Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Jun 1.
Published in final edited form as: Drug Alcohol Depend. 2024 Apr 5;259:111286. doi: 10.1016/j.drugalcdep.2024.111286

Effect of the Communities That HEAL intervention on receipt of behavioral therapies for opioid use disorder: A cluster randomized wait-list controlled trial

LaShawn Glasgow 1, Christian Douglas 1, Joel G Sprunger 2, Aimee N C Campbell 3, Redonna Chandler 4, Anindita Dasgupta 3, JaNae Holloway 1, Katherine R Marks 5, Sara M Roberts 6, Linda Sprague Martinez 7, Katherine Thompson 8, Roger D Weiss 9, Arnie Aldridge 1, Kat Asman 1, Carolina Barbosa 1, Derek Blevins 10, Deborah Chassler 7, Lindsay Cogan 11, Laura Fanucchi 12, Megan E Hall 1, Timothy Hunt 13, Elizabeth Jadovich 14, Frances R Levin 10, Patricia Lincourt 15, Michelle R Lofwall 12, Vanessa Loukas 14, Ann Scheck McAlearney 16, Edward Nunes 3, Emmanuel Oga 1, Devin Oller 12, Maria Rudorf 14, Ann Marie Sullivan 17, Jeffery Talbert 12, Angela Taylor 12, Julie Teater 16, Nathan Vandergrift 1, Kristin Woodlock 18, Gary A Zarkin 1, Bridget Freisthler 19,*, Jeffrey Samet 20,*, Sharon L Walsh 12,*, Nabila El-Bassel 13,*
PMCID: PMC11111326  NIHMSID: NIHMS1988390  PMID: 38626553

Abstract

Background:

The U.S. opioid overdose crisis persists. Outpatient behavioral health services (BHS) are essential components of a comprehensive response to opioid use disorder and overdose fatalities. The Helping End Addiction Long-Term (HEALing) Communities Study developed the Communities That HEAL (CTH) intervention to reduce opioid overdose deaths in 67 communities in Kentucky, Ohio, New York, and Massachusetts through the implementation of evidence-based practices (EBPs), including BHS. This paper compares the rate of individuals receiving outpatient BHS in Wave 1 intervention communities (n = 34) to waitlisted Wave 2 communities (n = 33).

Methods:

Medicaid data included individuals ≥18 years of age receiving any of five BHS categories: intensive outpatient, outpatient, case management, peer support, and case management or peer support. Negative binomial regression models estimated the rate of receiving each BHS for Wave 1 and Wave 2. Effect modification analyses evaluated changes in the effect of the CTH intervention between Wave 1 and Wave 2 by research site, rurality, age, sex, and race/ethnicity.

Results:

No significant differences were detected between intervention and waitlisted communities in the rate of individuals receiving any of the five BHS categories. None of the interaction effects used to test the effect modification were significant.

Conclusions:

Several factors should be considered when interpreting results—no significant intervention effects were observed through Medicaid claims data, the best available data source but limited in terms of capturing individuals reached by the intervention. Also, the 12-month evaluation window may have been too brief to see improved outcomes considering the time required to stand-up BHS.

Keywords: opioids, overdose, behavioral health, community intervention, retention, Medicaid

1. INTRODUCTION

The United States continues to experience an unprecedented drug overdose epidemic, with synthetic opioids—primarily fentanyl—involved in the majority of overdose deaths (Hoots, 2021). Provisional national data from the National Center for Health Statistics project more than 108,000 overdose deaths for the 12-month period ending November 2022, with approximately 81,000 overdose deaths involving opioids (Ahmad et al., 2023). Sociodemographic disparities in fatal overdoses are widening, with rates increasing by approximately 40% among non-Hispanic Black and non-Hispanic American Indian or Alaska Native (AI/AN) persons from 2019 to 2020 (Han et al., 2022; Kariisa et al., 2022; Larochelle et al., 2021). Beneficiaries of Medicaid, which is one of the nation’s largest health care payers and provides health coverage for people with low income, are at higher risk for opioid overdose and overdose fatalities (Garg et al., 2017; MACPAC, 2017; Saunders, 2023). Individuals who have opioid use disorder (OUD) with co-occurring mental illness are also at increased risk for morbidity and mortality (Jones and McCance-Katz, 2019; Novak et al., 2019).

Pairing medication with counseling and behavioral therapies is a critical “whole patient” approach to treating substance use disorders (SUD) (Centers for Disease Control and Prevention, 2022; Substance Abuse and Mental Health Services Administration, 2023). Methadone, buprenorphine, and extended-release naltrexone are U.S. Food and Drug Administration (FDA)-approved medications for opioid use disorder (MOUD) (National Institute on Drug Abuse (NIDA), 2021). However, these medications are underutilized, with only 22% of people diagnosed with OUD receiving treatment in the last year (Jones et al., 2023; Mauro et al., 2022; Substance Abuse and Mental Health Services Administration, 2022). Further, following successful MOUD initiation, many patients are not retained in care; retention rates generally fall within the 30–50% range within the first year (Chan et al., 2021; Timko et al., 2016). Addressing the opioid overdose crisis and related disparities requires widespread adoption of evidence-based behavioral health interventions (Novak et al., 2019; Ober et al., 2022).

1.1. Evidence-Based Behavioral Health Practices

Behavioral health interventions are important components of comprehensive and individualized treatment for addiction that help to address psychosocial functioning, mental health comorbidities, and quality of life (National Institute on Drug Abuse (NIDA), 2018). Further, providers in general health care settings may be more likely to deliver MOUD when behavioral health care is readily available or co-located with their program (Marino et al., 2019; McCollum et al., 2023). There is extensive evidence that behavioral interventions such as cognitive behavioral therapy, motivational interviewing, and relapse prevention improve SUD treatment engagement and retention rates and can reduce substance use (Jhanjee, 2014). Twelve-step facilitation and mutual–help support groups, often associated with specialty programs for addiction, can provide social support and encouragement to people with opioid and other SUDs (National Institute on Drug Abuse (NIDA), 2018), and there is some evidence that mutual-help groups—with and without agonist treatment—are associated with better long-term opioid abstinence outcomes(Weiss et al., 2019). Linking individuals to MOUD, along with utilizing the range of behavioral health supports and tools to enhance engagement in long-term treatment and recovery, can assist in reducing the risk of overdose and improve overall health outcomes.

1.2. HEALing Communities Study

The multisite Helping End Addiction Long Term (HEALing) Communities Study (HCS) assesses the effectiveness of the Communities That HEAL (CTH) intervention, which was designed to reduce opioid-involved overdose deaths in 67 highly impacted communities across four research sites (Walsh et al., 2020). The CTH core components are (1) the Opioid-overdose Reduction Continuum of Care Approach (ORCCA) (Winhusen et al., 2020), a compendium of evidence-based practices (EBPs) and resources developed to facilitate the implementation of interventions to reduce overdose deaths; (2) community engagement, including the use of community data dashboards, to increase adoption and support sustainability of EBPs (Sprague Martinez et al., 2020; Wu et al., 2020); and (3) health communication campaigns to reduce stigma associated with OUD and treatment and to drive demand for EBPs (Lefebvre et al., 2020) (). HCS communities were expected to implement all components of the CTH intervention; they were encouraged to implement EBPs in behavioral health, healthcare, and criminal legal settings and required to implement at least three MOUD strategies. The CTH intervention and ORCCA menu are described in detail in other publications (SAMHSA, 2023; Walsh et al., 2020).

The ORCCA compendium includes strategies for increasing linkage to MOUD treatment and retention within behavioral health settings, including addiction treatment and recovery facilities. ORCCA behavioral health strategies include enhancing clinical delivery approaches through case management, for example, to enhance patient engagement and support retention in MOUD treatment; implementing or expanding retention care coordinator programs; and integrating mental health and polysubstance use treatment into MOUD care. As part of the CTH intervention, HCS communities assessed the current landscape of OUD treatment and retention services, set priorities for addressing the opioid overdose crisis and related disparities, and selected specific evidence-based strategies to implement in behavioral health settings based on their unique community context. The research objective of this paper is to assess the impact of CTH on the rate of receiving services that align with ORCCA behavioral health strategies among Medicaid beneficiaries with OUD.

2. METHODS

Methods are summarized in accordance with the Consolidated Standards of Reporting Trials guidelines (Schulz et al., 2010). The CONSORT checklist and flow diagram are included in the supplemental files.

2.1. HCS Trial Design

The multisite HEALing Communities Study employs a parallel-group, community-level cluster randomized, unblinded, waitlist-controlled comparison design. The study population includes 67 communities across four research sites: Kentucky (KY), Massachusetts (MA), New York (NY), and Ohio (OH). More than 10.1 million people reside in the HCS communities, which consist of counties (N = 48) or cities/towns and city/town clusters (N = 19; 16 in MA and three in NY). Communities are the unit of analysis for the study and were assigned to study arms based on a covariate constrained randomization procedure conducted by the data coordination center. Urban/rural classification, opioid overdose death rate, and community population variables were used for constraints. There were 34 communities assigned to Wave 1 (intervention arm) and 33 communities assigned to Wave 2 (waitlist control arm). Each community had an equal probability of being allocated to either arm. One community withdrew from the study after randomization and before CTH implementation began.

The study protocol (Pro00038088) was approved by Advarra Inc., the HCS single Institutional Review Board (sIRB). HCS was granted a Waiver of Consent and a Full Waiver of HIPAA Authorization for secondary data analysis from the sIRB (Advarra, 10/25/2019, MOD00521925). The National Institute on Drug Abuse (NIDA) chartered an independent data and safety monitoring board to oversee the trial. See previous publications for additional information on HCS procedures and the CTH intervention, including an overview of the study’s conceptualization (Chandler et al., 2020), the HCS protocol (Walsh et al., 2020), a detailed description of outcome measures (Slavova et al., 2020), and the approach for adapting the RE-AIM/PRISM framework to assess implementation context and monitor fidelity to the CTH intervention (Knudsen et al., 2020)

2.2. Research Objective and Statistical Methods

The research objective of this paper is to compare Waves 1 and 2 on the rate of receiving the following behavioral health services among adult Medicaid recipients age 18 or older with OUD, irrespective of MOUD receipt: intensive outpatient (American Society of Addiction Medicine [ASAM] level 2), outpatient (ASAM level 1), case management, peer support, and case management or peer support (Guyer, 2021). The application of ASAM levels and revenue, place of service, and procedure codes were reviewed across all four research sites to maximize alignment around the services covered under each outcome (see supplemental files for measures specifications). The rate of receiving case management and peer support among adult Medicaid recipients were measures identified for the ORCCA EBPs, “Enhance clinical delivery approaches that support engagement and retention” and “Utilize retention care coordinators.” The intensive outpatient, outpatient, case management and peer support Medicaid measures all mapped to the “Mental health and polysubstance use treatment integrated into MOUD care” EBP in the ORCCA menu. The time period for analyses corresponds to the comparison period for the CTH intervention (July 1, 2021, through June 30, 2022), which was the active intervention period for Wave 1 and pre-intervention period for Wave 2.

The models used to address the research objectives utilize methods described by Westgate at al. (2022). For each objective, a negative binomial regression model was fitted to estimate the rate of receiving behavioral health services among adult Medicaid recipients for each arm (Wave 1 and Wave 2). The estimated rates were adjusted by research site, rurality (rural/urban), observed baseline community-level opioid overdose mortality rate, and the observed baseline community-level rate of the behavioral health service being modeled (or natural log of the observed baseline rate, depending on the health service). Small-sample adjusted empirical standard error estimates were used to construct inferential statistics, 95% CIs, and p-values (Mancl and DeRouen, 2001). All adjusted rates and 95% CIs were calculated using least square means. Missing data arising from suppression was imputed in the primary models using multiple imputation. The models were estimated using the intention-to-treat (ITT) population for the primary analyses and the per protocol (PP) population for sensitivity analyses.

Effect modification (subgroup) analyses to evaluate any changes in the effect of the CTH intervention between Wave 1 and Wave 2 communities were conducted for all outcomes for the following potential effect modifiers: research site, rurality, age (18–34 years old, 35–54 years old, and 55–64 years old), sex (male and female), and race/ethnicity (Hispanic, non-Hispanic white, non-Hispanic Black, and non-Hispanic other). For each potential effect modifier, a separate model was fitted using the ITT population. Each effect modification model contained the same covariates as the primary models, plus the corresponding potential modifier variable (if not part of the primary model), and an interaction between intervention and the modifier variable. If a group within a modifier variable was missing due to suppression, it was dropped from the analysis and not imputed.

To account for multiple comparisons arising from subgroup analyses, the false discovery rate (FDR) adjustment was applied to the p-values for the test of interaction between intervention and effect modifier using the Benjamini-Hochberg method (Benjamini et al., 2001). Estimates and p-values for pairwise tests between levels of a modifier were only reported if the FDR-adjusted p-value from the test of interaction was significant. Results were considered statistically significant for p-values < 0.05. All analyses were conducted using SAS Version 9.4.

2.3. Methods for Capturing EBP Implementation

HCS researchers developed a REDCap instrument, the ORCCA Tracker (ORCCAT), to document the EBPs HCS communities selected and implemented (Chandler et al., 2023). Staff at each research site entered EBP details—including menu (overdose education and naloxone distribution, MOUD, and safer prescribing and disposal), sector (health care, behavioral health, criminal legal) and venue (e.g., emergency departments, addiction treatment and recovery facilities, jails)—into site-specific REDCap databases, which were securely transferred to the data coordinating center for analysis. For this research objective, descriptive statistical analysis was conducted to determine the number and proportion of behavioral health EBPs implemented in outpatient settings by research site and overall. Analysis was limited to actively implemented EBPs (i.e., strategies that had advanced to the stage of delivering services to individuals intended to benefit from them).

3. RESULTS

3.1. Baseline Characteristics

Baseline characteristics used in the constrained randomization (rurality, rate of opioid overdose deaths, and population) are displayed in Table 1 and support expected balance between the study arms. The baseline distribution of age, sex, and race/ethnicity of residents 18 years of age and older were similar between Wave 1 and Wave 2. Wave 1 had a slightly higher percentage of non-Hispanic Black individuals than Wave 2 (16.4% versus 14.5%), and Wave 2 had a slightly higher percentage of Hispanic individuals than Wave 1 (8.6% versus 6.3%). The median baseline rates of adult Medicaid recipients with OUD receiving behavioral health (BH) services were similar between Wave 1 and Wave 2 for intensive outpatient (IOP) treatment (55.9 per 1,000 adult Medicaid recipients versus 53.2 per 1,000 adult Medicaid recipients), outpatient treatment (680.3 per 1,000 adult Medicaid recipients versus 681.1 per 1,000 adult Medicaid recipients), and case management (177.2 per 1,000 adult Medicaid recipients versus 176.4 per 1,000 adult Medicaid recipients). The baseline median rate of adult Medicaid recipients with OUD receiving BH via peer support was higher in Wave 2 than Wave 1 (51.0 per 1,000 adult Medicaid recipients versus 44.3 per 1,000 adult Medicaid recipients). However, the baseline median rate of adult Medicaid recipients with OUD receiving BH via any case management or peer support was higher in Wave 1 than Wave 2 (241.4 per 1,000 adult Medicaid recipients versus 193.4 per 1,000 adult Medicaid recipients).

Table 1.

Baseline Demographic Characteristics of N=67 Communities Participating in the HEALing Communities Study, by Intervention Wave

Intervention Wave Overall
Characteristic, statistic Wave 1 Wave 2
Number of Randomized Communities 34 33 67
Research Site, n(%)
 Kentucky 8 (23.5%) 8 (24.2%) 16 (23.9%)
 Massachusetts 8 (23.5%) 8 (24.2%) 16 (23.9%)
 New York 8 (23.5%) 8 (24.2%) 16 (23.9%)
 Ohio 10 (29.4%) 9 (27.3%) 19 (28.4%)
Urban-Rural Classification, n(%)
 Urban 19 (55.9%) 19 (57.6%) 38 (56.7%)
 Rural 15 (44.1%) 14 (42.4%) 29 (43.3%)
Population Aged 18+ 1 4,439,170 3,772,336 8,211,506
Age 1 , n(%)
 18–34 Years 1,334,880 (30.1%) 1,178,210 (31.2%) 2,513,090 (30.6%)
 35–54 Years 1,353,341 (30.5%) 1,180,392 (31.3%) 2,533,733 (30.9%)
 55+ Years 1,750,949 (39.4%) 1,413,734 (37.5%) 3,164,683 (38.5%)
Race/Ethnicity 1 , n(%)
 Non-Hispanic White 3,229,233 (72.7%) 2,750,369 (72.9%) 5,979,602 (72.8%)
 Non-Hispanic Black 728,037 (16.4%) 545,357 (14.5%) 1,273,394 (15.5%)
 Non-Hispanic Other 200,571 (4.5%) 153,956 (4.1%) 354,527 (4.3%)
 Hispanic 281,329 (6.3%) 322,654 (8.6%) 603,983 (7.4%)
Sex 1 , n(%)
 Male 2,133,827 (48.1%) 1,825,776 (48.4%) 3,959,603 (48.2%)
 Female 2,305,343 (51.9%) 1,946,560 (51.6%) 4,251,903 (51.8%)
Rate of Opioid Overdose Deaths 2
 Mean (SD) 38.2 (22.8) 37.1 (20.3) 37.7 (21.4)
 Median (Q1, Q3) 35.2 (21.6, 49.3) 32.7 (23.6, 48.6) 34.6 (22.9, 49.1)
 Missing due to Suppression, n(%) 0 (0%) 0 (0%) 0 (0%)
Rate of Adult Medicaid recipients with OUD receiving BH treatment (IOP, ASAM level 2) 3
 Mean (SD) 84.3 (66.9) 71.7 (57.6) 78.1 (62.3)
 Median (Q1, Q3) 55.9 (42.2, 104.1) 53.2 (32.5, 97.1) 55.4 (40.4, 101.6)
 Missing due to Suppression, n(%) 1 (2.9%) 1 (3.0%) 2 (3.0%)
Rate of Adult Medicaid recipients with OUD receiving BH treatment (outpatient, ASAM level 1) 3
 Mean (SD) 707.1 (118.1) 693.6 (155.4) 700.5 (136.9)
 Median (Q1, Q3) 680.3 (624.9, 777.1) 681.1 (610.8, 747.8) 681.1 (612.2, 777.1)
 Missing due to Suppression, n(%) 0 (0%) 0 (0%) 0 (0%)
Rate of Adult Medicaid recipients with OUD receiving BH treatment (case management) 3
 Mean (SD) 196.7 (113.7) 172.1 (101.3) 184.8 (107.8)
 Median (Q1, Q3) 177.2 (90.8, 306.4) 176.4 (97.7, 234.3) 176.4 (94.3, 258.8)
 Missing due to Suppression, n(%) 0 (0%) 1 (3.0%) 1 (1.5%)
Rate of Adult Medicaid recipients with OUD receiving BH (peer support) 3
 Mean (SD) 85.1 (95.8) 99.1 (120.5) 91.9 (107.8)
 Median (Q1, Q3) 44.3 (16.0, 140.0) 51.0 (18.9, 107.5) 49.3 (16.8, 131.7)
 Missing due to Suppression, n(%) 2 (5.9%) 3 (9.1%) 5 (7.5%)
Rate of Adult Medicaid recipients with OUD receiving BH treatment (any of case management or peer support) 3
 Mean (SD) 235.1 (122.9) 213.0 (130.5) 224.2 (126.2)
 Median (Q1, Q3) 241.4 (112.2, 324.5) 193.4 (143.8, 267.5) 214.3 (112.2, 310.0)
 Missing due to Suppression, n(%) 0 (0%) 0 (0%) 0 (0%)
Open in a new tab

%: Percentages may not add up to 100 due to rounding

1

For communities that represent counties (n=48 of 67), population estimates are from 2020 Bridged-Race Population Estimates retrieved via https://www.cdc.gov/nchs/nvss/bridged_race.htm on May 13, 2023. For communities that represent units smaller than counties (n=19 of 67), population estimates are from 2017–2021 American Community Survey 5-Year Estimates retrieved via https://data.census.gov/cedsci on May 13, 2023.

2

Rate per 100,000 community residents aged 18+ calculated as the number of events measured from January 2019 - December 2019 divided by the observed community population of individuals aged 18+ from the 2020 Bridged-Race Population Estimates or the 2017–2021 American Community Survey 5-Year Estimates, multiplied by 100,000.

3

Rate per 1,000 Medicaid recipients aged 18–64 with an OUD diagnosis calculated as the number of events measured from January 2019 - December 2019 divided by the observed number of Medicaid recipients aged 18–64 with an OUD diagnosis between January 2019 - December 2019, multiplied by 1,000.

3.2. EBP Implementation

HCS communities fully implemented a total of 256 MOUD-related EBPs in Wave 1 (Table 2). Twenty-four (9%) of the total 256 MOUD-related EBPs are strategies intended to affect rates of receiving the BHS assessed in this study, such as use of retention care coordinators. Most (87.5%) of the 24 relevant EBPs were implemented in addiction treatment or recovery facilities.

Table 2.

Implemented EBPs from ORCCAT Menu 2: MOUD by Research Site for N = 33* Wave 1 Communities Participating in HCS†

Research Site Rural/Urban Total
KY MA NY OH Rural Urban
Number of Communities 8 8 8 9 15 18 33
Total MOUD EBPs Implemented, n 92 52 55 57 102 154 256
Total MOUD EBPs Related to BH Services Examined in Study, n 17 1 1 5 12 12 24
 Enhancement of Clinical Delivery Approaches That Support Engagement and Retention 9 (52.9%) 1 (100.0%) 1 (100.0%) 5 (100.0%) 8 (66.7%) 8 (66.7%) 16 (66.7%)
 Use of Retention Care Coordinators 8 (47.1%) 0 (0%) 0 (0%) 0 (0%) 4 (33.3%) 4 (33.3%) 8 (33.3%)
 Mental Health and Polysubstance Use Integration into MOUD Treatment 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%)
Venue for EBPs Related to BH Services, n(%)
 Addiction Treatment and Recovery Facilities 16 (94.1%) 1 (100.0%) 1 (100.0%) 3 (60.0%) 10 (83.3%) 11 (91.7%) 21 (87.5%)
 Mental/Behavioral Health Treatment Facilities 1 (5.9%) 0 (0%) 0 (0%) 2 (40.0%) 2 (16.7%) 1 (8.3%) 3 (12.5%)
Open in a new tab
*

n=1 community randomized to Wave 1 withdrew prior to strategy selection.

†

Results based on data pulled March 27, 2023.

3.3. Statistical Results

The average raw rate ratios between Wave 1 and Wave 2 of Medicaid recipients with OUD receiving BH via IOP, outpatient, case management, peer support, or any case management or peer support during the evaluation period were 1.26, 1.01, 1.10, 0.89, and 1.04, respectively. The adjusted relative rates and 95% confidence intervals of Wave 1 versus Wave 2 that account for rurality, research site, baseline opioid overdose rate, and baseline outcome rate (or natural log of the baseline outcome rate) were 1.27 (CI 95%: 0.97, 1.65) for IOP; 1.00 (CI 95%: 0.97, 1.03) for outpatient; 0.99 (CI 95%: 0.85, 1.15) for case management; 1.05 (CI 95%: 0.83, 1.33) for peer support; and 1.01 (0.88, 1.15) for any case management or peer support (Table 3). No treatment effect was statistically significant for any of the outcomes. The sensitivity analysis completed on the PP population yielded similar results. The investigation of a difference in the treatment effect by subgroups did not indicate any significant difference for research site, rural/urban, age, sex, or race/ethnicity subgroups for any of the five outcomes.

Table 3.

Analysis of Efficacy Outcomes for Hypotheses 1–5 During the Evaluation Period (July 1, 2021–June 30, 2022) Using the ITT Population

Outcome Adjusted Rate (95% CI)4 Adjusted Relative Rate (95% CI)5 p-value % Community values Imputed
Wave 1 Wave 2
Adult Medicaid Recipients with OUD Receiving BH Treatment (IOP, ASAM level 2) 1, 3 54.3 (45.2, 65.1) 42.8 (34.7, 52.8) 1.27 (0.97, 1.65) 0.081 4.5
Adult Medicaid Recipients with OUD Receiving BH Treatment (outpatient, ASAM level 1) 2 727.6 (714.4, 741.1) 727.3 (710.0, 745.0) 1.00 (0.97, 1.03) 0.970 0
Adult Medicaid Recipients with OUD Receiving BH Treatment (case management) 1, 3 153.8 (140.9, 168.0) 156.0 (136.1, 178.8) 0.99 (0.85, 1.15) 0.857 1.5
Adult Medicaid Recipients with OUD Receiving BH (peer support) 1, 3 80.4 (68.5, 94.3) 76.5 (63.9, 91.6) 1.05 (0.83, 1.33) 0.681 7.5
Adult Medicaid Recipients with OUD Receiving BH Treatment (any of case management or peer support) 1 214.9 (200.5, 230.5) 213.3 (188.5, 241.5) 1.01 (0.88, 1.15) 0.913 0
Open in a new tab
1

Negative binomial model adjusting for rurality (urban, rural), research site (KY, MA, NY, OH), baseline opioid overdose death rate, and baseline outcome rate. Dispersion parameters (95% CI) for models: k=0.240 (IOP), k=0.077 (case management), k=0.195 (peer support), and k= 0.067 (0.046, 0.098) for case management or peer support.

2

Negative binomial model adjusting for rurality (urban, rural), research site (KY, MA, NY, OH), baseline opioid overdose death rate, and natural log of the baseline outcome rate. Dispersion parameter (95% CI): k=0.001 (0.001, 0.002)

3

Results based on imputed data.

4

Model estimated marginal event rate expressed as per 1,000 Medicaid recipients aged 18–64 with an OUD diagnosis.

5

Adjusted relative rate of Wave 1 communities over Wave 2 communities.

4. DISCUSSION

The present study investigated whether the CTH intervention increased rates of receipt for behavioral therapy interventions among adult beneficiaries of Medicaid with OUD—a population at higher risk for opioid-related overdose and overdose fatalities. Using the HCS multisite, parallel-group, community-level cluster randomized, unblinded, waitlist-controlled comparison design, we compared rates of receipt for intensive outpatient, outpatient, case management, peer support, and case management or peer support among Medicaid recipients between intervention (Wave 1) and waitlist control (Wave 2) communities. Accounting for rurality, research site, baseline opioid overdose death rate, and baseline rates for the outcomes, the results indicated comparable rates of receipt of all BHS in CTH intervention and waitlist communities. Results were consistent for the ITT and PP populations.

4.1. Interpretation of Findings

The effect of the CTH intervention on differences in rates of outpatient BHS was not statistically significant; however, findings must be interpreted within the context of the full intervention model and data source limitations. The CTH model employs a community-engaged and data-driven action planning process in which communities identified and prioritized EBPs that best aligned with their needs and strengths. MOUD-related EBPs in the ORCCA menu focused on (1) starting or expanding access to MOUD services, (2) effectively linking individuals at-risk to MOUD, and (3) retaining individuals in MOUD care. Ultimately, communities implemented a total of 256 MOUD-related strategies in Wave 1; however, only 9% of those were EBPs that mapped to Medicaid rates of intensive outpatient and outpatient care, case management, or peer support services in BH settings, as conceptualized in our study design (Table 2). Moreover, some communities did not implement any EBPs relevant to the BH outcomes in this study. The relatively low “dose” of relevant EBPs may help explain the nonsignificant findings. EBP implementation data indicate that HCS communities prioritized connecting individuals at high risk with new or expanded options for lifesaving MOUD to address opioid-related overdose deaths, which were rising during the overlapping COVID-19 pandemic and CTH intervention period (Table 2).

Another key consideration is the intervention’s focus on adoption of EBPs, not merely on the direct provision of the service. This approach involves changing policy and practice for existing service delivery organizations, which is a difficult undertaking requiring an understanding of the regulations and billing mechanisms that allow for integration of BHS and MOUD care. The heavy toll the COVID-19 pandemic exacted on resources and personnel, including an unprecedented behavioral health workforce shortage, likely exacerbated typical EBP adoption challenges (Gilbert et al., 2023).

4.2. Limitations

Despite design strengths, the study had some limitations that are important to take into account. First, the one-year evaluation window (July 1, 2021, through June 30, 2022) may have been too short to see a meaningful impact on BH outcomes as it takes time to scale-up and reach eligible clients with new services. All things considered, getting to EBP prioritization, adoption, and delivery of services to priority individuals are noteworthy accomplishments for HCS communities.

Notably, the study excluded rate of Medicaid recipients receiving inpatient services (ASAM levels 3, 4). Inpatient services were excluded because we were unable to extract BH services from the mix of codes for medical and behavioral services in our source data. For example, although withdrawal management (detoxification) can lead to MOUD initiation, it is more accurately classified as medical treatment rather than a BH service and, therefore, is outside of the scope of this analysis. Furthermore, inpatient services are typically oriented toward abstinence or withdrawal management and may not include MOUD maintenance treatment, a key focus of the CTH. Although excluding inpatient services from this study was an imperfect solution, the outcomes we included align with the intervention’s focus on enhancing the delivery of outpatient BH services.

Another constraint is that the data were limited to individuals receiving services charged to Medicaid with an OUD diagnosis tied to the encounter. Therefore, our study gaps include undiagnosed Medicaid beneficiaries and individuals receiving services from community-based social service agencies or other organizations that do not bill Medicaid. This exclusion also applies to organizations using study funds to support the start-up of peer support programs and that may not have submitted services to Medicaid for reimbursement. Additionally, peer support services that individuals seek on their own were not included in our analyses. Nor does this study capture the delivery of outpatient BHS to individuals who are uninsured or insured by other payors. Notably, all-payor claims databases, the best data source for testing the effectiveness of the CTH intervention on receipt of BHS, do not exist in all four HCS study sites.

Beyond insurance-related gaps, the data allowed for effect modification analyses, but were insufficient for additional subgroup analyses to investigate health inequities. NIDA leaders have shared HCS lessons about the limitations of administrative data sources, including Medicaid claims data, and have called for more comprehensive, timely and disaggregated data to inform and monitor our nation’s response to the opioid overdose epidemic (Volkow et al., 2022). Lastly, although we have EBP implementation data, which provide some measure of adoption of EBPs, detailed reach data are not available. Therefore, we cannot explore the extent to which limited reach may help explain results, and we cannot assess reach for different population groups. Over the course of the study, HCS researchers developed a REDCap instrument to capture available reach data, and our approach and lessons will be shared in future publications.

4.3. Conclusion

We set out to examine the impact of the CTH intervention on the rate of receiving ORCCA-related BHS among Medicaid beneficiaries with OUD. No significant intervention effects were observed through Medicaid claims data, which was the best available data source for our analyses but admittedly limited in terms of capturing individuals reached by the CTH behavioral health EBPs. Contextual factors that may help explain the nonsignificant findings include a relatively short (12-month) evaluation window for complex EBPs that require time to stand-up and increase reach, as well as communities’ data-driven prioritization of EBPs that focused on connecting individuals at high-risk for opioid-related overdose deaths with new or expanded options for MOUD, rather than those EBPs specifically classified as behavioral health and relevant to BH outcomes addressed in this study. Findings point to the complexity of testing the impacts of community-engaged interventions.

Although no significant intervention effects were observed, the scale and breadth of the HCS created an unprecedented opportunity to better understand how communities prioritize, integrate, and tailor EBPs to address the opioid crisis; the EBP implementation challenges communities face; and promising approaches for evaluating the implementation and impact of comprehensive community-driven interventions designed to reduce opioid overdose deaths. These insights can help strengthen the design of future large multi-site trials, particularly those with community engagement components. For example, HCS has elucidated gaps in administrative datasets that are commonly used in addiction research. Future research may prioritize sites with access to all-payor claims databases or incorporate a sub-study focused on the development and validation of de novo data sources for documenting the reach (including demographics) and dose of BHS across participating service delivery organizations.

The CTH intervention was designed with community engagement as a core component to align EBP adoption with communities’ intervention needs and resources. Admittedly, community engagement requires time and flexibility, and these requirements can pose challenges for testing the effectiveness of interventions in rigorous studies, such as the HCS (Clinical and Translational Science Awards Consortium, 2011W). Similar future studies may benefit from a bi-phasic research approach and longer time to examine individual level outcomes. Phase I could provide sufficient time and support for community-driven activities and pilot work, as well as coordinated planning at the service delivery organization level—as depicted in the adapted RE-AIM/PRISM framework. (Knudsen et al., 2020). Those demonstrating readiness for full EBP implementation could progress to the second trial phase with additional time to address barriers and measure distal outcomes including the reach and impact of EBPs.

Beyond the time and flexibility requirements, community-led adoption of EBPs introduces variance in implementation, which is reflected in the low dose of BHS interventions of interest in this analysis. While randomized control designs are the gold standard, large multi-site trials with community-driven components may benefit from supplemental studies leveraging other rigorous designs, such as interrupted time series, which may be better suited for instances in which a small number of communities implement certain EBPs. Indeed, lessons and administrative datasets from HCS, which will be shared in a data repository at the conclusion of the study, will help researchers and community partners better plan for future studies of interventions designed to increase BHS and reduce overdose-related fatalities among individuals with OUD.

Supplementary Material

1

HIGHLIGHTS.

  1. Evidence-based practices to reduce opioid overdose deaths implemented in 67 communities.

  2. Outpatient behavioral health services are key element of opioid addiction treatment.

  3. Similar receipt of behavioral services in control and intervention communities.

  4. More research needed to test the impact of comprehensive, community-driven interventions.

Acknowledgements

The authors wish to acknowledge the participation of the HEALing Communities Study communities, coalitions, partner organizations and agencies, and Community Advisory Boards, as well as the state government officials who partnered with us on this study.

Role of Funding Source

This work was supported by the following National Institutes of Health grants: RTI International: UM1DA049394, University of Kentucky: UM1DA049406, Boston Medical Center: UM1DA049412, Columbia University: UM1DA049415, Ohio State University: UM1DA049417. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health, the Substance Abuse and Mental Health Services Administration or the NIH HEAL Initiative. Dr. Redonna Chandler was substantially involved in UM1DA049394, UM1DA049406, UM1DA049412, UM1DA049415, UM1DA049417 consistent with her role as a National Institutes of Health Science Officer.

Footnotes

Declarations of Interest: In last three years, MRL has served as a scientific consultant for treatments in development for substance use disorders to Journey Colab, Titan, Braeburn, and Berkshire Biomedical. RW has consulted to Alkermes.

DECLARATION OF INTEREST STATEMENT

Michelle Lofwall has served as a scientific consultant for treatments in development for substance use disorders to Journey Colab, Titan, Braeburn, and Berkshire Biomedical in the last three years.

Roger Weiss has consulted to Alkermes.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

  1. Ahmad F, Cisewski J, Rossen L, Sutton P, 2023. National Center for Health Statistics: Provisional drug overdose death counts. https://www.cdc.gov/nchs/nvss/vsrr/drug-overdose-data.htm
  2. Benjamini Y, Drai D, Elmer G, Kafkafi N, Golani I, 2001. Controlling the false discovery rate in behavior genetics research. Behav Brain Res 125(1–2), 279–284. [DOI] [PubMed] [Google Scholar]
  3. Centers for Disease Control and Prevention, 2022. Linking people with opioid use disorder to medication treatment: A technical package of policy, programs, and practices. National Center for Injury Prevention adn Control, CDC, Department of Health and Human Services,. [Google Scholar]
  4. Chan B, Gean E, Arkhipova-Jenkins I, Gilbert J, Hilgart J, Fiordalisi C, Hubbard K, Brandt I, Stoeger E, Paynter R, Korthuis PT, Guise JM, 2021. Retention Strategies for Medications for Opioid Use Disorder in Adults: A Rapid Evidence Review. J Addict Med 15(1), 74–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chandler R, Nunes EV, Tan S, Freeman PR, Walley AY, Lofwall M, Oga E, Glasgow L, Brown JL, Fanucchi L, Beers D, Hunt T, Bowers-Sword R, Roeber C, Baker T, Winhusen TJ, 2023. Community selected strategies to reduce opioid-related overdose deaths in the HEALing (Helping to End Addiction Long-term SM) communities study. Drug and Alcohol Dependence 245, 109804. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chandler RK, Villani J, Clarke T, McCance-Katz EF, Volkow ND, 2020. Addressing opioid overdose deaths: The vision for the HEALing communities study. Drug and Alcohol Dependence 217, 108329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Clinical and Translational Science Awards Consortium, Community Engagement Key Function Committee., Task Force on the Principles of Community Engagement. “Principles of community engagement; ”, 2011. [Google Scholar]
  8. Garg RK, Fulton-Kehoe D, Franklin GM, 2017. Patterns of Opioid Use and Risk of Opioid Overdose Death Among Medicaid Patients. Med Care 55(7), 661–668. [DOI] [PubMed] [Google Scholar]
  9. Gilbert M, Hartnett T, Hoagland GW, Serafini MW, Strong K, 2023. Filling the Gaps in the Behavioral Health Workforce. Washington, D.C. [Google Scholar]
  10. Guyer J, 2021. Speaking the same language: a toolkit for strengthening patient-centered addiction care in the United States https://www.asam.org/docs/default-source/quality-science/final---asam-toolkit-speaking-same-language.pdf?sfvrsn=728c5fc2_2. (Accessed April 24 2023).
  11. Han B, Einstein EB, Jones CM, Cotto J, Compton WM, Volkow ND, 2022. Racial and Ethnic Disparities in Drug Overdose Deaths in the US During the COVID-19 Pandemic. JAMA Netw Open 5(9), e2232314. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hoots BE, 2021. Opioid Overdose Surveillance : Improving Data to Inform Action. Public Health Rep 136(1_suppl), 5S–8S. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Jhanjee S, 2014. Evidence based psychosocial interventions in substance use. Indian J Psychol Med 36(2), 112–118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Jones CM, Han B, Baldwin GT, Einstein EB, Compton WM, 2023. Use of Medication for Opioid Use Disorder Among Adults With Past-Year Opioid Use Disorder in the US, 2021. JAMA Netw Open 6(8), e2327488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Jones CM, McCance-Katz EF, 2019. Co-occurring substance use and mental disorders among adults with opioid use disorder. Drug and Alcohol Dependence 197, 78–82. [DOI] [PubMed] [Google Scholar]
  16. Kariisa M, Davis NL, Kumar S, Seth P, Mattson CL, Chowdhury F, Jones CM, 2022. Vital Signs: Drug Overdose Deaths, by Selected Sociodemographic and Social Determinants of Health Characteristics - 25 States and the District of Columbia, 2019–2020. MMWR Morb Mortal Wkly Rep 71(29), 940–947. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Knudsen HK, Drainoni M-L, Gilbert L, Huerta TR, Oser CB, Aldrich AM, Campbell ANC, Crable EL, Garner BR, Glasgow LM, Goddard-Eckrich D, Marks KR, McAlearney AS, Oga EA, Scalise AL, Walker DM, 2020. Model and approach for assessing implementation context and fidelity in the HEALing Communities Study. Drug and Alcohol Dependence 217, 108330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Larochelle MR, Slavova S, Root ED, Feaster DJ, Ward PJ, Selk SC, Knott C, Villani J, Samet JH, 2021. Disparities in Opioid Overdose Death Trends by Race/Ethnicity, 2018–2019, From the HEALing Communities Study. Am J Public Health 111(10), 1851–1854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Lefebvre RC, Chandler RK, Helme DW, Kerner R, Mann S, Stein MD, Reynolds J, Slater MD, Anakaraonye AR, Beard D, Burrus O, Frkovich J, Hedrick H, Lewis N, Rodgers E, 2020. Health communication campaigns to drive demand for evidence-based practices and reduce stigma in the HEALing communities study. Drug and Alcohol Dependence 217, 108338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. MACPAC, 2017. Medicaid and the Opioid Epidemic, Report to Congress on Medicaid and CHIP. Medicaid and CHIP Payment and Access Commission (U.S.), Washington, DC, pp. 59–97. [Google Scholar]
  21. Mancl LA, DeRouen TA, 2001. A covariance estimator for GEE with improved small-sample properties. Biometrics 57(1), 126–134. [DOI] [PubMed] [Google Scholar]
  22. Marino LA, Campbell AN, Nunes EV, Sederer LI, Dixon LB, 2019. Factors Influencing Buprenorphine Prescribing among Physicians in New York State. J Addict 2019, 7832752. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Mauro PM, Gutkind S, Annunziato EM, Samples H, 2022. Use of Medication for Opioid Use Disorder Among US Adolescents and Adults With Need for Opioid Treatment, 2019. JAMA Netw Open 5(3), e223821. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. McCollum CG, Eaton E, Creger T, Lee A, Gagnon K, Li L, 2023. Physicians’ experiences with buprenorphine: A qualitative study of motivations for becoming X waivered and barriers to and facilitators of prescribing the medication for opioid use disorder. Drug Alcohol Depend 244, 109777. [DOI] [PubMed] [Google Scholar]
  25. National Institute on Drug Abuse (NIDA), 2018. Principles of Drug Addiction Treatment: A Research-Based Guide (Third Edition). National Institutes of Health; U.S. Department of Health and Human Services. [Google Scholar]
  26. National Institute on Drug Abuse (NIDA), 2021. Medications to Treat Opioid Use Disorder Research Report. https://nida.nih.gov/publications/research-reports/medications-to-treat-opioid-addiction/overview (Accessed April 24, 2023 2023).
  27. Novak P, Feder KA, Ali MM, Chen J, 2019. Behavioral health treatment utilization among individuals with co-occurring opioid use disorder and mental illness: Evidence from a national survey. J Subst Abuse Treat 98, 47–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Ober AJ, Hunter SB, McCullough CM, Leamon I, McCreary M, Beas I, Montero A, Tarn DM, Bromley E, Hurley B, Sheehe J, Martinez J, Watkins KE, 2022. Opioid Use Disorder Among Clients of Community Mental Health Clinics: Prevalence, Characteristics, and Treatment Willingness. Psychiatr Serv 73(3), 271–279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Substance Abuse and Mental Health Services Administration: Opioid-Overdose Reduction Continuum of Care Approach (ORCCA) Practice Guide. Rockville, MD: National Mental Health and Substance Use Policy Laboratory. Substance Abuse and Mental Health Services Administration, 2023. [Google Scholar]
  30. Saunders H, 2023. A Look at Substance Use Disorders (SUD) Among Medicaid Enrollees https://www.kff.org/medicaid/issue-brief/a-look-at-substance-use-disorders-sud-among-medicaid-enrollees/ (Accessed April 24, 2023).
  31. Schulz KF, Altman DG, Moher D, for the CONSORT Group. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials [DOI] [PubMed] [Google Scholar]
  32. Slavova S, LaRochelle MR, Root ED, Feaster DJ, Villani J, Knott CE, Talbert J, Mack A, Crane D, Bernson D, Booth A, Walsh SL, 2020. Operationalizing and selecting outcome measures for the HEALing Communities Study. Drug and Alcohol Dependence 217, 108328. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Sprague Martinez L, Rapkin BD, Young A, Freisthler B, Glasgow L, Hunt T, Salsberry PJ, Oga EA, Bennet-Fallin A, Plouck TJ, Drainoni M-L, Freeman PR, Surratt H, Gulley J, Hamilton GA, Bowman P, Roeber CA, El-Bassel N, Battaglia T, 2020. Community engagement to implement evidence-based practices in the HEALing communities study. Drug and Alcohol Dependence 217, 108326. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Substance Abuse and Mental Health Services Administration, 2022. Key substance use and mental health indicators in the United States: Results from the 2021 National Survey on Drug Use and Health. [Google Scholar]
  35. Substance Abuse and Mental Health Services Administration, 2023. Medications for substance use disorders. https://www.samhsa.gov/medications-substance-use-disorders. (Accessed April 24 2023).
  36. Timko C, Schultz NR, Cucciare MA, Vittorio L, Garrison-Diehn C, 2016. Retention in medication-assisted treatment for opiate dependence: A systematic review. J Addict Dis 35(1), 22–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Volkow ND, Chandler RK, Villani J, 2022. Need for comprehensive and timely data to address the opioid overdose epidemic without a blindfold. Addiction 117(8), 2132–2134. [DOI] [PubMed] [Google Scholar]
  38. Walsh SL, El-Bassel N, Jackson RD, Samet JH, Aggarwal M, Aldridge AP, Baker T, Barbosa C, Barocas JA, Battaglia TA, Beers D, Bernson D, Bowers-Sword R, Bridden C, Brown JL, Bush HM, Bush JL, Button A, Campbell ANC, Cerda M, Cheng DM, Chhatwal J, Clarke T, Conway KP, Crable EL, Czajkowski A, David JL, Drainoni M-L, Fanucchi LC, Feaster DJ, Fernandez S, Freedman D, Freisthler B, Gilbert L, Glasgow LM, Goddard-Eckrich D, Gutnick D, Harlow K, Helme DW, Huang T, Huerta TR, Hunt T, Hyder A, Kerner R, Keyes K, Knott CE, Knudsen HK, Konstan M, Larochelle MR, Craig Lefebvre R, Levin F, Lewis N, Linas BP, Lofwall MR, Lounsbury D, Lyons MS, Mann S, Marks KR, McAlearney A, McCollister KE, McCrimmon T, Miles J, Miller CC, Nash D, Nunes E, Oga EA, Oser CB, Plouck T, Rapkin B, Freeman PR, Rodriguez S, Root E, Rosen-Metsch L, Sabounchi N, Saitz R, Salsberry P, Savitsky C, Schackman BR, Seiber EE, Slater MD, Slavova S, Speer D, Martinez LS, Stambaugh LF, Staton M, Stein MD, Stevens-Watkins DJ, Surratt HL, Talbert JC, Thompson KL, Toussant K, Vandergrift NA, Villani J, Walker DM, Walley AY, Walters ST, Westgate PM, Winhusen T, Wu E, Young AM, Young G, Zarkin GA, Chandler RK, 2020. The HEALing (Helping to End Addiction Long-term SM) Communities Study: Protocol for a cluster randomized trial at the community level to reduce opioid overdose deaths through implementation of an integrated set of evidence-based practices. Drug and Alcohol Dependence 217, 108335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Weiss RD, Griffin ML, Marcovitz DE, Hilton BT, Fitzmaurice GM, McHugh RK, Carroll KM, 2019. Correlates of Opioid Abstinence in a 42-Month Posttreatment Naturalistic Follow-Up Study of Prescription Opioid Dependence. J Clin Psychiatry 80(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Westgate PM, Cheng DM, Feaster DJ, Fernandez S, Shoben AB, Vandergrift N, 2022. Marginal modeling in community randomized trials with rare events: Utilization of the negative binomial regression model. Clin Trials 19(2), 162–171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Winhusen T, Walley A, Fanucchi LC, Hunt T, Lyons M, Lofwall M, Brown JL, Freeman PR, Nunes E, Beers D, Saitz R, Stambaugh L, Oga EA, Herron N, Baker T, Cook CD, Roberts MF, Alford DP, Starrels JL, Chandler RK, 2020. The Opioid-overdose Reduction Continuum of Care Approach (ORCCA): Evidence-based practices in the HEALing Communities Study. Drug and Alcohol Dependence 217, 108325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Wu E, Villani J, Davis A, Fareed N, Harris DR, Huerta TR, LaRochelle MR, Miller CC, Oga EA, 2020. Community dashboards to support data-informed decision-making in the HEALing communities study. Drug and Alcohol Dependence 217, 108331. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

1
NIHMS1988390-supplement-1.pdf (80.5KB, pdf)

RESOURCES