Harm reduction in practice: baseline cohort description and early engagement trends in an incentive-based program for PWUD

Abstract

Background:

Contingency management (CM) has been used to reinforce abstinence in the treatment of substance use disorders (SUD). Novel applications of CM in people who use drugs (PWUD) have been used to facilitate other desirable behaviors.

Objective:

Describe and assess preliminary outcomes of a program intended to reduce risk for HIV and related infections in a population of PWUD through increased healthcare engagement.

Design:

Patients receiving care for SUD at a collocated clinic receive shaping CM-based incentives for risk assessment, testing, and clinic attendance.

Methods:

Baseline cohort characteristics are assessed, and engagement in clinical care during the early period of the program is described.

Results:

Participants are majority African American, female, and meet criteria for experiencing financial resource strain. During the first year of the program, no significant changes in clinic appointment attendance were observed.

Conclusion:

Patient-centered CM-based incentivization implemented in a low-barrier, harm-reduction setting may facilitate incremental health behavior changes to reduce infection-related comorbid risk. There is a need to expand approaches to those with high risk and barriers.

Keywords

contingency management harm reduction HIV risk behavior sexually transmitted infections substance use disorders

Background

Substance use disorders (SUDs) represent a significant and worsening public health crisis. Drug overdoses have accounted for over 100,000 deaths annually in the United States in recent years.¹ People who use drugs (PWUD) are at increased risk for infections with bloodborne pathogens like HIV, viral hepatitis, and sexually transmitted infections (STIs) due to practices like sharing needles or paraphernalia and high-risk sexual behaviors associated with substance use.^2,3 They also face challenges to health access and adherence to interventions to modify risk due to the high prevalence of adverse social determinants of health (SDoH).⁴

The opioid syndemic has severely impacted Baltimore City, with high rates of drug overdose, and with existing HIV prevalence three to eight times the national rates.⁵ People who inject drugs (PWID) in Baltimore have an 18% prevalence of HIV, which is over twice the 7% national average for PWID.^6,7 Additionally, in 2022, Maryland saw the most syphilis cases for the state since 1991, with a 65% rise between 2018 and 2022 (Maryland Department of Health STI Report, 2024).⁸

A widely used evidence-based treatment for SUD, especially with stimulant use, is contingency management (CM). CM is founded on the behavioral theory of operant conditioning, using prizes, vouchers, and money to reinforce desired behaviors.⁹ It has also been used to facilitate desirable health behaviors besides abstinence, like medication adherence,¹⁰ reducing high-risk sexual behaviors,¹¹ and clinic attendance.¹²

Our study aims to build on this use of CM by designing a program that reinforces healthcare engagement among patients with SUD to reduce the risk of HIV and related infections, aligning incentivization with patient-centered goals and measures of success to allow for implementation in a real-world clinical setting.¹³

The study is conducted in a collocated clinic that provides low-barrier access to medical care and supportive services for people with SUD based on a harm reduction, trauma-informed framework. Treatment retention at the clinic is low, with appointment cancellation/no-show rates around 45% in 2022. Attendees have high rates of acute care utilization and high assessed HIV risk, with three incident cases of HIV reported in 2022–2023. There was an acute need to implement contextually relevant interventions to reduce both high-risk behavior and enhance intervention adherence in a population with many barriers to care.

Objectives

The primary objective of this study is to implement and evaluate the effectiveness of a program pairing contingency management-based incentivization of healthcare engagement with interventions to reduce the risk of HIV and related infections (hepatitis C and syphilis), including periodic risk assessment, testing, behavior modification, and access to preventive services within a population of PWUD. The secondary objective is to increase healthcare engagement and retention in PWUD.

Methods

The study is conducted in a collocated clinic within an urban opioid treatment program in West Baltimore. The site offers low-barrier on-demand primary and infectious disease care, preventive and mental health care, as well as integrated case management, supportive and wraparound services by staff who are trained in trauma-informed care and within a harm reduction framework. Eligible patients are incentivized to attend appointments and utilize services. Services offered off-site are also paired with incentives if this aligns with the patient’s goals. Completion of activities is verified by a healthcare provider’s attestation on a standard paper form. The project will enroll 475 participants over 5 years. The study was reviewed by the University of Maryland, Baltimore (UMB) IRB and approved as non-human subject’s research. See the “Ethics Approval” section for details.¹⁴ The reporting of this study conforms to the Standards for Quality Improvement Reporting Excellence (SQUIRE 2.0) publication guidelines.¹⁵

Measures

Patients are recruited by the provider and self-referral. Participants complete baseline interviews and point-of-care (POC) HIV, hepatitis C, and syphilis testing. The interview consists of a biopsychosocial assessment provided by the Substance Abuse and Mental Health Services Administration (SAMHSA), the Generalized Anxiety Disorder 7-item scale (GAD-7), the Patient Health Questionaire (PHQ-9,) screening for interpersonal violence,¹⁶ HIV risk behavior scale (HRBS),¹⁷ and a physician-developed pre-exposure prophylaxis (PrEP) Questionnaire. Electronic health records are reviewed to assess overall risk and prior test results to guide screening procedures and testing frequency.

Patients are deemed high risk based on their self-reported risk behaviors on the HRBS as well as physician assessment. If a participant discloses sharing needles, and/or unprotected sex, they are deemed high risk and are seen at 3-month intervals for reassessment of risk, POC HIV/STI testing and education, and referral for PrEP.

The intended primary outcome of the program, a reduction in risk behavior, is evaluated by a change in risk assessment from baseline to the end of intervention. The secondary outcome of improved clinic attendance and utilization is assessed by comparing rates of completed clinic appointments at enrollment with rates observed at follow-up.

Participants are assessed at baseline, 6 months, and 12 months. Their eligibility window to complete a follow-up assessment begins 1 month prior to the exact date and remains open 2 months after the exact date.

The study is ongoing and we report baseline characteristics of the initial cohort and 6-month data of self-reported risk behaviors. Future publications will include data on the 12-month timepoint for this sample.

We intend to analyze full data using interrupted time series (ITS) analyses to assess for change in risk behavior and attendance pre- to post-intervention. We will also use ITS to assess outcomes like self-reported mental health symptoms, days of drug use, medication prescribed and taken, and other parameters. For the final year, we aim to conduct qualitative interviews with patients who demonstrated the most improvement in health risk and attendance to identify impactful methods of implementation.

Intervention

The intervention consists of payment for completing goal behaviors. Compensated tasks are determined individually for each participant, and may include comprehensive testing for HIV/STI, keeping clinical appointments, or completing follow-up risk assessments with providers. When participants receive compensation for their first task, their following incentive-eligible activities are discussed, and the next eligible appointment is determined. Payment for appointments is on a one-to-one basis. Patients may complete multiple incentivized activities aligned with a shared health goal within a short time frame and are incentivized for each activity as a method of shaping such that patients’ successive approximations toward the desired target behavior are reinforced.

For patients deemed at high risk for HIV acquisition, most of the incentive-paired activities may include testing for HIV/STI, treatment or prophylaxis for STI (as appropriate), PrEP visits (or HIV care visits for those who have HIV), treatment visits for hepatitis C, and wound care for those with cutaneous wounds (all available on-site). Payment is provided immediately post visit ($15 each). Each participant is eligible for five payments per year ($75/year) in accordance with the HHS-OIG Final Rule regarding the use of cash equivalents.

Patients who screen low risk for HIV/STI may have incentives assigned to tasks like establishing primary care or attending clinic appointments. If they desire to establish a goal of abstinence, once clinic engagement is improved, the study team will assign incentives to be linked to negative urine toxicology screens. A visual representation of care pathways for low and high-risk individuals is depicted in Figure 1.

Figure 1.

Care pathway for new and existing patients engaged in harm reduction and incentive-based programming.

A participant will not receive the payment until they meet the assigned goal. Further, they are not assigned their next goal until they meet the first. This creates variation in the intervals for which patients receive their payment.

A goal is redefined if an individual’s infection risk on rescreening changes from high to low or low to high at subsequent assessments, and goals may be reassigned or reoriented based on achievement of initial goals to reprioritize preventive measures or other emergent health issues.

Results

A subset of the anticipated final sample (N = 475) was enrolled in 2024. The current samples’ (N = 113) descriptive data, HIV, HCV, syphilis diagnosis, and specific SDoH are provided in Table 1. Participants are categorized as high versus low risk based on the above-described criteria for risk behaviors (i.e., injecting drugs and/or unprotected sex) and final physician interpretation of patients endorsed behaviors.

Table 1.

Baseline sample characteristics by risk group (N = 113).

Characteristic	Low risk (N = 72)	High-risk (N = 41)
Age, mean (SD)	53.6 (10.5)	44.6 (12.4)
Gender, female (%)	31 (43.05)	19 (46.34)
Race
African American	42 (58.3%)	16 (39%)
White	29 (40.3%)	25 (61%)
Other	1(1.4%)	0 (0)
HCV diagnosis^a (%)	28 (38.9)	17 (41.5)
HCV diagnosis in last 2 years (%)	3/28 (10.7)	2/17 (11.8)
Syphilis diagnosis^b (%)	2 (2.8)	3 (7.3)
Syphilis diagnosis in last 2 years (%)	0 (0)	2/3 (66.67)
HIV diagnosis (%)	10 (13.9)	7 (17.1)
HIV diagnosis in last 2 years (%)	2/10 (20)	3/7 (42.9)
Financial resource strain^c (%)	39/69 (56.5)	24/39 (61.5)
Housing instability^d (%)	19/70 (27.1)	15/41 (36.6)
Food insecurity^e (%)	33/70 (47.1)	22/40 (55)

Hepatitis C Virus (HCV) diagnosis defined by positive HCV RNA (viral load), indicating chronic infection.

Syphilis diagnosis defined by Rapid Plasma Reagin (RPR)-positive test result with corresponding clinical or provider documentation indicating need for treatment during the study period.

Single question screen based on Centers for Medicare & Medicaid (CMS) Accountable Health Communities (AHC) or custom Electronic Medical Record (EMR)-based SDoH assessments: How hard is it for you to pay for the very basics?

Four question screen adapted from PRAPARE tool developed by the NACHC.

Using the Hunger Vital Sign™ screening tool based on the U.S. Household Food Security Survey Module.

NACHC, National Association of Community Health Centers; PRAPARE, Protocol for Responding to and Assessing Patients’ Assets, Risks, and Experiences; SDoH, social determinants of health.

Table 2 reflects available paired participants’ risk behaviors at baseline and 6-month follow-up, keeping in mind the ongoing nature of the study and that the initial cohort of participants’ 6-month follow-up windows have not all closed at the time of publication.

Table 2.

Change in individual HIV risk behaviors between baseline and at 6 months for patients who have paired data.

HIV risk behaviors	Baseline (%)	6 months (%)
Injection drug use	11/85 (12.9%)	9/85 (10.6%)
Needle sharing	2/11 (18.2%)	2/9 (22.2%)
Condomless sex**	26/85 (30.6%)	17/85 (20.0%)*
Multiple partners	6/85 (7.1%)	6/85 (7.1%)
Mean HRBS score	3.3 (SD 4.9)	2.5 (SD 3.9)

Statistically significant at p value < 0.05. Other individual items are not significant. Note the p-value for individual items of HRBS adjusted for multiple comparisons.

Using McNemar’s test for paired proportions.

HRBS, HIV risk behavior scale.

Finally, clinic attendance and utilization rates are available for 2024 from the University of Maryland Medical System electronic health database. During the enrollment period, among individuals with at least one scheduled appointment in both Q1 and Q4 of 2024, completion rates were largely stable across the year. A Wilcoxon signed-rank test showed no significant change in appointment completion over time (V = 852.5, p = 0.66), with a small median increase in completion of 0.02 that was not statistically meaningful (95% CI (−0.08, 0.11)).

Discussion

The purpose of this project is to identify CM-based incentivization-paired interventions that may influence a reduction in risk of HIV and related infectious disease in PWUD in real-world clinical settings.

Newer efforts with CM have begun in preventative measures, treatment attendance, and similar behaviors to the ones described in this program. However, literature in this area of CM is limited and findings are mixed.^11,18,19

Strengths of our study include improved access through collocated care, immediacy of reinforcement, and a focus on patient-expressed health goals, which may improve salience of and promote greater behavior change.²⁰ We also prioritize POC testing to limit phlebotomy when possible. Lastly, the quasi-experimental approach allows all participants to receive the benefits of the intervention. However, the study has a number of limitations.

First, analyses of the effectiveness of the intervention are limited without a control group and by the multi-pronged nature of the intervention. We wish to maximize the number of patients who can receive the intervention due to the high rates of adverse SDoH in the study population, impacting access. We may not be able to reliably compare this group with non-enrolled patients in the clinic due to low baseline engagement, expected high attrition, and systematic baseline differences in clinic-engaged populations who have low risk. We will aim to serially assess risk in enrolled populations over the entire period of the project to assess the change in risk. We acknowledge the limitations of the HRBS, and while residual confounding cannot be eliminated, this offers us the most practical assessment of impact of intervention.

Another limitation is that outcomes may not be comparable between participants who receive different incentivized tasks or have varying baseline risks. Also, SAMHSA regulations impose annual disbursement limits for CM per patient (HHS-OIG Final Rule) and limited flexibility in payment structure. A total of $15 may be deemed too low to reinforce a task with a high perceived barrier for the patient, even though data about the effectiveness of the total cash value of CM are variable.²¹ Within limitations of needing to stay within the total permitted per patient CM incentive value limit allowed by the SAMHSA and implementation in the clinical context, we felt an escalating schedule was not feasible. Fixed payments have been used for adherence intervention.^22–24 We hope that more immediate disbursement of funds may address more pressing healthcare issues and help improve overall engagement. However, the variation in payment intervals will need to be adjusted during data analysis. Further, while abstinence may not be the goal for everyone, abstinence does lead to a greater quality of life, self-esteem, and recovery capital.²⁵

Discussion of baseline cohort

The sample reported in this paper has HIV prevalence representative of rates in PWID reported from Baltimore City (18%).⁷ Our sample is made up entirely of PWUD but does not distinguish between injection drug use versus other forms of drug use. Further, we only measure current drug use.

The proportion of patients with previous or current hepatitis C is high and comparable to estimates among those with ever injection drug use.²⁶ The rate of syphilis in our sample is significantly higher than the prevalence in Baltimore City (12% vs 0.54%), but rates of syphilis among people who use or inject drugs in Baltimore City are not available.

There was an observed change in patients engaging in condomless sex from baseline to 6-month follow-up, although no significant changes for injection drug use, needle sharing, or having multiple partners (Table 2). However, the sample size for these items is small and causal inference cannot be established due to the incomplete data on incentivized attendance completion.

Lastly, because the initial cohort was enrolled throughout 2024, analyses of change in appointment attendance during the first year should not be interpreted as reflecting the impact of the intervention. Rather, these results reflect the overall patterns of clinic attendance during the enrollment period. Data on appointment completion stratified by enrollment in the incentivization program will be presented in future publications.

Conclusion

PWUD may face many challenges with access or engagement. For such patients, abstinence may not be a realistic goal. We aim to facilitate incremental or continued changes in health behavior through engagement in care using CM-based incentivization. This is flexible, patient-centered, and encourages more autonomy and self-efficacy, which may affect overall health risk over time. It may also be more acceptable and relevant for vulnerable populations.²⁰ There is a need to pivot and adapt such interventions to meet people where they are.

Supplemental Material

sj-docx-1-tai-10.1177_20499361261420916 – Supplemental material for Harm reduction in practice: baseline cohort description and early engagement trends in an incentive-based program for PWUD

Supplemental material, sj-docx-1-tai-10.1177_20499361261420916 for Harm reduction in practice: baseline cohort description and early engagement trends in an incentive-based program for PWUD by Nadia Mattanah, Aditi Ringwala, Atri Surapaneni, Kellie Miller, Anne Sawyer and Shivakumar Narayanan in Therapeutic Advances in Infectious Disease

Footnotes

Acknowledgements

None.

Disclaimer

Initial participant incentives in this study were disbursed as cash, based on contemporaneous guidance provided during consultation with the assigned SAMHSA project officer. However, subsequent clarification from SAMHSA leadership indicated that direct cash payments are not permissible under current federal policy, as outlined in SAMHSA’s Contingency Management Advisory (PEP24-06-001). In alignment with this guidance, the disbursement method was revised to gift cards as of May 2025. All future incentive distributions fully comply with SAMHSA policy and promote program integrity and participant well-being.

Declarations

ORCID iDs

Nadia Mattanah

Atri Surapaneni

Shivakumar Narayanan

Supplemental material

Supplemental material for this article is available online.

References

Garnett

Miniño

AM.

Drug overdose deaths in the United States, 2003-2023. Data Brief, National Health Center Statistics, USA, December 2024.

Murali

Jayaraman

Substance use disorders and sexually transmitted infections: a public health perspective. BJPsych Adv 2018; 24(3): 161–166.

Flores Anato

Panagiotoglou

Greenwald

, et al. Chemsex and incidence of sexually transmitted infections among Canadian pre-exposure prophylaxis (PrEP) users in the l’Actuel PrEP Cohort (2013-2020). Sex Transm Infect 2022; 98(8): 549–556.

Heidari

, et al. Barriers and facilitators to primary care engagement for people who inject drugs: a systematic review. J Nurs Scholarsh 2023; 55(3): 605–622.

Maryland Department of Health. Baltimore City Annual HIV epidemiological profile. Baltimore (MD): Maryland Department of Health, 2024.

Handanagic

Finlayson

Burnett

, et al. HIV infection and HIV-associated behaviors among persons who inject drugs—23 Metropolitan statistical areas, United States. Report, Morbidity and Mortality Weekly Report, USA, 2018.

Irwin

Jozaghi

Weir

, et al. Mitigating the heroin crisis in Baltimore, MD, USA: a cost-benefit analysis of a hypothetical supervised injection facility. Harm Reduct J 2017; 14(1): 29.

Maryland Department of Health. 2022 Annual Maryland STI Report. Report, Maryland, USA, April 2024.

Prendergast

Podus

Finney

, et al. Contingency management for treatment of substance use disorders: a meta-analysis. Addiction 2006; 101(11): 1546–1560.

10.

Ribeiro

Pinto

DGA

Trevisol

, et al. Can contingency management solve the problem of adherence to antiretroviral therapy in drug-dependent individuals? Health Educ Behav 2023; 50(6): 738–747.

11.

Letourneau

McCart

Sheidow

, et al. First evaluation of a contingency management intervention addressing adolescent substance use and sexual risk behaviors: risk reduction therapy for adolescents. J Subst Abuse Treat 2017; 72: 56–65.

12.

Milward

Lynskey

Strang

Solving the problem of non-attendance in substance abuse services. Drug Alcohol Rev 2014; 33(6): 625–636.

13.

Dennis

Foss

Scott

CK.

An eight-year perspective on the relationship between the duration of abstinence and other aspects of recovery. Evaluation Rev 2007; 31(6): 585–612.

14.

U.S. Department of Health and Human Services. Protection of human subjects. 45 CFR §46.101. https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-A/part-46 (2018).

15.

Ogrinc

Davies

Goodman

, et al. SQUIRE 20 (Standards for QUality Improvement Reporting Excellence): revised publication guidelines from a detailed consensus process. BMJ Quality Saf 2016; 25: 986–992.

16.

Iverson

King

Gerber

, et al. Accuracy of an intimate partner violence screening tool for female VHA patients: a replication and extension. J Trauma Stress 2015; 28(1): 79–82.

17.

Ward

Darke

Hall

, et al. The HIV Risk-taking behaviour scale (HRBS) manual. Report, University of New South Wales, Australia, 1990.

18.

Sung

Viera

Esserman

, et al. Contingency Management and Pre-Exposure Prophylaxis Adherence Support Services (CoMPASS): a hybrid type 1 effectiveness-implementation study to promote HIV risk reduction among people who inject drugs. Contemp Clin Trials 2023; 125: 107037.

19.

Pfund

Ginley

Rash

, et al. Contingency management for treatment attendance: a meta-analysis. J Subst Abuse Treat 2022; 133: 108556.

20.

Kullgren

Williams

Resnicow

, et al. The promise of tailoring incentives for healthy behaviors. Int J Workplace Health Manag 2016; 9(1): 2–16.

21.

Kropp

Lewis

Winhusen , et al. The effectiveness of ultra-low magnitude reinforcers: findings from a “Real-World” application of contingency management. J Subst Abuse Treat 2017; 72: 111–116.

22.

Tuten

Svikis

Keyser-Marcus

, et al. Lessons learned from a randomized trial of fixed and escalating contingency management schedules in opioid-dependent pregnant women. AM J Drug Alcohol Abuse 2012; 38(4): 286–292.

23.

Herrmann

Matusiewicz

Stitzer

, et al. Contingency management interventions for HIV, tuberculosis, and hepatitis control among individuals with substance use disorders: a systematized review. J Subst Abuse Treat 2017; 72: 117–125.

24.

Kirby

Benishek

Tabit

MB.

Contingency management works, clients like it, and it is cost-effective. AM J Drug Alcohol Abuse 2016; 42(3): 250–253.

25.

Eddie

Bergman

Hoffman

, et al. Abstinence versus moderation recovery pathways following resolution of a substance use problem: prevalence, predictors, and relationship to psychosocial well-being in a US national sample. Alcohol Clin Exp Res 2022; 46(2): 312–325.

26.

Falade-Nwulia

Mehta

Lasola

, et al. Public health clinic-based hepatitis C testing and linkage to care in Baltimore. J Viral Hepat 2016; 23(5): 366–374.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB