BI-B indicates brief intervention with telephone boosters; MSO, minimal screening only; SAR, screening, assessment, and referral; and TLFB, timeline follow-back.
aOne of the 6 sites was not able to include prisoners in research, so participants at that site were withdrawn from the study if incarcerated.
bThe number of participants included in the primary analysis is greater than the number who completed the 3-month follow-up because 3-month TLFB data were collected retrospectively for patients who completed the 6-month but not the 3-month visit.
Bogenschutz MP, Donovan DM, Mandler RN, Perl HI, Forcehimes AA, Crandall C, Lindblad R, Oden NL, Sharma G, Metsch L, Lyons MS, McCormack R, Macias-Konstantopoulos W, Douaihy A. Brief Intervention for Patients With Problematic Drug Use Presenting in Emergency DepartmentsA Randomized Clinical Trial. JAMA Intern Med. 2014;174(11):1736-1745. doi:10.1001/jamainternmed.2014.4052
Medical treatment settings such as emergency departments (EDs) present important opportunities to address problematic substance use. Currently, EDs do not typically intervene beyond acute medical stabilization.
To contrast the effects of a brief intervention with telephone boosters (BI-B) with those of screening, assessment, and referral to treatment (SAR) and minimal screening only (MSO) among drug-using ED patients.
Design, Setting, and Participants
Between October 2010 and February 2012, 1285 adult ED patients from 6 US academic hospitals, who scored 3 or greater on the 10-item Drug Abuse Screening Test (indicating moderate to severe problems related to drug use) and who were currently using drugs, were randomized to MSO (n = 431), SAR (n = 427), or BI-B (n = 427). Follow-up assessments were conducted at 3, 6, and 12 months by blinded interviewers.
Following screening, MSO participants received only an informational pamphlet. The SAR participants received assessment plus referral to addiction treatment if indicated, and the BI-B participants received assessment and referral as in SAR, plus a manual-guided counseling session based on motivational interviewing principles and up to 2 “booster” sessions by telephone during the month following the ED visit.
Main Outcomes and Measures
Outcomes evaluated at follow-up visits included self-reported days using the patient-defined primary problem drug, days using any drug, days of heavy drinking, and drug use based on analysis of hair samples. The primary outcome was self-reported days of use of the patient-defined primary problem drug during the 30-day period preceding the 3-month follow-up.
Follow-up rates were 89%, 86%, and 81% at 3, 6, and 12 months, respectively. For the primary outcome, estimated differences in number of days of use (95% CI) were as follows: MSO vs BI-B, 0.72 (−0.80 to 2.24), P (adjusted) = .57; SAR vs BI-B, 0.70 (−0.83 to 2.23), P (adjusted) = .57; SAR vs MSO, −0.02 (−1.53 to 1.50), P (adjusted) = .98. There were no significant differences between groups in self-reported days using the primary drug, days using any drug, or heavy drinking days at 3, 6, or 12 months. At the 3-month follow-up, participants in the SAR group had a higher rate of hair samples positive for their primary drug of abuse (265 of 280 [95%]) than did participants in the MSO group (253 of 287 [88%]) or the BI-B group (244 of 275 [89%]). Hair analysis differences between groups at other time points were not significant.
Conclusions and Relevance
In this sample of drug users seeking emergency medical treatment, a relatively robust brief intervention did not improve substance use outcomes. More work is needed to determine how drug use disorders may be addressed effectively in the ED.
Recent years have seen a marked increase in efforts to develop, implement, and evaluate models for integration of substance use disorder interventions into health care settings. Specialty treatment for addictions has important limitations. Of the 23.1 million Americans needing treatment for substance use disorder, only 2.5 million (10.8%) receive specialty treatment annually,1 whereas 82.6% of adults see a health care professional annually.2 Therefore, many individuals with harmful or hazardous substance use are not receiving treatment but are potential candidates for brief interventions in medical settings (with or without further treatment). The Affordable Care Act strongly emphasizes and incentivizes the integration of behavioral health and medical treatment.3 SBIRT models, comprising Screening, Brief Intervention, and Referral to Treatment, have been promoted as an important strategy for addressing substance use problems in medical settings.4,5
Results of SBIRT interventions for alcohol problems, although mixed, provide evidence of efficacy across settings. Meta-analyses of the fairly extensive literature on SBIRT for alcohol in primary care demonstrate significant although fairly modest effects on subsequent drinking over 12 months of follow-up.6,7 Importantly, these studies primarily included nondependent drinkers. The more limited literature on SBIRT in trauma centers suggests that such interventions can result in decreases in drinking and subsequent arrests for driving under the influence.8,9 A meta-analysis of emergency department (ED) SBIRT interventions for alcohol use disorders did not demonstrate beneficial effects on drinking but found significant decreases in alcohol-related injury.10 Subsequently, a well-designed study demonstrated significant decreases in both drinking and driving while intoxicated in harmful and hazardous drinkers who received a brief intervention in the ED.11
Data on SBIRT for drug use problems are much more limited. One single-site study demonstrated decreases in heroin and cocaine use among dependent primary care patients following a brief intervention.12 An international World Health Organization (WHO) study also found decreases in drug use in patients receiving a brief intervention using feedback based on the results of the WHO ASSIST (Alcohol, Smoking and Substance Involvement Screening Test) (although not among participants in the United States).13 In EDs, observational studies have demonstrated decreases in drug use following SBIRT interventions.5,14 However, very few controlled trials have been published of SBIRT approaches in drug-using ED patient populations.15- 18
The Screening, Motivational Assessment, Referral, and Treatment in Emergency Departments (SMART-ED) study was designed to address this gap by contrasting substance use and substance-related outcomes among patients endorsing problematic drug use during an ED visit who are randomly assigned to 1 of 3 treatment conditions: (1) minimal screening only (MSO); (2) screening, assessment, and referral to treatment (if indicated) (SAR); and (3) screening, assessment, and referral plus a brief intervention (BI) with 2 telephone follow-up booster sessions (BI-B).
All study procedures were overseen by an independent Data and Safety Monitoring Board and reviewed and approved by the institutional review boards (IRBs) of each site. The study was conducted under a Certificate of Confidentiality from NIDA. Detailed methods and rationale have been described previously.19 The study was a multisite, randomized, prospective trial of 3 groups. Individuals presenting in the ED who endorsed problematic drug use on screening were randomized in 1:1:1 ratio to MSO vs SAR vs BI-B. The SAR group was included to evaluate the effects of assessment and referral procedures independent of those of the brief intervention (ie, attention control).20 Follow-up assessments of all 3 groups were conducted by blinded interviewers at 3 months, 6 months, and 12 months after enrollment.
The study was conducted in 6 EDs of urban academic hospitals, each of which partnered with a node of the National Institute on Drug Abuse (NIDA) National Drug Abuse Treatment Clinical Trials Network (CTN). Three sites were on the East Coast, 1 in the Midwest, 1 in the South, and 1 in the Southwest.
Participants were men and women 18 years or older who were seeking medical treatment at the ED, had adequate English language proficiency, were capable of providing informed consent, had a score of 3 or greater on the 10-item Drug Abuse Screening Test (DAST-10)21 indicating moderate to severe problems related to drug use, reported at least 1 day of drug use in the 30 days prior to screening, were willing to participate in the protocol, and had access to a telephone. Individuals were excluded if they were prisoners or in police custody, were currently engaged in or actively seeking addiction treatment, resided more than 50 miles from the follow-up location, were unable to provide sufficient contact information, or had already participated in the study. Participants were compensated $50 for the screening/baseline visit and $75 for each of the 3 follow-up visits.
During defined recruitment hours, research staff screened ED patients who were possibly eligible for the study. Prior to screening, age, sex, and reason for ED visit were collected from the electronic medical record. Research staff then obtained verbal consent for the anonymous collection of screening data, using a brief IRB-approved script. The screening instrument consisted of 4 sections: the Heavy Smoking Index22; the 3 alcohol consumption questions from the Alcohol Use Disorders Identification Test (AUDIT-C)23; the DAST-1021; and questions to determine primary substance of abuse, days of use of the primary substance, and substance-relatedness of the ED visit. A secondary screening form addressed additional exclusion criteria. Participants eligible to this point were invited to complete the full informed consent. Prior to randomization, consenting participants completed a demographic questionnaire, provided locator information, and provided a hair sample for use as an objective measure of substance use. The Psychemedics Corporation performed the hair analysis for the study, using hair samples covering a period of approximately 90 days. Samples testing positive during the preliminary screening radioimmunoassay were confirmed using chromatographic and mass spectrometric methods.24
The randomization procedure was conducted through a centralized, web-based process set up by the CTN Data and Statistics Center (DSC). Participants were stratified by site, drug problem severity, and alcohol use severity. The randomization schedules consisted of balanced varied size blocks within strata. Allocation was revealed in 2 stages. Initially, the staff member performing the randomization was informed whether the participant was in the MSO group or not. Those not in the MSO group received the baseline assessment of substance use and consequences, consisting of a 30-day timeline follow-back (TLFB) interview25 and the NIDA-Modified version (NM-ASSIST) of the WHO ASSIST.13 After completion of the baseline assessment, staff were informed whether participants were in the SAR or BI-B group.
The MSO participants did not receive further assessment or treatment following randomization but were given an informational pamphlet about drug use and misuse, its potential consequences, and treatment options.
The SAR participants were provided with the same information pamphlet as the MSO group. In addition, following assessment, SAR participants with a NM-ASSIST score of 27 or greater for 1 or more substances (indicating high risk of dependence) and any who requested referral were also provided a referral to treatment, consisting of a recommendation to seek treatment and a standardized list of available treatment options.
Individuals randomized to the BI-B condition received the same information and referral (if indicated or requested) as those in SAR. In addition, while in the ED the BI-B group received a manual-guided brief intervention based on motivational interviewing principles,26 with content patterned on that of motivational enhancement therapy,27 including use of feedback based on screening information and development of a change plan if indicated. Consistent with the spirit of motivational interviewing, the BI focused initially on the primary problem drug identified by the participant, but also addressed concerns about other substance use if these came up in the session. In addition, participants in the BI-B group received up to 2 telephone “booster” sessions to check whether they had entered treatment, review change plans, and reinforce motivation. The booster calls occurred within 7 days of the ED visit if possible, but up to 1 month was allowed to complete the calls if necessary. Booster calls were made using a centralized, studywide intervention booster call center.
Interventions were performed by staff hired for the study. Interventionist were not required to have prior clinical training. They received a 2-day training in basic motivational interviewing skills, followed by a second 2-day training devoted to teaching the details of the specific brief intervention used in the trial. On completion of the basic training, interventionists were required to complete practice sessions including at least 2 with consenting pilot/training patients and receive satisfactory fidelity ratings in order to be certified by the central monitoring center. They received ongoing supervision and fidelity monitoring during the course of the study.
Follow-up assessments were conducted by interviewers blinded to treatment assignment at a site separate from the ED. A total of 2915 interviews (91.8%) were conducted face-to-face, and 261 (8.2%) were conducted by telephone. The primary outcome was days of use of the patient-defined primary problem drug, assessed by the TLFB for the 30-day period preceding the 3-month follow-up. Secondary outcomes included days of use of the primary substance at 6 and 12 months; the number of days abstinent from all drugs at 3, 6, and 12 months; days of heavy drinking at 3, 6, and 12 months; and objective evidence of drug use based on analysis of hair samples.
The primary analysis contrasted MSO, SAR, and BI-B groups with respect to the primary outcome variable (days of use of the primary drug of abuse in the 30 days preceding 3-month follow-up) using a linear mixed model with a random site effect and fixed treatment effect and intercept, as well as fixed effects for baseline DAST-10 score, baseline AUDIT-C score, and baseline days of use of the primary substance reported during screening. Following the a priori analysis plan, a preliminary analysis included a site-by-treatment interaction, which was not statistically significant and was therefore excluded from the final model. Three pairwise contrasts were made with an overall type I error rate of α = .05. We considered a difference of 3 days to be a clinically significant difference in past 30-day use. A total of 1285 participants yielded 90% power to detect this difference, allowing for 15% attrition and assuming distributions equivalent to those observed in another ED study.14
Secondary self-reported substance use outcomes were analyzed using analogous methods. Hair sample results were analyzed using a generalized linear mixed model approach (logistic regression) with treatment arm, the 2 stratification variables (DAST-10 and AUDIT-C scores), and the corresponding baseline hair analysis result as fixed effects and site as a random effect.
The Figure summarizes the patient flow through the study. Staff identified 20 762 patients as potentially eligible, of whom 15 224 underwent screening. Of these, 13 939 were excluded, and 1285 were randomized to MSO (n = 431), SAR (n = 427), or BI-B (n = 427). Emergency departments enrolled 135 to 287 participants, with a mean of 214 per ED.
Participant characteristics are summarized in Table 1. The mean (SD) age was 36 (12) years; 70% of participants were men; and 50% were white, 34% black/African American, and 24% Hispanic. The most common primary drugs of abuse were cannabis (44%), cocaine (27%), street opioids (17%), prescription opioids (5%), and methamphetamine (4%). Socioeconomic status of the sample as a whole was low, with 63% having an annual household income less than $15 000, 42% being unemployed, and 32% not graduating from high school. Mean (SD) DAST-10 score was 5.8 (2.3), with 652 participants (51%) scoring 6 or higher, indicating substantial or severe problems related to drug use. The mean (SD) AUDIT-C score was 5.4 (3.8), and participants reported using their primary substance of abuse at a mean (SD) 16.2 (11.6) days during the past 30 days.
Within the BI-B group, 421 (99%) received the initial brief intervention in the ED, 243 (57%) received the first booster session, and 166 (39%) received the second booster session. Thirty-one interventionists were trained and certified across the 2 waves of the study, plus 3 booster counselors. Treatment fidelity was assessed by scoring of treatment session audiotapes using the Motivational Interviewing Treatment Integrity coding system (MITI 3.1).28 Interrater reliability assessed on a random sample of 124 tapes was excellent, with intraclass correlations averaging 0.81 for global scores and 0.93 for behavior counts. A total of 380 initial sessions (90%) and 83 booster sessions (20%) were MITI coded. Mean global scores for the initial sessions ranged from 4.25 to 4.67, and for the booster sessions, from 4.64 to 4.86. These scores are well above the proficiency benchmark of 4.0.
A total of 233 participants (54.6%) in the BI-B group and 255 participants (59.7%) in the SAR group had ASSIST scores of 27 or higher (4 participants in the BI-B left before completing the ASSIST). A total of 250 participants (58.5%) in the BI-B group and 265 participants (62.1%) in the SAR group were referred for treatment. At the 3-month follow-up, 292 of 1136 participants (25.7%) across the 3 groups had at least 1 formal treatment contact (including inpatient treatment, any form of medication or counseling, and urine drug monitoring), with a median of 8.5 contacts, and 132 participants reported attending Narcotics Anonymous and Cocaine Anonymous with a median of 11 contacts. There were no significant between-group differences in any form of treatment attendance at any follow-up point (independent samples Kruskal-Wallace tests).
Primary outcome data were available for 1139 participants (89%). Interviews were conducted with 1026 participants at 3 months (80%), 1107 at 6 months (86%), and 1043 participants at 12 months (81%). The TLFB data for the 3-month time point were collected retrospectively from 113 participants at the 6-month visit (Figure).
For the primary outcome variable (days of use of the patient-defined primary problem drug during the 30-day period preceding the 3-month follow-up) there were no statistically significant treatment effects (Table 2). The effects of baseline DAST-10 score, AUDIT-C score, and use days were all significant at the .05 level. The site effect was not significant. It was noted that the data violated one of the assumptions of the model, the normal distribution of errors. Because the primary outcome fit the β-binomial distribution well, we reanalyzed the primary outcome using a β-binomial regression. The results of the β-binomial model are similar to those of the primary outcome model, indicating that the violation of the normality assumption does not have a serious impact on the outcome of the trial. Because the analyses assumed that data were missing at random, sensitivity analyses were conducted for missing data under various scenarios, including imputing negative for drug use, imputing positive for drug use, best and worst cases (in which imputation is positive in one arm and negative in the other), and a full range of intermediate cases. When the proportion of missing assigned to drug use was the same in the 2 arms, there was never a significant treatment effect between any pair of arms. Best and worst cases were usually but not always significant. Of the 729 imputation scenarios, only 32.9%, 24.1%, and 20.6% were significant for comparing B-IB vs SAR, BI-B vs MSO, and SAR vs MSO, respectively.
Parallel mixed model analyses were conducted for secondary outcomes from the TLFB (Table 3). At 3, 6, or 12 months, there were no significant effects of treatment on days of primary substance use (during the 30 days prior to assessment), days of any drug use, or heavy drinking days. Both BI-B and SAR groups showed decreased use of the primary substance from baseline to the 3-, 6-, and 12-month follow-up.
For the primary problem drug identified by participants, hair analysis data were available for 1044 participants (81%) at baseline, 842 (66%) at 3 months, 858 (67%) at 6 months, and 802 (62%) at 12 months (Table 4). At the 3-month follow-up, participants in the SAR group had a significantly higher rate of positive hair samples (265 of 280 [95%]) than did participants in the MSO group (253 of 287 [88%]) or the BI-B group (244 of 275 [89%]) (P = .02). Differences between groups at other time points were not significant. There were no differences between groups at any time point with respect to hair samples positive for any drug.
Given the heterogeneity of the study sample, it was important to explore whether there was evidence for treatment-by-attribute interactions or treatment effects in clinically meaningful subgroups within the sample. Separate parallel analyses adding fixed effects for sex, race, and ethnicity, as well as the treatment-by-attribute interaction, revealed no significant effects of sex, race, or ethnicity on the primary outcome, indicating that these attributes did not moderate the effects of treatment. The primary outcome analysis was repeated separately for the subgroups identifying cannabis (n = 567), cocaine (n = 349), or opioids (n = 287) as the primary substance of abuse. No significant treatment effect was found for any of these subgroups.
Despite robust implementation of a relatively extensive brief intervention, the BI-B strategy used in this study for ED patients screening positive for moderate to severe problematic drug use did not improve outcomes over those found with MSO or SAR, and SAR was not superior to MSO. These findings appeared to be consistent across sites and racial, ethnic, sex, and substance use categories. Overall, drug use decreased over time in all treatment groups, suggesting that the ED visit may mark a turning point for many drug-using patients, regardless of what specific treatment they receive. The study design does not allow any inference to be drawn as to the causal role of the ED visit itself.
The interventions used in the trial represent a fairly broad range of interventions ranging from minimal (a 20-item screen) to screening, assessment and referral procedures which could be considered a very brief intervention, to a 3-session brief intervention using motivational interviewing, comparable to a somewhat abbreviated version of motivational enhancement therapy. We cannot rule out the possibility that brief screening alone was efficacious and that the more intensive interventions did not add to its efficacy. The study design does not provide the opportunity to evaluate the efficacy of screening vs no intervention. It is also possible that other types of brief intervention would have had a greater effect than the interventions used in this study. However, the BI-B used in this study was similar to interventions that have had significant effects in other populations.
The results based on hair sample data differed from those based on the TLFB in that the analyses based on hair found greater rates of samples positive for the primary problem drug at 3 months in the SAR group than in the other 2 groups. This result should be treated with caution because it is contrary to the primary analyses based on the TLFB and for several other reasons. The analyses based on hair had considerably more missing data than those based on the TLFB, and these P values were not adjusted for multiple testing. The significant difference observed was an isolated finding and represents a fairly small effect size, so it could easily have resulted from chance. Finally, it is difficult to find a satisfactory explanation for the finding in which the addition of assessment and referral led to an outcome worse than that found in those receiving only minimal screening. Further analyses of the concordance between self-report and hair results may shed light on these findings and provide more information as to the usefulness of hair analysis in drug use disorder trials with infrequent follow-up contacts.
The findings of this study are relevant to the population represented by the sample, the types of intervention used in the trial, and the outcomes that were examined. The relatively high problem severity in the sample, as well as its heterogeneity, may have contributed to the lack of efficacy in this study. Most of the evidence for the efficacy of brief interventions for alcohol use disorders comes from studies with samples on the milder end of the severity spectrum.6,7,11 To take an example from the alcohol literature, in a study focused on alcohol use in trauma center patients, Gentilello et al8 found that a brief intervention was effective only in the subgroup with mild to moderate severity of alcohol use disorder, not in the group with high severity levels. Our results may not generalize to populations with less severe substance use disorders or those presenting in other settings. Because this intervention was focused primarily on drugs other than alcohol, the findings are not directly relevant to the efficacy of interventions focused on alcohol. This study also does not provide information on other important outcomes such as injuries, accidents, overdose, arrests, or violent behavior.
The findings of this study suggest that even a relatively robust brief intervention such as the one implemented in this trial is unlikely to be useful as a general strategy for the population recruited for this trial: ED patients with relatively severe drug problems and other life challenges. Further research will be needed to explore more intensive interventions targeting the most severely affected patients with substance use disorder visiting the ED and to ascertain whether screening and brief interventions play a useful role in the treatment of ED patients less severely affected by drug use disorders.
Accepted for Publication: June 28, 2014.
Corresponding Author: Michael P. Bogenschutz, MD, Department of Psychiatry, Center for Psychiatric Research, University of New Mexico Health Sciences Center, MSC11 6035, 1 University of New Mexico, Albuquerque, NM 87131-0001 (firstname.lastname@example.org).
Published Online: September 1, 2014. doi:10.1001/jamainternmed.2014.4052.
Author Contributions: Dr Bogenschutz had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Bogenschutz, Donovan, Mandler, Perl, Forcehimes, Crandall, Lindblad, Lyons, Douaihy.
Acquisition, analysis, or interpretation of data: Bogenschutz, Perl, Forcehimes, Crandall, Oden, Sharma, Metsch, Lyons, McCormack, Macias-Konstantopoulos, Douaihy.
Drafting of the manuscript: Bogenschutz, Donovan, Perl, Forcehimes, Crandall, Sharma.
Critical revision of the manuscript for important intellectual content: Bogenschutz, Donovan, Mandler, Perl, Forcehimes, Crandall, Lindblad, Oden, Metsch, Lyons, McCormack, Macias-Konstantopoulos, Douaihy.
Statistical analysis: Bogenschutz, Oden, Sharma.
Obtained funding: Bogenschutz, Donovan, Crandall, Douaihy.
Administrative, technical, or material support: Bogenschutz, Mandler, Perl, Forcehimes, Crandall, Lindblad, Lyons, Douaihy,
Study supervision: Bogenschutz, Forcehimes, Crandall, Metsch, Lyons.
Conflict of Interest Disclosures: Dr Bogenschutz reports grants from the National Institute on Drug Abuse (NIDA) during the conduct of the study and grants from the Lundbeck Foundation and the Heffter Research Institute, outside the submitted work. Dr Lindblad reports grants from the National Institutes of Health (NIH) during the conduct of the study. Dr Lyons reports grants from NIDA Clinical Trials Network (CTN), Ohio Valley Node, during the conduct of the study. Dr Macias-Konstantopoulos reports grants from NIDA-NIH through McLean Hospital (Belmont, MA) during the conduct of the study. No other disclosures are reported.
Funding/Support: The study was supported by the following grants from NIDA: HHSN271200900034C (EMMES Corporation); U10DA015833 (Principal Investigator [PI], Michael P. Bogenschutz); U10DA013714 (PI, Dennis M. Donovan); U10DA013720 (PIs, José Szapocznik, PhD, and Lisa Metsch); U10DA013732 (PI, Theresa Winhusen); U10DA013046 (PI, John Rotrosen); U10DA015831 (PIs, Roger D. Weiss and Kathleen M. Carroll, PhD); U10DA020036 (PI, Dennis C. Daley); and U10DA013035 (PIs, John Rotrosen and Edward V. Nunes).
Role of the Funder/Sponsor: Staff of NIDA Center for the Clinical Trials Network played an advisory role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. NIDA appointed members and coordinated meetings of the data safety monitoring board. The manuscript was reviewed and approved by the Publications Committee of the National Drug Abuse Treatment CTN.
Disclaimer: The authors are solely responsible for the content of this article, which does not necessarily represent the official views of NIDA or NIH. Dr Mandler and Dr Perl, employees of NIDA, are authors and did review and approve the manuscript as a part of their authorship roles.
Additional Contributions: We thank the following persons for their contributions to the study. From EMMES Corporation, Bethesda, Maryland: for trial coordination, Ro Shauna Rothwell, Eve Jelstrom, Radhika Kondapaka, and Maria Campanella; and for data management, Lauen Yesko, Colleen Allen, MPH, CRRA, and Paul Van Veldhuisen, PhD. From the University of New Mexico, Albuquerque: for fidelity monitoring, Karin M. Wilson and Christina Ripp; for quality assurance, Roberta Chavez, Rena Treacher, and Amber Martinez; and for trial management and data collection, Lindsay Worth, MS, Christine Lizarraga, Meredith M. Davis, Carolyn Camplain, Jill Gatwood, MS, and Craig Pacheco. From the University of Cincinnati, Cincinnati, Ohio: for site coordination and data collection, D. Beth Wayne, BSN, JD, Emily Dorer, Andy Ruffner, and Ron Coleman. From Jackson Memorial Hospital, Miami, Florida: for site management, John Cienki, MD. From the University of Miami Miller School of Medicine, Miami, Florida: for site management, coordination, and data collection, Lisa Abreu, MPH, Jessica Ucha, MSEd, Xavier Pereira, Oliene Toussaint, MSW, Daniel Glaser, Cheryl Walker, Richard Walker, Silvia Mestre, MSEd, Pedro Castellon, MPH. From New York University, New York: for site management, coordination, and data collection: Agatha Kulaga, Alexandra Schepens, Phoebe Gauthier, MA, Erica Silen, Sean Sobin, Lauren Moy, Bridget McClure, Shirley Irons, Sarah Farkas, Alexandra Kvernland, and for administrative support, John Rotrosen, MD. From West Virginia University, Morgantown: for site management and coordination, Owen Lander, MD, Marilyn Byrne, ACSW, Kelly Gurka, PhD, Stephen M. Davis, MPA, MSW; and for study interventions and data collection, Robert A. Wilson Jr, LPC, AADC, Gary D. Thompkins Jr MSW, LCSW, Kimberly E. Hotlosz, MS, CRC, LPC, Kathleen Chiasson-Downs, LPC, ALPS, Jodie Russell CRC, LPC, Shelley Layman, MPH, Casey Clark, BS, MPH, and Blair Lord, MSW. From McLean Hospital/Harvard University, Belmont, Massachusetts: for administrative support and site management, Hilary Connery Smith, PhD, Roger D. Weiss, MD, and Jessica Dreifuss, PhD.
Correction: This article was corrected on January 8, 2015, to correct an author’s name in the byline and Article Information.