Abstract
Background:
Drug overdose remains a major crisis in the United States. Expanding substance use disorder (SUD) treatment and recovery support services is critical for reducing overdose risk during disasters such as the Severe acute respiratory syndrome coronavirus 2 (SARS-CoV2) pandemic. We evaluated the outcomes of an innovative multicomponent service, inclusive of medications for SUD, and peer support, co-located in an outpatient infectious disease clinic in Baltimore City. Our goal was to examine whether a multicomponent SUD program can support patients in recovery during a pandemic.
Methods:
105 patients in the RESTORE service between 2019–2020 completed baseline, 3-month and 6-month surveys. Telemedicine and phone-based support groups were implemented in March 2020 following statewide restrictions on face-to-face services due to SARS-CoV2. Data from surveys and electronic medical records were integrated and analyzed using mixed-effects regression models.
Results:
At baseline, most (88%) patients reported using drugs/alcohol in the preceding 30 days; 48% of patients reported a history of drug overdose, as well anxiety (23%) and depression (28%) symptoms. Despite pandemic-related disruptions and procedural changes, retention in RESTORE was high (83% after 3 months, 76% after 6 months). Mixed-effects regression models indicated decreased anxiety, alcohol use, heroin use, and non-fatal overdose after 6 months of enrollment (all p<0.05).
Conclusions:
Multicomponent SUD services that are co-located within infectious disease specialty services could help patients to successfully manage their overdose risk and mental health even during future disasters. This model of care could be implemented in other specialty settings that see high rates of SUD.
Keywords: drug treatment, overdose, mental health, substance use disorder
Introduction
Drug overdose remains a persistent public health issue in the U.S.,1 and rates have worsened since the advent of Severe acute respiratory syndrome coronavirus 2 (SARS-CoV2). Over 100,000 overdose deaths were recorded in 2021.2 In March of 2020, shelter-in-place measures were implemented in Maryland to reduce the spread of SARS-CoV2. Such policies raised concerns regarding their effects on social isolation, mental health and overdose.3,4 Patients with substance use disorder (SUD) are especially vulnerable to social isolation,6 for which the negative impact on mental health has been documented during the pandemic.7 Patients’ limited access to comprehensive SUD treatment and recovery support services may increase psychological distress and overdose risk.8,9 The purpose of this study is to examine whether a multicomponent SUD service can support patients in recovery even during a pandemic. We examined multiple behavioral health outcomes, including anxiety and depression symptoms, substance use, and overdose over 6-months of follow up.
Methods
The RESTORE service (REcovery in Specialty care Through medication and OutREach), established in January 2019, is a new multicomponent SUD treatment service designed to simultaneously improve access to low threshold SUD, mental health and recovery support services to patients attending an established outpatient infectious disease clinic affiliated with an academic hospital in Baltimore, Maryland. SUD care is provided utilizing a team-based approach, with patient case discussions occurring at a weekly team meeting among the infectious disease clinician, addiction psychiatrist, a licensed therapist, and peer recovery specialists. RESTORE provides a variety of as-needed services, including low threshold access to medications (e.g., sublingual buprenorphine/naloxone, extended-release buprenorphine, oral and injectable naltrexone, and acamprosate); linkage to mental health evaluation and management; peer recovery coaching and support to facilitate treatment initiation, engagement, and retention; a prescription for take-home naloxone; and linkage to community-based SUD treatment where a higher level of care is appropriate. Additionally, peer specialists prompt patients to remember clinic appointments and teach them to utilize clinic resources such as case management for social service needs. Patients were primarily referred for treatment and support by providers and staff at the infectious disease (ID) clinic and from inpatient settings.
All RESTORE patients were invited to participate in a cohort study consisting of three survey visits (baseline, 3-, 6-month regardless of treatment status), a $20 gift card, and release of electronic medical records (EMR). 128 patients were approached, and 23 patients declined to participate or were unable to be enrolled. The study was approved by the [Johns Hopkins Medicine] Institutional Review Board. Care occurred in-person until March 2020, after which all services transitioned to video/phone-based telemedicine including weekly peer-support calls, and home delivery of medications from the ID clinic-based pharmacy. Under the Health and Human Services Public Health Emergency Declaration, our practitioners were permitted to initiate and prescribe buprenorphine-naloxone via telemedicine to patients.5
The survey included questions from the SAMHSA Center for Substance Abuse Treatment Government Performance and Results Act Client Outcome Tool.10,11 Mental health symptoms were measured using the Generalized Anxiety Disorder Questionnaire (GAD-7) (cutoff:10, range:0–21), and the Patient Health Questionnaire-2 (PHQ-2) (cutoff:3, range:0–6).12,13 Social connectedness (SC) to support groups and family/friends was measured using a 4-item index (range:0–4).11 Given the dominance of fentanyl in the local epidemic, we added questions on past 30 day illicit fentanyl use. The AUDIT scale was used to detect hazardous drinking (cutoff:8, range:0–40).14 We asked about past 3 month overdose, and carrying naloxone. The presence of SUD (diagnosed by the treating clinician), medication history, and drug toxicology results were abstracted from EMR. Fatal overdoses were directly ascertained through the Maryland Office of the Chief Medical Examiner’s office.
In addition to describing the cohort, mixed-effects regression models accounting for enrollment period (before/after closures on March 30, 2020) and age/gender/race, were used to assess temporal changes in each health outcome. Linear models were applied for continuous outcomes (anxiety/depression/social connectedness/hazardous drinking) and logistic for dichotomous outcomes (fentanyl/heroin/overdose). We also explored the association between non-fatal overdose and drugs used using post-hoc Fisher’s exact tests (small cell sizes precluded regression). Complete case analysis was used for data missingness. All data were analyzed using Stata/MP 16.1 (StataCorp, College Station, TX).
Results
Among 105 enrolled patients, treatment retention remained high after 3- and 6-months (83% and 76%). Participants lost to follow-up did not differ in age, race, gender, or any of the outcomes at baseline from those who were retained (all p>0.05). Most patients were male (61%), and Black (79%). A quarter exhibited symptoms of anxiety (23%) and/or depression (28%). Most (86%) were connected to at least one organization or person who supported their recovery (average SC score=1.4, standard deviation=0.9, range: 0–4). Most patients reported substance use (88%) in the preceding 30 days, including heroin (49%), fentanyl (36%), cocaine/crack (55%) or alcohol (59%). The majority carried naloxone (82%). Most had been diagnosed with OUD (80%); however, alcohol use disorder (AUD) was also common (42%). Half (48%) had ever experienced an overdose. Over half (57%) received buprenorphine-naloxone, 11% were prescribed other SUD medications, and 33% did not receive medications.
After 3 months, 66% (n=55) continued to use drugs and reported a variety of reasons, including wanting to feel good/for pleasure/enjoyment/boredom (49%), mental health/recent trauma/stressor/grief (31%), chronic pain (26%), drug withdrawal (15%), social/environmental pressure (9%), and to avoid medication side effects (6%).
No fatal overdose deaths occurred during the study. Despite significant reductions in social connectedness after 6 months of recovery support services engagement, we observed reductions in anxiety symptoms (b=−1.73, p=0.002) and hazardous drinking (b=−2.20, p<0.001), as well as decreases in the odds of heroin use (b=0.04, p=0.001) and non-fatal overdose (b=0.04, p=0.02). To explore the overdose finding further, we compared the substance use profiles of those who did and did not report overdosing during the study through post-hoc analysis. Rates of overdose were higher among patients who used fentanyl (74% vs. 27%, p<0.001) and heroin (79% vs. 36%, p<0.001) but not cocaine/crack use (63% vs. 45%, p=0.15), non-medical prescription opioid use (3% vs. 9%, p=0.49), or alcohol use (53% vs. 54%, p=1.00).
Discussion
RESTORE, a multicomponent SUD treatment and recovery support service within an outpatient infectious disease clinic successfully supported patients through mental distress and overdose risk amidst pandemic-related disruptions to in-person services. Regulatory changes provided a means to mitigate treatment disruption and extended telemedicine access to persons previously not engaged in SUD treatment. Weekly check-in calls from peer recovery specialists and phone-based support groups benefited many patients during this period. Our preliminary findings indicate that comprehensively addressing comorbid substance use, mental health, and infectious disease conditions, could help to buffer the deleterious impacts of large-scale disruptions among patients in recovery.
Nationally, social isolation, psychological distress, and overdose rates rose during SARS-CoV2. However, we did not observe worsening of mental distress, substance use, or overdose among RESTORE patients, despite continued illicit opioid use. Team-based services for SUD remain uncommon in specialty outpatient settings throughout the US. Future research should evaluate the potential benefits of the RESTORE model to understand the value of scaling up such services. Overdose rates were highest among patients using fentanyl, which highlights the need for additional interventions for this group. In conclusion, these findings could be useful in preparing for periods where in-person care is not possible, during shelter-in-place due to SARS-CoV2 and other future disasters.
Table 1.
Baseline Characteristics of the RESTORE Cohort (N=105), Baltimore City, Maryland
| n (%) | |
|---|---|
|
| |
| Age, mean (sd) | 52 (10) |
| Gender | |
| Male | 64 (61.0) |
| Female | 40 (38.1) |
| Other | 1 (1.0) |
| Race/ethnicity | |
| Black | 83 (79.1) |
| White | 14 (13.3) |
| Hispanic/Latinx | 2 (1.9) |
| Multiracial | 6 (5.7) |
| Social connectedness (SC) score: mean, sd | 1.4 (0.9) |
| GAD-7 ≥ 10 | 24 (23.1) |
| PHQ-2 ≥ 3 | 29 (27.9) |
| AUDIT ≥ 8 | 34 (32.4) |
| Self-reported drug and alcohol use, past 30 days | |
| Heroin use | 51 (49.0) |
| Fentanyl use | 37 (35.6) |
| Prescription opioid use | 12 (11.4) |
| Cocaine/crack use | 58 (55.2) |
| Alcohol use | 62 (59.1) |
| If used drugs in past 30 days: Solitary drug use, past 30 days (n=87) | 45 (51.7) |
| Currently carry naloxone | 86 (81.9) |
| Current SUD diagnosis | |
| OUD (n=98) | 78 (79.6) |
| Diagnosed AUD (n=94) | 39 (41.5) |
| Prescribed SUD medications at baseline (select all that apply) | |
| Buprenorphine/naloxone (Suboxone, Zubsolv, Subutex, Sublocade) | 60 (57.1) |
| Methadone (Methadose, Dolophine) | 5 (4.8) |
| Naltrexone (Vivitrol, Revia) | 6 (5.7) |
| Disulfiram (Antabuse) | 1 (1.0) |
| Acamprosate (Campral) | 0 (0.0) |
| No medications prescribed | 35 (33.3) |
| Overdose | |
| Yes, ever | 50 (47.6) |
| Yes, past 3 months | 14 (13.3) |
AUD = Alcohol Use Disorder
AUDIT-c = Alcohol Use Disorder Identification Test-Concise
GAD = Generalized Anxiety Disorder
OUD = Opioid Use DisorderPHQ = Patient Health Questionnaire
SC = Social Connectedness
SUD = Substance Use Disorder
Table 2:
Temporal Changes in Mental Health, Social Connectedness, Substance Use and Overdose: RESTORE Cohort (N=104), Baltimore City, Maryland
| Linear models | b | 95% CI | p |
|
| |||
| Model A: Anxiety symptoms (GAD-7) 1 | |||
| Enrolled after March 30, 2020 | −2.00 | −3.33, −0.67 | 0.003 |
| Change in outcome (Ref: baseline) | |||
| 3-month | −1.02 | −2.06, 0.01 | 0.05 |
| 6-month | −1.73 | −2.83, −0.64 | 0.002 |
|
| |||
| Model B: Depression symptoms (PHQ-2) 1 | |||
| Enrolled after March 30, 2020 | −0.25 | −0.66, 0.15 | 0.22 |
| Change in outcome (Ref: baseline) | |||
| 3-month | −0.23 | −0.57, 0.11 | 0.19 |
| 6-month | −0.25 | −0.61, 0.11 | 0.18 |
|
| |||
| Model C: Social connectedness 1 | |||
| Enrolled after March 30, 2020 | −0.44 | −0.71, −0.17 | 0.001 |
| Change in outcome (Ref: baseline) | |||
| 3-month | −0.02 | −0.22, 0.18 | 0.85 |
| 6-month | −0.30 | −0.52, −0.09 | 0.005 |
|
| |||
| Model D: Hazardous drinking (AUDIT) 1 | |||
| Enrolled after March 30, 2020 | 1.10 | −0.78, 2.99 | 0.25 |
| Change in outcome (Ref: baseline) | |||
| 3-month | −1.04 | −2.13, 0.03 | 0.06 |
| 6-month | −2.20 | −3.38, −1.02 | <0.001 |
|
| |||
| Logistic models | aOR | 95% CI | p |
|
| |||
| Model E: Fentanyl use, past 3 months 1 | |||
| Enrolled after March 30, 2020 | 0.41 | 0.13, 1.29 | 0.13 |
| Change in outcome (Ref: baseline) | |||
| 3-month | 0.44 | 0.18, 1.08 | 0.07 |
| 6-month | 0.49 | 0.19, 1.25 | 0.14 |
|
| |||
| Model F: Heroin use, past 3 months 1 | |||
| Enrolled after March 30, 2020 | 0.11 | 0.01, 0.83 | 0.03 |
| Change in outcome (Ref: baseline) | |||
| 3-month | 0.09 | 0.02, 0.44 | 0.003 |
| 6-month | 0.04 | 0.007, 0.27 | 0.001 |
|
| |||
| Model G: Non-fatal overdose, past 3 months 1 | |||
| Enrolled after March 30, 2020 | 0.49 | 0.09, 2.72 | 0.42 |
| Change in outcome (Ref: baseline) | |||
| 3-month | 0.21 | 0.05, 0.93 | 0.04 |
| 6-month | 0.04 | 0.003, 0.54 | 0.02 |
Mixed-effects regression adjusted for age (years), gender (male vs other), race (Black vs other).
aOR = adjusted odds ratio; CI = confidence interval
Acknowledgements and funding
We gratefully acknowledge the study participants, the dedicated RESTORE team, as well as Yifan Zhang from the Johns Hopkins Institute for Clinical and Translational Research (ICTR), and C To for assistance with copy editing and manuscript submission. The RESTORE service was implemented with the support of a Substance Abuse and Mental Health Services Administration grant (H79TI084291). Research reported in this publication was supported by the National Institute on Drug Abuse (R01DA052297 K24DA035684) of the National Institutes of Health (NIH). Dr. Park was supported by the Rhode Island Center of Biomedical Research Excellence (COBRE) on Opioids and Overdose (5P20GM125507) and the Johns Hopkins Center for AIDS Research (1P30AI094189). The content is solely the responsibility of the authors and does not necessarily represent the official views of SAMHSA or the NIH.
Footnotes
Conflicts of interest: None.
References
- 1.Daly M, Gargano LM. Factors Associated with Substance Use Disorder Treatment Completion, Rhode Island, USA, 2018. Substance Use & Misuse. 2021;56(6):793–800. [DOI] [PubMed] [Google Scholar]
- 2.Ahmad FB, Cisewski JA, Rossen LM, Sutton P. Provisional drug overdose death counts. National Center for Health Statistics. 2022; accessed online on 10/12/22 (https://www.cdc.gov/nchs/nvss/vsrr/drug-overdose-data.htm). [Google Scholar]
- 3.McGinty EE, Presskreischer R, Han H, Barry CL. Psychological distress and loneliness reported by US adults in 2018 and April 2020. JAMA. 2020;324(1):93–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Cornwell EY, Waite LJ. Measuring social isolation among older adults using multiple indicators from the NSHAP study. J Gerontol B Psychol Sci Soc Sci. 2009;64 Suppl 1(Supplement 1):i38–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.COVID-19 Information Page. US Department of Justice and Drug Enforcement Administration Diversion Control Division. https://www.deadiversion.usdoj.gov/coronavirus.html. Accessed June 23, 2021. [Google Scholar]
- 6.Chou K-L, Liang K, Sareen J. The association between social isolation and DSM-IV mood, anxiety, and substance use disorders: wave 2 of the National Epidemiologic Survey on Alcohol and Related Conditions. J Clin Psychiatry. 2011;72(11):1468–76. [DOI] [PubMed] [Google Scholar]
- 7.Panchal N, Kamal R, Cox C, Garfield R. The Implications of COVID-19 for Mental Health and Substance Use. KFF. https://www.kff.org/coronavirus-covid-19/issue-brief/the-implications-of-covid-19-for-mental-health-and-substance-use/. Published February 10, 2021. [Google Scholar]
- 8.Volkow ND. Collision of the COVID-19 and addiction epidemics. Ann Intern Med. 2020;173(1):61–2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Dunlop A, Lokuge B, Masters D, Sequeira M, Saul P, Dunlop G, et al. Challenges in maintaining treatment services for people who use drugs during the COVID-19 pandemic. Harm Reduct J. 2020;17(1):26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Government Performance Results Act of 1993. Office of Management and Budget. https://obamawhitehouse.archives.gov/omb/mgmt-gpra/gplaw2m. Accessed December 11, 2021. [Google Scholar]
- 11.Substance Abuse and Mental Health Services Administration (SAMHSA), Center for Substance Abuse Treatment (CSAT). Government Performance and Results Act (GPRA) Client Outcome Measures for Discretionary Programs. https://spars.samhsa.gov/sites/default/files/CSATGPRATool1.pdf. Published March 2021.
- 12.Spitzer RL, Kroenke K, Williams JBW, Löwe B. A brief measure for assessing generalized anxiety disorder: the GAD-7: The GAD-7. Arch Intern Med. 2006;166(10):1092–7. [DOI] [PubMed] [Google Scholar]
- 13.Gilbody S, Richards D, Brealey S, Hewitt C. Screening for depression in medical settings with the Patient Health Questionnaire (PHQ): a diagnostic meta-analysis. J Gen Intern Med. 2007;22(11):1596–602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Babor TF, de la Fuente JR, Saunders J, & Grant M (1989). AUDIT: The Alcohol Use Disorders Identification Test: Guidelines for use in primary health care (WHO/MNH/DAT 89.4). Geneva, Switzerland: World Health Organization. [Google Scholar]