A Pilot Trial of Contingency Management for Stimulant Use in a Low-Barrier Syringe Service Program

Abstract

Objective:

To pilot test the efficacy of contingency management (CM) for stimulant use disorder (StimUD) among syringe service program (SSP) recipients.

Methods:

This randomized, parallel-groups, 12-week pilot trial was conducted in Burlington, Vermont from 02/24–07/24. Twenty-seven individuals with StimUD participating in SSP services were randomized to health education (HE) plus CM for session attendance (HE+Att-CM) (n=13) or HE plus CM for stimulant abstinence (HE+Abs-CM) (n=14). Stimulant use monitoring was examined twice-weekly using point-of-care urine drug tests (UDTs). HE+Att-CM participants earned incentives for UDT samples independent of results. HE+Abs-CM participants earned incentives for stimulant-negative UDT samples. The primary outcome was the percentage of participants stimulant-negative by week. Missing UDTs were treated as stimulant positive. Outcomes were compared between conditions using repeated measures mixed models for categorical data based on generalized estimating equations using a logistic link function.

Results:

The percentage of participants stimulant-negative was greater in the HE+Abs-CM condition averaging 22.6% (weekly range, 7.1%–28.6%) compared to 3.2% (weekly range, 0%–7.7%) in the HE+Att-CM condition (p=.014).

Conclusions:

Findings provide preliminary experimental evidence that CM reduces stimulant use among individuals participating in low-barrier services at a community-based SSP.

1. Introduction

The provision of evidence-based treatment for stimulant use disorders (StimUD) in community settings remains a notable research-to-practice gap in addiction health services and public health priority. To date, the most effective treatment for StimUD is contingency management (CM), a behavioral intervention wherein individuals earn financial incentives for objective evidence of target behavior achievement (e.g., recent stimulant abstinence).^1,2 As noted by the American Society of Addiction Medicine and Academy of Addiction Psychiatry in their clinical practice guidelines for StimUD, CM represents the current standard of care and should be a primary component of any StimUD treatment plan.³

Overdose deaths involving psychomotor stimulants (e.g., cocaine, methamphetamine) have drastically risen in the United States (US) in the current epidemic.^4,5 Between January 2021 and June 2024, stimulants were involved in 59.0% of all overdose deaths.⁶ More recently, whereas fatal overdoses involving fentanyl have begun to significantly decrease since their peak in 2023, stimulant-involved overdose deaths have continued to increase.⁷ Stimulant use patterns vary greatly by US geographic region, where in Vermont and most of the northeast cocaine is the most common stimulant.⁸ This differs from other US regions where methamphetamine is most prevalent.⁹ In 2024, the highest rate of cocaine-involved fatal overdoses was in Vermont (23.7/100,000 population) at almost four times the national average (8.1/100,000 population),¹⁰ with stimulant-involved overdose deaths for the first time surpassing those involving fentanyl.¹¹

Now, three decades after the seminal trials demonstrated the efficacy of CM for StimUD,¹² real-world statewide implementation initiatives are emerging across the US (e.g., California, Washington, Vermont, Michigan). In these initiatives, CM implementation is currently limited to providers in conventional substance use disorder (SUD) treatment settings (e.g., specialty SUD programs, opioid treatment programs). If we are to increase access to the current standard of care for StimUD treatment, these efforts will likely involve expanding CM beyond conventional SUD treatment settings into low-barrier community programs—where questions about efficacy and implementation remain largely unanswered empirically and in practice.

Syringe service programs (SSP) are one type of low-barrier community program that often provide an array of evidence-based SUD interventions for people who use drugs (PWUD).¹³ As nonjudgemental community settings, SSPs commonly provide services to many underserved populations at high risk for overdose and other harms.⁸ These populations include individuals interested in SUD treatment with preferences for services outside of conventional SUD treatment or healthcare settings due to stigma and other barriers.¹³ As a first step in assessing the feasibility of CM in SSPs, our research team examined interest in CM and reducing stimulant use among a sample of 139 individuals participating in SSP services, with approximately 83.0% of participants reporting interest in CM.¹⁴ Expanding upon these findings and reported interest, the purpose of the current study was to conduct a pilot randomized clinical trial to examine the efficacy of CM for StimUD among SSP service recipients.

2. Methods

2.1. Study Design and Population

Participants in this randomized, parallel-groups, 12-week pilot study were individuals with a StimUD who were currently receiving services at a SSP in Burlington, Vermont. Eligibility criteria included (a) ≥18 years of age, (b) self-reported past-30-day illicit stimulant use, (c) a moderate or severe StimUD (DSM-5th edition), and (d) no stimulant medication prescription. The study was conducted on-site at the SSP. Available SSP services included sterile syringes, narcan/overdose reversal kits, Human Immunodeficiency Virus (HIV) testing, counseling, medication for opioid use disorder (MOUD), drug test strips, wound care kits, and safer smoking supplies.

All prospective participants completed an intake assessment to assess eligibility that included questionnaires on sociodemographics (age, gender, race/ethnicity, housing status, employment, incarceration), substance use (primary drug, injection drug use), and substance use treatment (current, past-year). Written informed consent was provided by each participant; all individuals received $25 for completing the assessment. All study data were collected between February and July 2024 by a trained research assistant. This study was approved by the University of Vermont Institutional Review Board and pre-registered (NCT05787847).

Upon meeting eligibility, participants were randomly assigned to one of two treatment conditions: health education (HE) plus CM for session attendance (HE+Att-CM) or HE plus CM for stimulant abstinence (HE+Abs-CM). Random assignment was conducted using the National Cancer Institute’s software program. Treatment assignment was not masked after randomization. Forty individuals completed the study intake assessment, 4 were ineligible (no StimUD, n=2; prescription stimulant medication, n=2) and 9 did not attend their enrollment visit for condition randomization. Twenty-seven individuals met inclusionary criteria and enrolled in the study; 13 were randomized to the HE+Att-CM condition and 14 to the HE+Abs-CM condition.

Across both treatment conditions, monitoring of recent stimulant use was examined through twice-weekly urine drug tests (UDTs) across the 12-week intervention (24 total expected sessions) using a CLIA-waived 12-panel point-of-care test with protections to control for specimen tampering (e.g., temperature, adulterants), of which 3 panels examined stimulant use (cocaine, methamphetamine, amphetamine). Participants in both conditions received a brief 12-topic educational program addressing health, wellness, and lifestyle topics (e.g., stress relief, healthy relationships, hepatitis prevention, HIV prevention) adapted from a wellness manual used in previous studies.¹⁵ The research assistant discussed one education topic each week at one of the two weekly sessions.

2.1.1. Health Education Plus CM for Session Attendance.

In the HE+Att-CM condition, participants could earn $10.00 for attending the session and providing a urine specimen independent of the UDT results. This incentive was to generate UDT sample adherence. The maximum total incentives a participant could earn across the study was $240.00.

2.1.2. Health Education Plus CM for Stimulant Abstinence.

In the HE+Abs-CM condition, participants could earn financial incentives contingent on providing a stimulant-negative urine specimen. There was no separate incentive for urine specimen submission. The incentive amount for the first stimulant-negative UDT was $25.00. Every subsequent stimulant-negative UDT increased the incentive value by $2.50 to a maximum of $65.00. The maximum total incentives a participant could earn was $1,220.00 across the study. A reset to $25.00 occurred if a participant submitted a stimulant-positive specimen or had an unexcused absence, and a recovery of the pre-reset value occurred after a participant provided two consecutive stimulant-negative specimens.

2.2. Measures

The primary dependent outcome was biologically verified stimulant use, defined as the percentage of participants stimulant-negative by week. Secondary exploratory outcomes included the longest duration of continuous abstinence (whether participants achieved ≥ 4- or 8-week durations) and session attendance.

2.3. Statistical Analysis

Analyses of between-group differences on baseline demographics, substance use, and treatment characteristics were conducted using two-sample t-tests, Fisher’s exact tests, and χ2 tests. Outcome measures were examined using an intent-to-treat approach where all participants and expected UDT samples (n=24/person), excluding excused absences, were included in the analyses independent of early dropout. All non-excused missing UDTs were assumed to be stimulant-positive. The percentage of participants stimulant-negative and attending sessions were compared between conditions using repeated measures mixed models for categorical data based on generalized estimating equations using a logistic link function. Models included fixed factors of condition and time, with time (intervention week) a repeated factor, an independent working correlation structure, no inclusion of treatment-by-time interactions, and empirical/robust standard errors. Longest duration of continuous abstinence was compared between conditions using Wilcoxon rank sum tests. Dichotomous measures of longest duration were compared between conditions using Fisher’s Exact tests. Statistical significance level alpha was set a priori at 0.05. All statistical analyses were conducted using SAS ver. 9.4 statistical analysis software (SAS Institute, Inc., Cary, NC, USA).

3. Results

There were no significant differences between conditions on demographic, substance use, or treatment characteristics (Table 1). Overall, the most common primary drug was cocaine (37.0%), followed by methamphetamine (29.6%), and a combination of stimulants and opioids (22.2%). Fentanyl use and injection drug use were each reported by 70.4% of participants. Regarding treatment, the most common was MOUD, with 81.5% current and 88.9% within the past year.

Table 1.

Participant characteristics of syringe service program recipients with stimulant use disorder in Burlington, Vermont between 02/24–07/24, overall and by treatment condition.

Variable	All Participants (n = 27)		Health Education + CM for Session Attendance Condition (n = 13)		Health Education + CM for Stimulant Abstinence Condition (n = 14)		p- valuea
	mean	SD	mean	SD	mean	SD
Mean age (years)	41.8	7.7	40.1	6.1	43.4	8.8	0.27
	n	%	n	%	n	%
Gender							0.18
Female	16	59.3	6	46.2	10	71.4
Male	11	40.7	7	53.8	4	28.6
Race/Ethnicity							0.51
Non-Hispanic White	23	85.2	11	84.6	12	85.7
Non-Hispanic, > 1 race	3	11.1	2	15.4	1	7.1
Hispanic, > 1 race	1	3.7	0	0.0	1	7.1
Primary Drug							0.48
Cocaine	10	37.0	4	30.8	6	42.9
Methamphetamine	8	29.6	3	23.1	5	35.7
Cocaine & heroin/fentanyl	3	11.1	2	15.4	1	7.1
Methamphetamine & heroin/fentanyl	3	11.1	2	15.4	1	7.1
Heroin/fentanyl	2	7.4	2	15.4	0	0.0
Not reported	1	3.7	0	0.0	1	7.1
Injection Drug Use							0.47
Yes	19	70.4	10	76.9	9	64.3
Reported Fentanyl Use							0.47
Yes	19	70.4	10	76.9	9	64.3
Current Treatment
Medication for opioid use disorder	22	81.5	10	76.9	12	85.7	0.65
Outpatient	3	11.1	2	15.4	1	7.1	0.60
12-step/recovery group	3	11.1	2	15.4	1	7.1	0.60
None	3	11.1	2	15.4	1	7.1	0.60
Past Year Treatment
Medication for opioid use disorder	24	88.9	11	84.6	13	92.9	0.60
Outpatient	5	18.5	3	23.1	2	14.3	0.65
Inpatient	3	11.1	1	7.7	2	14.3	1.00
12-step/recovery group	3	11.1	2	15.4	1	7.1	0.60
None	3	11.1	2	15.4	1	7.1	0.60
Housing Status							0.40
Unhoused/Unstable	20	74.1	11	84.6	9	64.3
Permanent	6	22.2	2	15.4	4	28.6
Not reported	1	3.7	0	0.0	1	7.1
Employment Status							0.45
Unemployed	16	59.3	9	69.2	7	50.0
Disability	8	29.6	3	23.1	5	35.7
Part-time work	1	3.7	0	0.0	1	7.1
Full-time work	1	3.7	1	7.7	0	0.0
Other	1	3.7	0	0.0	1	7.1
Jail or Prison in Last 12 Months							0.17
Yes	6	22.2	1	7.7	5	35.7

Notes:

^ap-values based on two-sample t-tests, Fisher’s exact tests, and χ² tests. CM = contingency management.

Figure 1 (top panel) depicts the percentage of participants stimulant-negative by week and condition. There was a significant difference between conditions (p=.014), with the percentage of participants stimulant-negative in the HE+Abs-CM condition averaging 22.6% (weekly range, 7.1%–28.6%) compared to 3.2% (weekly range, 0%–7.7%) in the HE+Att-CM condition. Regarding the longest duration of continuous stimulant abstinence, the percentage of participants in the HE+Abs-CM condition and HE+Att-CM condition who achieved ≥4 weeks was 28.6% (n=4) versus 0.0% (n=0) (p=.098), and ≥8 weeks was 21.4% (n=3) versus 0.0% (n=0) (p=.222), respectively (Figure 1, middle panels). The percentage of participants attending sessions did not significantly differ between conditions (p=.147; Figure 1, bottom panel), with participants in the HE+Att-CM condition averaging 49.4% (weekly range, 38.5%–61.5%) compared to 27.4% (weekly range, 21.4%–57.1%) in the HE+Abs-CM condition.

Figure 1.

Outcome results by treatment condition among syringe service program recipients with stimulant use disorder in Burlington, Vermont between 02/24–07/24.

Notes: UDT = Urine Drug Test; CM = Contingency Management; HE = Health Education; Abs = Abstinence; Att = Attendance. Top Panel: Percentage of participants stimulant-negative by week and treatment condition. Middle Panel: Percentage of participants achieving continuous stimulant abstinence of ≥ 4 weeks and ≥ 8 weeks by treatment condition. Bottom Panel: Percentage of participants attending sessions by week and treatment condition.

4. Discussion

These findings provide preliminary evidence extending CM for StimUD to individuals participating in low-barrier SSP services—a marginalized, underserved population of PWUD¹⁶ at risk for drug- and other-related harms not actively engaged in StimUD treatment in other conventional SUD settings. With recent calls to expand CM access and consider how CM could be integrated into risk reduction frameworks,¹⁷ these data provide preliminary support for SSPs as a previously unexplored potential setting for StimUD treatment that warrants further exploration.¹⁸ Over 80.0% of participants were currently receiving MOUD, highlighting the notable public health efforts to expand access to the standard of care for opioid use disorder. This includes efforts within SSPs, where a growing number of programs report on-site MOUD treatment. Similar efforts to address disparities in access to CM for StimUD are greatly needed, with recent epidemiological findings showing deaths involving stimulants without fentanyl still rising.⁷

Collectively, the current findings have several implications for future CM research and implementation in low-barrier settings. Retention and early attrition, major challenges to effective StimUD treatment,¹² were observed in the present study, particularly among HE+Abs-CM participants where session attendance averaged 27.4%. Comparatively, session attendance among HE+Att-CM participants where incentives were contingent on attendance and independent of UDT results was higher (49.4%) although not significantly different. If a standalone therapeutic goal is to promote retention and frequent check-in opportunities, these findings suggest that future efforts to adapt CM for low-barrier settings could examine a combined intervention that includes contingencies for both stimulant abstinence and session attendance on a concurrent schedule where each behavior is separately reinforced.

By providing CM alongside other evidence-based services within SSPs, implications pertaining to the setting also merit comment. With CM for StimUD, individuals are participating in abstinence-focused treatment and asked to visit the SSP twice-weekly. With learning histories and repeated pairings of the SSP and drug use-related activities (e.g., picking up safer smoking supplies), for some individuals the SSP could function as a conditioned contextual cue associated with stimulant use, and elicit conditioned responses (e.g., craving, use expectancies) counter-therapeutic to stimulant abstinence. In these situations, options for CM sessions in an alternative setting could be provided.

The present study has several limitations to acknowledge. First, as a pilot study with heterogeneity in demographic characteristics (e.g., gender) observed between conditions due to small sample size, the current findings should be viewed as preliminary with future fully powered studies needed. Second, this study was conducted with a predominantly white population at one SSP in Vermont, limiting its generalizability to other populations and settings. Third, all study procedures were conducted by a trained research assistant; future studies should examine strategies for CM provision by staff within SSP organizations. These limitations notwithstanding, notable study strengths include the first randomized experimental evaluation of an evidence-based, abstinence-focused CM protocol among low-barrier SSP recipients, a clinically complex population with high rates of housing instability and polysubstance use, including fentanyl use.

5. Conclusion

This study provides preliminary experimental evidence that CM can reduce stimulant use among individuals participating in low-barrier services at a community-based SSP. Overall, we observed a between-groups difference wherein programmatic incentives contingent on a stimulant-negative specimen increased the percentage of stimulant-negative participants compared to incentives for session attendance. While current statewide initiatives are beginning to demonstrate how CM can be implemented in conventional SUD treatment settings, these preliminary findings suggests that CM provision in low-barrier settings like SSPs is a feasible approach to further expanding community access to this evidence-based, cost-effective treatment for StimUD.¹⁹

Highlights.

Syringe service programs (SSP) provide evidence-based interventions for substance use

We examined contingency management (CM) for stimulant use among SSP recipients

CM for stimulant abstinence increased stimulant-negative specimens

Findings provide preliminary support for SSPs as potential low-barrier settings for CM

Availability of Data and Materials

The datasets generated and/or analyzed during the current study are not publicly available due to institutionally approved protocols, but de-identified datasets can be made available from the corresponding author upon reasonable request.