Rapid Initiation of Drug Treatment Engagement

1. Study Objectives Aim 1: To evaluate the impact of the mobile, transitional MAT intervention on its ability to engage participants in targeted, existing MAT treatment slots at 1- and 6-month post-enrollment.

Aim 2: To evaluate the impact of the mobile intervention on subsequent drug use and overdoses at 6-month post-enrollment.

Aim 3: To assess the acceptability and costs of the intervention. The program and patient costs of delivering and participating in the intervention will be documented.

Study participants who will receive buprenorphine/naloxone will be provided with information, counseling, and support for selection of ongoing treatment options including methadone maintenance (MMT), extended-release naltrexone (XR-NTX), or continued treatment with buprenorphine/naloxone.

2. Study design Duration of Study Participation The duration of study participation including screening/baseline assessments to follow-up will be 6 months.

Total Number of Subjects A total of 250 participants will be engaged in the research at the time they seek treatment: 125 individuals will be enrolled as they engage with the mobile, rapid induction team and 125 individuals will be enrolled as they seek treatment from the BAC/CRC (treatment as usual).

Subject Recruitment The recruitment of participants for this study will be accomplished using a community awareness campaign involving street outreach, targeted social media, and local advertisement. This campaign will begin one month before activation of the rapid induction intervention. During this time, the mobile assessment team will begin to be present in the designated location where the Mobile Clinical Trials Unit (MCTU) will be parked.

Participant Education and the Informed Consent Process Potential participants who express interest in initiating treatment will be engaged in a structured approach to education about the study and its procedures. The information delivered at this point will be supportive and motivational and describe study participation as an opportunity to consider options for treatment. It will also be explained that their participation can contribute to a better understanding of the most effective ways to assist individuals who want to begin treatment. Staff will make it clear that even if they do not become engaged in treatment, their experiences are important and their contributions to the research will be valuable. This discussion will offer potential participants a chance to ask questions and receive prepared materials explaining the study and the requirements for participation.

Those who appear to be eligible and express interest will begin the informed consent process to allow for screening. The informed consent process will give the participant an accurate understanding of the research procedures and make it clear that participation is voluntary. The informed consent document will be read and reviewed with the participant. The consent form includes clear language with special attention to the provision of permissions to contact the participant and individuals they identify as acceptable contacts if the research assistants are unable to make direct contact with the participant for the 1- and 6-month assessments.

Screening assessments For those who remain interested following discussions and appear to meet eligibility criteria, the consent form will be signed and the screening visit will be conducted. This visit will include completion of assessments to confirm inclusion and exclusion criteria. The Nurse Practitioner will complete a medical history, a physical exam, a rapid blood test for HIV, a pregnancy test for women, and collect 10ml of blood for liver function tests. Research staff will complete baseline assessments. Those who are not eligible will receive a listing of treatment providers and encouraged to make contact for help with their substance use problems. For those who appear eligible following the screening visit, an appointment for initiating the intervention will be scheduled within 1 or 2 days.

Study Measures All interviews will be administered by experienced research assistants trained in the administration of the proposed assessments. Following confirmation of eligibility, the mobile research assistants will complete the baseline assessment (described below). The participant assessments in this research will be sharply focused on the valid and reliable measurement of its primary endpoints i.e. ongoing treatment engagement will be the primary endpoint on which the study is powered. The two groups will be compared with respect to the rate of continued opioid use and subsequent overdoses. All assessments will be completed by trained research staff (not clinicians) at baseline, and again at 1- and 6-month post-enrollment. A follow-up rates of 95% and 90% at Month 1 and 6, respectively is anticipated. The assessments will take approximately 45 minutes to complete.

- Physical history and examination including a blood test to evaluate the liver function.

- Locator Data Collection: A detailed contact sheet (Locator Form) will be completed for all subjects as part of the baseline assessment. The form will be updated at each scheduled assessment visit.

- DSM-5 Substance Use Disorder Checklist: This form is a listing of the eleven DSM-5 criteria for substance use disorder. Opioids, cocaine, and alcohol use disorders will be screened for confirmation of inclusion and exclusion criteria. A score of 4-6 is considered moderate severity and from 7-11 is considered severe substance use disorder. An opioid use disorder score of 4 or greater will be required for study eligibility. Alcohol use disorder and benzodiazepine scores of 4 or greater will be exclusionary.

- A study specific questionnaire that gathers socio-demographic information (e.g. age, gender, race and ethnic identity, marital status, sexual orientation, educational level, employment status, income and income sources) and descriptive information about the subject's living situation and housing stability, history of involvement with the criminal justice system, drug use, and drug treatment history.

o Addiction Severity Index (ASI) sections for Drugs and Alcohol and Medical Status: The ASI is a structured interview developed to assess the range of problems seen in drug users. The ASI produces severity ratings and composite scores in each of seven areas, and each type score has been assessed with regard to validity and reliability. Severity ratings and composite scores have demonstrated high levels of inter-rater, test-retest, and concurrent reliability (Cacciola et al., 2008; McLellan, Cacciola, Alterman, Rikoon, & Carise, 2006; McLellan et al., 1992). The Medical and the Drug and Alcohol sections will be administered at baseline, 1- and 6-month follow-ups.

- Patient Health Questionnaire (PHQ-9): Given the important role of depression in accessing and adhering to treatments for substance use disorder and other chronic medical problems (Mahajan, Avasthi, Grover, & Chawla, 2014; Tavakkoli, Ferrando, Rabkin, Marks, & Talal, 2013), the severity of depression using the 9-item Patient Health Questionnaire (PHQ-9) will be assessed. This brief assessment asks participants to indicate the frequency of occurrence of each of the nine DSM-IV diagnostic criteria for depression during the previous two weeks. This instrument has been widely used in clinical research and has strong psychometric properties. The PHQ-9 will be completed at baseline and month-1 and -6.

- Urine drug screen using the CLIA waived® ACON Dip-and-Read 8-panel test for THC, cocaine, opiates, amphetamines, PCP, methamphetamine, benzodiazepines, barbiturates, and buprenorphine. The urine sample will be provided using standard procedures with temperature monitoring to preclude tampering and dilution.

- HIV testing: Chembio SURE CHECK HIV 1/2® rapid-HIV test kits will be used to detect antibodies. These FDA approved and CLIA Waived kits produce results in 15 minutes from a blood drop. For those who test antibody positive, additional blood will be drawn for confirmatory testing via RNA viral load assay. Approximately 10% of participants will test positive for HIV.

- Economic evaluation: The cost-effectiveness of the intervention will be performed using data collected with questionnaires widely used for cost-effectiveness analyses, the Drug Abuse Treatment Cost Analysis Program (DATCAP) and Non-Medical and Other Services Form (NSMOS). The DATCAP is a reliable instrument widely used by substance abuse treatment programs for the collection and organization of programmatic costs (French et al. 1997). The NSMOS is a questionnaire adapted from the Treatment Services Review (McLellan et al., 1992; Cacciola et al., 2008 ) for individuals receiving substance abuse treatment.

Overview of Interventions Rapid Induction Intervention (RI) The rapid induction intervention will be delivered by the project's mobile team led by a nurse practitioner supported by a certified peer recovery specialist (CPRS), and a case manager (CM). Team members will work closely with each other and coordinate screening and treatment initiation, medication adherence monitoring, and assessment and linkage to ongoing MAT. Two half-time research assistants (RA) will participate as team members but will have intervention responsibilities. The RAs will assist with the community awareness campaign and recruitment and will conduct all assessments for the project. The mobile team will be based at Prevention Point Philadelphia and will receive clinical supervision from the mobile team's buprenorphine physician.

Induction: Once eligibility has been determined, the NP will begin preparing the participant for induction, following the buprenorphine/naloxone induction protocol used by D'Onofrio et al., 2015.

Medication Assisted Treatment (MAT) engagement during the month of intervention will be facilitated by the Peer Recovery Specialist (PRS). The purpose of a PRS is to support the participant in their recovery process. In addition, the project will provide intensive training and ongoing consultation on MAT. The PRS hired for this study will also receive intensive training focused on addiction as a chronic medical condition, motivational enhancement strategies and facilitating short-term behavioral contracts. The PRS will meet with the participant a minimum of 3 times per week and be available for telephone counseling and support throughout the one-month intervention period.

The mobile team's Case Management (CM) function will be performed by a master's level social worker. The CM will have primary responsibility for identifying the treatment program that best meets the needs of the participant. The CM effort will be sharply focused on identifying and removing barriers to MAT treatment engagement. This will include completing medication authorization forms and insurance enrollment if necessary, and then making referrals to additional supports, including mental health treatment, housing, medical, legal support, and peer-based recovery supports in the community. The CM will work collaboratively with participants to remove barriers to care and to enhance program outcomes and to provide psychosocial support.

The mobile team will insure that there is a clear plan for rapid engagement with ongoing treatment as soon as possible. Buprenorphine/naloxone will be provided during the time of transition to their long-term provider for up to 1 month. Telephone support will be available should participants have questions or need assistance and the mobile team will also have daily contact with participants until they complete contact with their ongoing provider.

Comparison group - Treatment as usual During the same period of time, participants will also be recruited at the City of Philadelphia Behavioral Health System Behavioral Assessment Centers (BAC) and Crisis Response Centers (CRC) located at Episcopal Hospital serving Kensington area and Hall Mercer BAC/CRC (a University of Pennsylvania program) serving South Philadelphia area. Two research assistants from the TRI-PHMC will enroll and follow-up 125 participants at BAC/ CRC. The participants will complete the same assessments at baseline, 1- and 6-month follow-up. Authorization for the Philadelphia Department of Public Health to look shared information about their subsequent use of health department services (CARES database, http://www.phila.gov/hhs/data/Pages/Cares.aspx) will also be requested in the consent form. Research staff will only conduct assessments. Treatment engagement services will be provided by the staff of the BAC/CRC as usual.

Subject Withdrawal Participants may choose to withdraw from the study at any time. They do this by providing verbal or written communication to this effect. Withdrawal from the study will not impact access to care. The Principal Investigator may withdraw subjects for reasons related to safety or for administrative reasons. It will be documented whether or not each subject completes the study. Subjects who do not complete the intervention will continue to be contacted to complete follow-up visits (Month 1 and 6) to collect final evaluations and to assess adverse events.

3. Statistical Plan 3.1. Data Quality Trained staff will review all forms prior to the completion of the study visit in an effort to minimize problems associated with missing values and incorrect skip patterns. All assessments will be recorded on final version of the electronic Case Reporting Forms (eCRFs). eCRFs will also be developed to capture the results of all biological assessments. All eCRFs will be completed using only participant identifying numbers and will not include names, addresses, or other data that could possibly be used to disclose the identity of the participant. Research staff will be responsible for entering all data into the secure web-based developed on REDCap (Research Electronic Data Capture). REDCap is designed to comply with HIPAA regulations. Data will be input with secure web authentication, data logging, and Secure Sockets Layer (SSL) encryption. REDCap allows multisite access. It allows real-time data entry validation (e.g. for data types and range checks), audit trails, and the ability to set up a calendar to schedule and track critical study events such as participant visits.

The data quality control process associated with data processing will consist of the following stages: potential subject pre-screening for eligibility, registration, subject eligibility confirmation, first data entry, second data entry in the form of 100% interactive verification, data validation, and data auditing. A random 5% of the questionnaires in each wave of data collection (i.e., baseline, 1- and 6-month) will be subjected to database auditing throughout the trial. This data quality control process ensures that all stages of the data handling process will be subjected to data quality control.

3.2. Data analyses Propensity score model to account for non-randomized allocation: The participants are not randomly assigned to the mobile rapid induction intervention and BAC/CRC groups, so it is possible that there will be confounding of the group effect with the effects of characteristics of the participants. Propensity score approach (Rosenbaum 2002, Rothman, Greenland et al. 2008) will be used to account for this. A broad range of baseline variables will be entered as covariates in a logistic regression model predicting treatment group, including age, race/ethnicity, gender, number of previous overdoses, number of previous treatments for opiate use disorder, type of opioid used, and depression. The model will yield predicted probabilities of BAC/CRC group for all participants (the propensity score), and a five-level ordinal variable based on the quintiles of the propensity score will be created. This ordinal variable will be included as a stratum variable in our models.

Analyses for the primary aim: i.e. to evaluate the impact of the mobile, transitional MAT intervention on its ability to engage participants in targeted, existing MAT treatment slots at 1- and 6-month post-enrollment. For each of the 1- and 6-month time points, the participants will be classified as being engaged/not-engaged (where non-engaged will include dropped out or missing). The groups will be compared on engagement rates at the two time points using a repeated measures logistic regression model, incorporating the propensity score as described above (Diggle, Heagerty et al. 2002). The fixed effects will be binary indicators of treatment group and time point; test for a group by time interaction to assess whether the group effect differs at month 1 and month 6. As there are only two time points, the use an unstructured covariance matrix to accommodate within-participant covariance.

Analyses for Aim 2: Repeated measures model will be run to compare the groups on the ASI (addiction severity index) drug composite score at month-1 and 6. The model will be similar to that described for the engagement analyses of Aim 1, but with a beta distribution assumed for the response, to accommodate the bounded range (zero to one) of the ASI composite scores. The groups will be compared on the number of overdoses per participants using repeated measures Poisson regression models, if there is sufficient variability in the number of overdoses reported. If a low rate of overdoses causes convergence issues for the Poisson model, participants will be classified as having no overdose versus at least one, and compare the groups using repeated measures logistic regression models.

Missing data. For the longitudinal analyses described above, premature discontinuation from treatment and occasional missing daily use indicators will lead to incomplete data. The repeated measures models described above can make use of all available data provided by subjects, but the inferences drawn from them will be unaffected by the missing data only if the missing data can be regarded as ignorable, essentially meaning that missingness can be predicted/explained from the baseline data and from responses obtained prior to drop-out. The sensitivity of our main analyses to the presence of missing data will be assessed by performing a sensitivity analysis, in which estimates of the group effect will be obtained under various non-ignorability assumptions. Selection models will be used to examine the effects of missing data (Robins, Rotnitzky et al. 1995), in which the probability of premature discontinuation at a time point as a function of baseline characteristics and responses at previous time points will be modeled using logistic regression models, and incorporate the predicted probabilities into a weighted analysis of the main hypotheses.

Analyses for Aim 3: The total costs and per-participant costs of rapid initiation of MAT for opioid treatment engagement will be estimated using a cost-offsets approach to determine the economic value of this novel strategy. A cost-offsets approach requires a comparison of the total cost of the intervention (proposed enhanced procedure or treatment as usual) to the future costs of healthcare utilization avoided (i.e., benefits) due to the intervention. Healthcare utilization and medical costs will be estimated before and after participants receive either rapid initiation of MAT for opioid treatment engagement or treatment as usual. Information on healthcare utilization will be collected from participants at each assessment when study-staff administer the Non-Medical and Other Services Form (NSMOS; detailed description in Instruments section). Medical costs associated with these treatments/services will be estimated using Medicaid data.

4. Safety and Adverse Events 4.1. Data Safety Monitoring Plan (DSMP) For this study, the University of Pennsylvania established procedures and infrastructure for data and safety monitoring will be used. During the course of the study, safety and data quality monitoring will be performed on an ongoing basis by the Principal Investigators and the study staff. Study staff members are responsible for collecting and recording all clinical data using the established MOP. This includes ensuring that all source documents exist for the data on the Case Report Forms, ensuring all fields are completed appropriately, and ensuring that all corrections are done according to Good Clinical Practice (GCP). Any inconsistencies/deviations will be documented. The study Key Personnel will review data on an ongoing basis and will document reviews by initialing and dating reports. Study staff members conduct 100% quality assurance on data.

Staff training will consist of an explanation of the protocol and review of the e-Case Report Forms. In addition, the duties of each staff person will be outlined and all applicable regulations will be reviewed. Senior personnel will supervise junior staff and provide re-training in the study protocol as needed.

The Independent Monitor for this study is the University of Pennsylvania Center for Studies on Addiction DSMB. The DSMB will review the study every 6 months for all the duration of the study. The DSMB form could be find in Appendix at the end of this document. The PI will provide a summary of the status of the project that occurs during the reporting period. These data will also be reported to CDC on an annual basis as part of the progress report. The DSMB report will include the participants' socio-demographic characteristics, expected versus actual recruitment rates, treatment retention rates, any quality assurance or regulatory issues that occurred during the past year, summary of SAEs, and any actions or changes with respect to the protocol.

4.2. Adverse Events Adverse events that occur at any point in the trial will be identified, managed, and documented in accordance with reporting requirements of the the City IRB (the IRB of record) and the sponsor (the CDC). Adverse events are defined as negative biologic events and/or social harms that occur during the course of the trial. When needed, referrals to medical treatment or specialists will be made. All patients will have comprehensive psychiatric and medical screening prior to randomization and at each counseling and assessment where evaluations for AEs will be routine. A member of the research team will be available at all times to answer questions and assess possible AEs. Participants will be withdrawn from the study if they show severe deterioration or if determined clinically necessary for other reasons.

4.3 Internal Monitoring and Auditing

Oversight and Monitoring: The City of Philadelphia IRB will monitor the protection of human subjects and the safe and secure collection and storage of data. This committee assesses all studies before study initiation and then reviews protocols annually. The committee ensures the scientific, technical, and statistical soundness of the research and guarantees that methods for the ethical and safe treatment of human subjects are in place. The committee scrutinizes the scientific and ethical aspects of protocols and provides for an objective and ongoing assessment of the study's scientific and ethical integrity.

Protocol monitoring will ensure that the research protocol specified is being followed without unauthorized deviations. Weekly meetings will be held to monitor the progress of the trial. These meetings will involve the Principal Investigators, Co-Investigators, Project Director, Site coordinator and other study staff. This will help to ensure standardized application of the protocol and will serve as an ongoing mechanism by which project staff and investigators will communicate in order to maintain a consistently high quality of study conduct. Concerns identified will be addressed through training and retraining of personnel.

Database Auditing: Project director and RAs will review data entered into the database versus that recorded on the CRFs. All accrued cases will be subjected to database auditing throughout the duration of the trial. Depending on the data management findings, re-training will be provided, should problems such as increased errors be detected.

Data Auditing: Project director and staff RAs will review safety data recorded on the CRF versus that contained on the actual source document (client chart, EHR). All accrued cases will be subjected to auditing throughout the duration of the trial. A Regulatory Binder Review by OHR will include the following essential documents: IRB Protocol, Consent Form and Amendment Approvals, IRB Closure Letter, List of Authorized Signatures, Laboratory Certifications, Protocol and Amendment Signature Pages, Financial Disclosure Questionnaires, and Monitoring Log. Additional monitoring by OHR may include: source documentation verification; adverse event documentation; and facility assessment.

Data Security: Using network firewall technologies, the database will prevent the three major sources of data security problems: unauthorized internal access to data, external access to data, and malicious intent to destroy data and systems. Controlled user access will ensure that only appropriate and authorized personnel are able to view, access, and modify trial data. All modifications to data will document user access and data associated with the modification, as well as values prior to modification.

Evidence of Training in Human Subject Research: All personnel working on this project will be required to review the protocol, complete training in the protection of human subjects and undergo training.

Confidentiality Since self-report and medical data will be collected and stored as part of this study, it is possible that subject privacy or confidentiality can be threatened. To address this concern, the Data Management System has set up several safeguards to prevent unauthorized access to study data. An automatically generated index number is assigned to a subject's study identification number (unique for personnel and clients). A linked subject identification table is created for the storing of subject name, address and telephone contact information. This table uses the automatically generated index number rather than the study identification number. The master subject map and subject identification information tables are maintained in a separate database. Using this method, no identifying subject information is directly linked to medical information or other study data. The present research team has not experienced the unauthorized use of study data. A web-based data collection procedure will minimize the possibility of loss of privacy or confidentiality. The risk of a potential breach of confidentiality is addressed in the informed consent documents.

Computers and Databases Trained staff will review all forms prior to the completion of the study visit in an effort to minimize problems associated with missing values and incorrect skip patterns. All assessments will be recorded on final version of the electronic Case Reporting Forms (eCRFs). eCRFs will also be developed to capture the results of all biological assessments. All eCRFs will be completed using only participant identifying numbers and will not include names, addresses, or other data that could possibly be used to disclose the identity of the participant. Research staff will be responsible for entering all data into the secure web-based developed on REDCap (Research Electronic Data Capture). REDCap is designed to comply with HIPAA regulations. Data will be input with secure web authentication, data logging, and Secure Sockets Layer (SSL) encryption. REDCap allows multisite access. It allows real-time data entry validation (e.g. for data types and range checks), audit trails, and the ability to set up a calendar to schedule and track critical study events such as participant visits.