fellow in Infectious Diseases and Addiction Medicine, Boston Medical Center

Alexander Y. Walley, MD, MSc, FASAM – Director, Addiction Medicine Fellowship, Boston Medical Center, Boston University School of Medicine

Dana Berson, MPH – Assistant Director, Special Analytic Projects, MA Department of Public Health

Marc Larochelle, MD, MPH – Assistant Professor of Medicine, Boston Medical Center and Boston University School of Medicine

Introduction: People who inject drugs are at risk for endocarditis, a serious infection of the heart which requires prolonged intravenous antibiotics for treatment. Concurrent with the opioid crisis, injection drug associated endocarditis (IDU-IE) cases have increased. Hospitalizations for IDU-IE among patients with opioid use disorder are opportunities to initiate treatment with medications for opioid use disorder (MOUD). Though initiation of MOUD is recommended and feasible, the proportion receiving MOUD following discharge for IDU-IE and subsequent association of MOUD with rehospitalization and mortality is unknown. Methods: We performed a retrospective cohort study of Massachusetts residents ages 18 to 64 with a first hospital encounter for IDU-IE from July 2011 to September 2015. We employed a state-maintained population registry that includes person-level medical claims, prescription monitoring program, mortality, and state substance use treatment data. We defined IDU-IE as any inpatient discharge with a primary diagnosis of endocarditis and at least one claim in the prior 6 months with a diagnosis of opioid use disorder, injection drug use, or Hepatitis C. We examined the proportion of individuals who received MOUD at least once in the 3 months following discharge, defined as treatment in a methadone maintenance program or receipt of buprenorphine or naltrexone. The outcomes were rehospitalization and death in the 6 months following discharge. We used multivariable Cox Proportional Hazard Models to examine the association of MOUD receipt with rehospitalization and mortality, controlling for gender, age, medical and psychiatric comorbidity, homelessness, and alcohol use. We conducted a sensitivity analysis considering MOUD exposure as a monthly time-varying variable. Results: Among 653 individuals identified with an IDU-IE discharge, 128 (19.6%) individuals had received MOUD in the 3 months before hospitalization and 145 (22.5%) received MOUD in the 3 months following discharge. Of those receiving MOUD after discharge, 39 (27%) received methadone, 101 (70%) buprenorphine, <10 naltrexone and <10 received more than one MOUD. 128 (20%) individuals died within 6 months discharge, with 20 (16%) due to opioid overdose. MOUD receipt within 3 months following discharge was not associated with mortality (HR 0.81, 0.5-1.3) or rehospitalization (HR 1.0, 0.8-1.2). In the sensitivity analysis analyzing MOUD as a monthly time varying exposure, there was no association with reduced rehospitalization (HR 0.96, 0.7-1.3); however, there were no deaths in the month an individual received MOUD (HR 0). Conclusion: Among the minority of individuals who received MOUD following endocarditis, initiation of MOUD was not associated with reduced mortality or rehospitalization, but no deaths were observed in the month MOUD was received. Efforts to understand and reduce barriers to MOUD treatment engagement and retention following IDU-IE are necessary. Financial Support: Support from ASAM 2017 Fellowship Award, Fellow Immersion training Program in Addiction Medicine (R25DA013582), Research in Addiction Medicine Scholars Program (R25DA033211), Boston University Clinical HIV/AIDS Research Training (5T32AI052074), National Center for Advancing Translational Sciences, National Institutes of Health, through BU-CTSI (1UL1TR001430), and Office of National Drug Control Policy subcontract to University of Baltimore (G1799ONDCP06B).


  View Poster
Poster Thumbnail
Risk factors and costs of 30-day readmissions following alcohol-related hospitalizations

Alison Silverstein, MPH – Research Scientist, Precision Health Economics

Carole D. Gleeson, MSc, MBA – Director, Health Economics and Outcomes Research, Alkermes, Inc.

Jesse Sussell, PhD – Senior Research Economist, Precision Health Economics

Deborah S. Hasin, PhD – Professor of Epidemiology, Columbia University

Anju Parthan, PhD – Director Health Economic and Outcomes Research, Alkermes

Introduction Patients with alcohol-related diagnoses are at high risk of hospital readmission within 30 days of initial discharge.[1,2] Among patients with alcohol-related diagnoses, alcohol misuse and high-risk drinking patterns,[3] as well as psychiatric comorbidities,[4] have been independently associated with increased risk of readmission, compared to those with lower-risk drinking and those without psychiatric comorbidity, respectively. Understanding risk factors for readmission among patients hospitalized with alcohol-related diagnoses may help to identify patients who would benefit from treatment in an effort to lower the chance of 30-day readmission.[2] Using a nationally representative sample, this study estimated 30-day all-cause readmission rates among U.S. patients with an index hospitalization for an alcohol-related diagnosis between 2010 and 2015, and evaluated risk factors and costs associated with these readmissions. Methods The Nationwide Readmissions Database (NRD)[5] was used to identify patients who were 18 years old at initial hospitalization, had an alcohol-related condition as the primary diagnosis (based on ICD-9-CM codes), and were discharged between 2010 and 2015. Patients were followed for 30 days post initial hospitalization within the same calendar year to identify all-cause readmissions. These initial hospitalizations were segmented into those with and those without a 30-day readmission. A logistic regression analysis was conducted to assess the association between the risk factors and likelihood of 30-day readmission. Risk factors of interest included patient demographics, characteristics of the initial hospitalization, and hospital attributes. Average costs of the initial admission and the readmission were estimated. Results The NRD contains 113,931,723 adult hospital admissions between 2010 and 2015, and 1,124,228 index admissions with alcohol-related primary diagnoses were included. Among these hospitalizations, patients had a mean age of 49 years, were primarily male (73%), and 45% had public insurance coverage. The proportion of alcohol-related hospitalizations between 2010 and 2015 followed by a 30-day all-cause readmission was 13.3% (vs. 10.0% among all patients with a hospitalization during the same period). The annual rate of 30-day readmissions among patients with alcohol-related initial hospitalizations increased from 119 readmissions per 1,000 admissions in 2010 to 140 per 1,000 in 2015, while the rate of 30-day readmissions among all patients with a hospitalization for any reason declined over that period, from 103 per 1,000 to 98 per 1,000. Results from the regression analysis suggested that age, male gender, presence of comorbid conditions (mental health, cancers, diabetes), discharge against medical advice, admission to large and teaching hospitals, and Medicaid vs. non-Medicaid payment were all risk factors for 30-day all-cause readmission. Mean costs of initial alcohol-related hospitalizations were greater among those with a 30-day readmission than without ($10,156 vs. $8,453, p<0.001), and the mean cost of 30-day readmission was $11,196. Conclusion The increase in rates of readmission following alcohol-related initial hospitalizations between 2010 and 2015 contrasts with the relatively stable readmission rates observed following all-cause hospitalizations. Identifying patients with alcohol-related diagnoses at high risk of readmission and implementing effective treatments are needed to begin to address the clinical and economic burden of 30 day readmission.


  View Poster
Poster Thumbnail
Urine drug test results track co-occurrence of illicit fentanyl in heroin-positive patients

Maria G. Guevara, Pharm.D., CPE – Director, Clinical Affairs, Education & Training, Millennium Health

Eric Dawson, Pharm.D. – Vice President of Clinical Affairs, Millennium Health

Angela Huskey, Pharm.D, CPE – Chief Clinical Officer, Millennium Health

Leah LaRue, Pharm.D., PMP – Associate Director, Clinical Affairs, Millennium Health

Purpose: In 2016, CDC and DEA issued warnings on increased prevalence of fentanyl in the illicit market. Up to 50 times more potent than heroin and up to 100 times more potent than morphine, fentanyl is often spiked into heroin or sold as heroin, frequently without knowledge of the consumer. Epidemiology of fentanyl-spiked heroin has relied largely on reports of seizures by law enforcement agencies and accounts of overdoses by emergency healthcare professionals, first responders, medical examiners and coroners. In an effort to provide additional insights into increasing use of fentanyl-spiked heroin, this study examines urine drug test results acquired by a national drug testing laboratory to track co-occurrence of non-prescribed fentanyl in national, regional and statewide populations of heroin-positive healthcare patients. Objective/Methods: A retrospective review was conducted for urine drug test results from January 1, 2013 through May 31, 2018 for patient samples submitted for testing by healthcare providers as part of therapeutic treatment. The analysis was completed on tests ordered for definitive drug testing by LC-MS/MS for 6-MAM (heroin metabolite) and fentanyl (fentanyl/norfentanyl). Patients reported to have been prescribed fentanyl were removed from the analysis. Heroin positives were examined, and within those heroin positives, fentanyl positives without a reported prescription were measured. Results Heroin positivity rates in a population of patients in healthcare remains approximately the same (0.94% in 2013, 1.15% in 2018). Positivity rates for non-prescribed fentanyl in the heroin-positive population have risen remarkably since 2013. Nationally, in 2013 we found that 2.04% of samples with heroin positives were concomitantly positive for non-prescribed fentanyl. In 2018, 38.72% of heroin positive samples were concomitantly positive for non-prescribed fentanyl, a 1800% rise. The national average of heroin positive samples with non-prescribed fentanyl has stayed fairly steady in the last year and a half (40.93% in 2017, 38.72% in 2018). However, this varies substantially by region of the United States. The percent of heroin with non-prescribed fentanyl ranged in 2018 from 4.82% in the Pacific region up to 83.93% in the East South Central region. There was an increase in the correlation between heroin and non-prescribed fentanyl from 2017 to five months into 2018 in the following regions: East North Central (58.09 % to 72.96%), East South Central (77.45% to 83.93%), Mid-Atlantic (55.11% to 61.24%), New England (66.28% to 76.09%), South Atlantic (62.16% to 67.87%), West North Central (46.61% to 68.04%), and West South Central (12.30% to 22.06%). Conclusion Though the source of non-prescribed fentanyl cannot be verified as definitively fentanyl-spiked heroin, the data suggests the average has remained fairly steady in the last year. Regionally, there continues to be increases in prevalence of fentanyl-spiked heroin. These results may help inform and focus clinicians, first responders, and public health officials to better leverage various modalities at their disposal (e.g.,MAT, naloxone, and fentanyl specific messaging) in their ongoing efforts to halt the devastating toll of fentanyl-spiked heroin. Secondarily, this study demonstrates the potential utility of UDTs from healthcare patients as an epidemiological tool to track emerging patterns of drug-use.


  View Poster
Poster Thumbnail
Evaluation of an Alcohol Withdrawal Protocol Using the Brief Alcohol Withdrawal Scale

Brian Lindner, PharmD – Pharmacy Resident, Johns Hopkins University

Darius Rastegar, MD – Associate Professor of Medicine, Johns Hopkins Bayview Medical Center

Anika Alvanzo, MD, MS, DFASAM – Assistant Professor/ Associate Medical Director, Johns Hopkins University School of Medicine

Andrew Jarrell, PharmD – Clinical Specialist Pharmacist, Johns Hopkins University

Background: Symptom-triggered benzodiazepine protocols are the standard of care for treatment of alcohol withdrawal syndrome (AWS). While the most common assessment tool is the Clinical Institute of Withdrawal Assessment for Alcohol, revised (CIWA-Ar), shorter scales may be helpful for clinician efficiency. In 2016, the Brief Alcohol Withdrawal Scale (BAWS) was developed; this assessment tool consists of five items (sweats, tremor, agitation, orientation and hallucinations), each scored from 0 to 3. A corresponding treatment protocol provides dosing of benzodiazepines and monitoring parameters based on the most recent BAWS score. The initial study evaluating the BAWS protocol showed efficacy in treating alcohol withdrawal in a dedicated unit focused on treatment of substance withdrawal, but there was no evaluation of its use on other units. Methods: This was a retrospective review of all patients who had the BAWS protocol ordered between August 1, 2016 and July 31, 2017 at Johns Hopkins Hospital. We included all admissions on which this protocol was ordered, with the exception of those who had the protocol ordered and discontinued within one minute and subsequent encounters by the same patient. Charts were reviewed to collect information on demographics, admitting service, development of severe withdrawal (defined as BAWS>8 or transfer to a higher level of care), duration of treatment, provision of benzodiazepines and adjunctive medications. Results: There were a total of 799 admissions that met our criteria. The mean age of the patients was 53 and 74% were male; 79% were admitted to the medical service, 21% to a surgical service. Patients received no benzodiazepines on 415 (52%) of the admissions. The mean time on the protocol was 45 hours. Severe withdrawal was observed in 17 admissions (2%); 4 required transfer to a higher level of care. Adjunctive medications were given in 86 (11%) of the admissions, the most common was antipsychotic medications (68), followed by antiepileptics (10), dexmedetomidine (12), propofol (18), lorazepam infusion (8) and ethyl alcohol (5). Conclusions: Most of the patients on the BAWS protocol received little to no benzodiazepines; severe withdrawal was uncommon and few received additional medication. These findings support the BAWS as a reasonable symptom-triggered alcohol withdrawal scale and our protocol as one that can be used across a variety of hospitalized patient populations.


  View Poster
Poster Thumbnail
Buprenorphine MAT: Low Threshold, Patient-centered, and Low Cost Care

Neil Flynn, MD, M.P.H. – Professor Emeritus, University of California Davis

Andrew Gilbert, B.S. – Medical research assistant, Transitions Buprenorphine Treatment Clinic of Sacramento

Phillip Summers, M.D., M.P.H. – Resident physician, University of California Davis

Christina Bourne, M.D., M.P.H. – Resident physician, University of California Davis

Intro/background: Buprenorphine (B) medication assisted therapy (BMAT) is a grossly underutilized modality for the treatment of opioid use disorder (OUD). Current BMAT practice is rule-bound, not readily available, and may rely on inadequate doses of B. Studies have shown reductions in overdose mortality while on B of 3 – 20-fold. Clinic: The clinic serves 580 people living with OUD (PLWOUD). It follows harm reduction principles, accepts all patients, requires monthly visits, applies motivational interviewing, charges a fee-for-service of $200 monthly, has a no-charge homeless program, offers phone interviews for patients unable to attend clinic, does not urge weaning, does not discharge patients who use other drugs or relapse to their opioid use, does replace lost or stolen medications, conducts urine testing at least every 3 months, utilizes CURES/PDMP at most visits, offers case management. Staff: 3-6 physicians, 1 FNP, clinic manager, 3 clerical personnel. Physicians are paid $50 per patient seen, up to 3 patients per hour. We train FP, Int Med, Psych residents and community physicians. Residents moonlight with us. Treatment goals are to de-stigmatize, prevent relapse, support abstinence, and diagnose and treat underlying psychiatric and painful disorders. Dosing is adjusted on the basis of reported “craving”, to maximum dose of 32 mg. Methods: We examined a random sample of the 1,614 patients treated in our clinic over the past 10 years. Chart abstraction was performed by one of the authors (AG). The resulting data base was queried for clinically important variables. Results: We reviewed 341 currently-enrolled and inactive patients, spanning 10 years. Among the 100 currently-enrolled patients were 35 females and 65 males. Mean, median, range of ages at entry into treatment were, respectively: 34, 32, and 19-63 years; of duration of treatment was 27, 18, and 0.5 – 91 mos.; of current doses were 23, 24, and 2-32 mg. Drugs of choice are 35% heroin, 47% pills, 15% unknown. Psych co-morbidities at entry included major depression 18%, bipolar 8%, anxiety/panic disorder 13%, PTSD 2%, ADHD 2%. Pain: 7%. Plain B in prescribed in 51%. Insurance: 72%. Among 241 inactive patients were 96 females, 145 males, with a mean, median and range in age at entry of 36, 34, and 17-67 years, respectively; of duration of treatment was 7, 2, and 0.03 – 47 months. Doses at discharge from clinic were 19, 16, and 2-32 mg. Drugs of choice at entry were 51% heroin, 33% pills, and 16% unknown. Psych comorbidities included depression 21%, anxiety/panic 11%, ADHD 3%, bipolar 7%, PTSD 4%, borderline 2%, OCD 1%, schizophrenia 1%, more than one disorder 8%, none 51%, no record 16%, benzo use 18%. Plain B prescribed in 67%. Insurance: 53%. Conclusions: Relatively low cost long-term BMAT is possible on a large scale. This clinic prescribes relatively high doses of B. To date, 8 of our trainees have confidently established their own BMAT clinics in our region. Benzodiazepine use is a major problem. Insurance coverage has increased.


  View Poster
Close this panel
Browse By Poster Author
Browse By Title
Close this panel
Stuff for Poster Tools
Stuff for Share


Technical Support

(877) 426-6323


SUBMIT FEEDBACKfeedback icon

We really appreciate your feedback on the eventScribe website. We use the data to improve the experience and simplify the process for users like you.


Log In / Sign Up

Already have an Event Scheduler or mobile app login? Login with those details. If not, create a login.

Log In   Sign Up
Access your bookmarked poster and notes by logging in ...   Sign up to take notes on poster, bookmark poster, and submit feedback.
  Lost your access key?      
You need to be logged in to bookmark posters, save notes, or rate posters.