Cocaine Use Disorder
Conditions
Keywords
Cocaine-Related Disorders, Contingency Management, Reward, Choice Behavior, Executive Function, Precision Medicine, Evidence-Based Practice, Prospective Thinking, Electroencephalography, Event-Related Potentials, Functional Connectivity
Brief summary
The proposed work will investigate changes in brain signaling and cognitive functioning that support recovery from addiction, as well as use of pretreatment neurocognitive functioning to inform substance use treatment planning. Substance use disorders are prevalent amongst Veterans. Cocaine addiction, in particular, has been shown to complicate treatment of other high priority behavioral health problems in the Veteran population (e.g., PTSD, opioid addiction). While there are currently no approved medications to support recovery from cocaine addiction, research indicates that Contingency Management (CM) - a behavioral intervention for cocaine users - can be effective. However, individual responses are variable and long-term benefits are limited. This CDA will test a new model of how CM works by examining brain-based predictors and indicators of treatment response. Results will have immediate implications for measurement-based implementation of existing CM variants within the VA, supporting access to the version of CM that is best aligned with each Veteran's needs.
Detailed description
Electrophysiological methods, including event-related potential and functional connectivity approaches, have potential to clarify mechanisms of substance use treatment response and characterize individual differences therein. Veterans are disproportionately affected by disorders of addiction, of which cocaine use disorder (CUD) is particularly problematic due to high relapse rates and the absence of approved pharmacotherapy options. Behavioral interventions for CUD have therefore become an important focus and Contingency Management (CM) has emerged as the best-supported approach. CM involves reinforcing cocaine abstinence (established through objective testing) with reliable, short-term reward, such as chances to win prizes (i.e., Prize-Based CM or PBCM). Given robust empirical support, nationwide dissemination of PBCM has been supported by a VHA initiative since 2011. However, PBCM response rates are variable and long-term benefits are limited - problems magnified by the cost of implementation with respect to staffing and prizes. Measurement-based approaches to PBCM implementation have promise to improve the effectiveness and efficiency of CM programming but have not yet been investigated within the VA or considered in relation to promising neuromarkers. Importantly, two versions of PBCM are already utilized at VA sites and may differentially benefit individuals with distinct neurocognitive profiles. Specifically, VA PBCM programs employ either abstract (voucher prize) or concrete (tangible prize) incentives, the latter of which may more effectively incentivize abstinence in Veterans with poor future-oriented thinking and planning ability. While selection between existing PBCM variants currently reflects practical considerations only, pretreatment neurocognitive functioning could meaningfully and realistically inform clinical decision-making in this regard. This project aims to advance measurement-based implementation of CM by testing a novel neurocognitive model with immediate implications for the use of abstract versus concrete PBCM incentives within the VA. Specifically, the future-minded decision-making (FMDM) model posits that CM scaffolds future-oriented goal representation and self-control to support abstinence during in the moment use-related decision-making. For individuals with greater FMDM impairment, concrete, readily-accessible incentives may be more effective than abstract voucher-based rewards (which require future-oriented thinking and planning to acquire value). To test this model, neurocognitive substrates of FMDM will be examined as predictors of differential treatment response in voucher (VoucherPBCM) versus tangible prize (TangiblePBCM) versions of PBCM. Treatment-related change in neural and cognitive-behavioral correlates of FMDM will also be evaluated in PBCM-adherent versus non-adherent subgroups. Veterans with CUD will be allocated to VoucherPBCM or TangiblePBCM conditions and followed for a 12-week treatment interval. Pre- and post-treatment electroencephalography (EEG) and cognitive-behavioral assessments will be used to measure FMDM-related constructs (working memory, self-control, future-oriented decision-making, future reward representation) and related neuromarkers. These measures will be investigated as predictors of differential treatment response in VoucherPBCM versus TangiblePBCM, as well as maintenance of gains during a post-treatment follow-up period. Change in FMDM-related neural and cognitive measures over the course of treatment will also be evaluated for evidence of neuroadaptation (e.g., changes in functional connectivity) and enhancement of FMDM function through PBCM. Taken together, results of the current research project will represent a first step toward precision implementation of CM within the VA.
Interventions
BEHAVIORALPrize-Based Contingency Management
Participants assigned to Prize-based Contingency Management (PBCM) conditions will receive PBCM as an adjunct to TAU. PBCM will involve twice weekly one-on-one sessions with a provider for 12-weeks. During each session, a urine specimen provided by the patient will be tested for cocaine using a point-of-care dip-test. Results of point-of-care testing will be shared with the patient and negative results will be reinforced with draws from a fish bowl containing 500 paper slips, 250 of which award small, large, or jumbo prizes (remaining slips deliver words of encouragement). Patients will be reinforced with a single prize draw for their first negative specimen; an additional prize draw will be added for each consecutive negative result (up to 8 prize draws per session). Abstinence-contingent prize draws will be reset to one upon either a positive test result or unexcused, missed appointment.
BEHAVIORALTreatment As Usual Outpatient Substance Use Treatment
All participants will receive treatment as usual outpatient substance use services during the 12-week treatment interval. TAU will specifically entail recommended participation in at least two outpatient group and/or individual psychotherapy encounters per week within the Center for Treatment of Addictive Disorders (CTAD) at VA Pittsburgh Healthcare System. Participants will additionally continue any previously prescribed pharmacotherapy for substance use and/or other mental health conditions, if applicable.
Sponsors
VA Office of Research and Development
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
NONE
Intervention model description
Participants will be randomly assigned to receive: (1) 12 weeks of TangiblePBCM (n = 70) or (2) 12 weeks of VoucherPBCM (n = 70). CM recipients will also be followed for 6 months post treatment. The proposed design enables evaluation of CM outcome predictors within 140 CM recipients - both with respect to initial treatment response and longer term (6 month) outcomes. All participants will receive a Baseline Assessment prior to the 12 Week Treatment interval, as well as a Follow-up Assessment at the conclusion of this period. Data from Baseline and Follow-up Assessments will enable longitudinal analysis of treatment-related change in EEG and cognitive-behavioral measures in treatment adherent versus treatment non-adherent subgroups.
Eligibility
Sex/Gender
ALL
Age
18 Years to 75 Years
Healthy volunteers
No
Inclusion criteria
* Military Veterans * DSM-5 Criteria for Cocaine Use Disorder (Mild, Moderate, or Severe) * Cocaine Use Within Past 60 Days * Stated Goal of Cocaine Abstinence or Reduced Cocaine Use * Normal or Corrected-to-Normal Vision * Average or Corrected Hearing
Exclusion criteria
* History of Severe Traumatic Brain Injury, Seizure Disorder, or other Neurological Illness * Severe or Unstable Medical or Psychiatric Condition * Pregnant or Lactating Women * Moderate-to-Severe Neurocognitive Impairment per Medical Record, SLUMS \< 21, or Mini MoCA \< 11 * In Ongoing Residential Treatment or Imminently Expected to Enter Residential Treatment During the Study Interval at Time of Screening
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Longest Duration of Cocaine Abstinence (LDCA) | 12-Week Treatment Interval | Longest period of objectively-verified abstinence from cocaine during treatment. It is noted that two TAU participants withdrew during the 12-week treatment interval. For these participants, the LDCA was computed for the pre-withdrawal period only. |
| % Cocaine-Negative Urine Specimens Per Participant | 12-Week Treatment Interval | Percentage of cocaine-negative urine specimens during the 12-week treatment interval out of the total number possible, computed on a per-participant basis. The denominator was adjusted for participants for whom a full course of treatment could not be delivered due to suspension of face-to-face research activities during the COVID-19 pandemic. Denominators were similarly adjusted for 2 TAU participants who withdrew from the study during the 12-week treatment interval. Descriptive statistics represent the mean and standard deviation of the per-participant percentage of cocaine-negative urine specimens, computed across participants within each arm. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Total Non-CM Treatment Encounters Per Participant | 12-Week Treatment Interval | Number of non-CM, recovery-oriented treatment encounters during the 12-week treatment interval, computed on a per-participant basis. Encounters were documented in the electronic health record (VA encounters) and/or self-reported (non-VA encounters, including interactions with mutual support communities). Because this measure reflects the absolute number of non-CM treatment encounters per participant, only individuals with complete data for the 12-week period were included. |
| % Self-Reported Cocaine-Abstinent Days During Treatment | 12-Week Treatment Interval | Percentage of self-reported cocaine-abstinent days during the 12-week treatment interval, computed on a per-participant basis. Only participants with complete self-report data for the full treatment interval were included. |
| % Contingency Management (CM) Sessions Attended Per Participant (CM Groups Only) | 12-Week Treatment Interval | Percentage of CM treatment sessions attended out of the total number possible, computed on a per-participant basis. The denominator was adjusted for participants for whom a full course of treatment could not be delivered due to suspension of face-to-face research activities during the COVID-19 pandemic. Descriptive statistics represent the mean and standard deviation of the per-participant percentage of CM sessions attended, computed across participants within each arm. |
| % Self-Reported Drug- and Alcohol-Abstinent Days During Treatment | 12-Week Treatment Interval | Percentage of self-reported drug and alcohol-abstinent days during the 12-week treatment interval, computed on a per-participant basis. Only participants with complete self-report data for the full treatment interval were included. |
| % Self-Reported Stimulant-Abstinent Days at Post-Treatment (CM Groups Only) | 6 Month Post-Treatment Interval | Percentage of self-reported stimulant-abstinent days during the 6 month post-treatment interval. For participants without complete data for the full 6-month post-treatment interval, the percentage of self-reported stimulant-abstinent days was computed based on the total number of days for which self-report data were available. |
| % Self-Reported Drug- and Alcohol-Abstinent Days at Post-Treatment (CM Groups Only) | 6 Month Post-Treatment Interval | Percentage of self-reported drug- and alcohol-abstinent days during the 6 month post-treatment interval. For participants without complete data for the full 6-month post-treatment interval, the percentage of self-reported stimulant-abstinent days was computed based on the total number of days for which self-report data were available. |
Other
| Measure | Time frame | Description |
|---|---|---|
| Difference in Control-related Theta Synchronization From Baseline to Post-Treatment | Baseline and Post-Treatment Follow-up | Theta synchronization between anterior cingulate cortex (Cz) and lateral prefrontal cortex (F3, F4, FC5, FC6) was measured to assess differences in functional connectivity between baseline and post-treatment (i.e., follow-up assessment conducted after the conclusion of the 12-week treatment interval) for high versus low conflict events. At each timepoint, phase locking value (PLV) was computed for 4 electrode pairs (Cz-F3, Cz-F4, Cz-FC5, Cz-FC6) across 3 stimulus conditions ranging from low to high conflict (Congruent, Intermediate-Incongruent, High-Incongruent) and for high-conflict error versus low-conflict correct responses in the Parametric Conflict Flankers task. A wavelet transformation isolated theta activity, and PLV was derived from normalized complex wavelet phases. PLV quantifies trial-to-trial phase synchrony (0-1), with 1 representing perfect synchrony. Descriptive data reported below represent the average PLV across the 4 electrode pairs. |
| Difference in Executive Working Memory From Baseline to Post-Treatment | Baseline and Post-Treatment Follow-up | Difference in Brown-Peterson working memory scores between baseline and post-treatment (i.e., follow-up assessment conducted after the conclusion of the 12-week treatment interval). We will specifically use a modified Brown-Peterson test (Auditory Consonant Trigrams) for which both age- and Veteran-specific norms exist. Summary scores for this measure (including 9-, 18-, and 36-second delay conditions) can range from 0-45, with higher scores indicating improved executive working memory performance. |
| Difference in Episodic Future Thinking Effect on Delay Discounting From Baseline to Post-Treatment | Baseline and Post-Treatment Follow-up | Delay discounting was quantified using the hyperbolic discounting parameter (k), estimated separately for the Episodic Future Thinking (EFT) condition-with personally meaningful event tags anchoring delayed rewards-and the Standard condition without such tags. Values of k were natural log-transformed (ln(k)), with higher ln(k) indicating steeper discounting of delayed rewards, reflecting greater devaluation of future outcomes and a stronger preference for immediate gratification. |
Countries
United States
Participant flow
Recruitment details
Recruitment took place through the VA Pittsburgh Healthcare System and other local sites serving Veterans with Substance Use Disorders. Methods included mailings, provider referrals, public advertisements, event info tables, EHR screening, and announcements during therapy groups. Recruitment began in November 2019 but was paused due to COVID-19 from March 17, 2020, to January 15, 2021. It resumed thereafter and continued until enrollment ended on August 31, 2023.
Pre-assignment details
Before randomization, participants completed a pre-screening interview. Those eligible were invited to enroll and proceed to full screening. Participants subsequently completed a baseline assessment visit that included biospecimen testing, interviews, questionnaires, cognitive-behavioral tasks, and an electroencephalography (EEG) session. Of the 122 unique Veterans prescreened, 63 met initial eligibility criteria and were enrolled in the study, with 45 subsequently proceeding to randomization.
Participants by arm
| Arm | Count |
|---|---|
| Tangible Prize-Based Contingency Management (TangiblePBCM) Participants in this condition will receive Prize-Based Contingency Management (PBCM) as an adjunct to treatment as usual (TAU). During twice-weekly sessions, urine samples will be tested for cocaine, and each negative result earns the participant draws from a bowl of 500 paper slips, 250 of which award small, large, or jumbo prizes (the remaining slips deliver words of encouragement). Each consecutive negative adds an additional draw (up to 8), while a positive or missed test resets the draw count.
For participants assigned to TangiblePBCM, prizes are awarded in the form of tangible items, which are chosen from a physical prize cabinet. Participants may redeem their draws for small, medium, large, or jumbo prizes, with medium prizes valued at approximately 5 small prizes. The cabinet is visible during sessions and restocked regularly, and the selection of prizes is guided by patient preference. All participants continue to receive their usual outpatient treatment and medications alongside PBCM. | 22 |
| Voucher Prize-Based Contingency Management (VoucherPBCM) Participants in this condition will receive Prize-Based Contingency Management (PBCM) as an adjunct to treatment as usual (TAU). During twice-weekly sessions, urine samples will be tested for cocaine, and each negative result earns the participant draws from a bowl of 500 paper slips, 250 of which award small, large, or jumbo prizes (the remaining slips deliver words of encouragement). Each consecutive negative adds an additional draw (up to 8), while a positive or missed test resets the draw count.
For participants assigned to VoucherPBCM, prize draws resulting in small, large, or jumbo wins are rewarded with VA Canteen vouchers of corresponding value. These vouchers can be redeemed at Veterans Canteen Service vendors, including the hospital cafeteria and Patriot Store. | 19 |
| Treatment As Usual (TAU) A treatment-as-usual (TAU) arm was originally included in the study design but was discontinued early in the trial due to evidence that its inclusion adversely affected participant accrual and retention among treatment-seeking individuals with cocaine use disorder. Participants randomized to TAU received standard care without the addition of prize-based contingency management (PBCM). Although they were still required to submit urine samples twice weekly for testing, they did not receive therapeutic incentives based on their urinalysis results. | 4 |
| Total | 45 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 | FG002 |
|---|---|---|---|---|
| 12-Week Treatment Interval | Withdrawal by Investigator | 0 | 0 | 1 |
| 12-Week Treatment Interval | Withdrawal by Subject | 0 | 0 | 1 |
| 6-Month Post-Tx Follow-Up Period | Lost to Follow-up | 12 | 10 | 0 |
| 6-Month Post-Tx Follow-Up Period | Withdrawal by Investigator | 1 | 0 | 0 |
| Post-Tx Evaluation Session | Declined or Unavailable | 1 | 2 | 0 |
| Post-Tx Evaluation Session | Lost to Follow-up | 4 | 4 | 0 |
Baseline characteristics
| Characteristic | Tangible Prize-Based Contingency Management (TangiblePBCM) | Voucher Prize-Based Contingency Management (VoucherPBCM) | Treatment As Usual (TAU) | Total |
|---|---|---|---|---|
| Age, Continuous | 61.9 age in years STANDARD_DEVIATION 7.2 | 57.5 age in years STANDARD_DEVIATION 7.9 | 61.5 age in years STANDARD_DEVIATION 4.4 | 60.0 age in years STANDARD_DEVIATION 7.5 |
| Ethnicity (NIH/OMB) Hispanic or Latino | 1 Participants | 1 Participants | 0 Participants | 2 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 21 Participants | 18 Participants | 4 Participants | 43 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Asian | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Black or African American | 18 Participants | 12 Participants | 4 Participants | 34 Participants |
| Race (NIH/OMB) More than one race | 1 Participants | 0 Participants | 0 Participants | 1 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) White | 3 Participants | 7 Participants | 0 Participants | 10 Participants |
| Region of Enrollment United States | 22 Participants | 19 Participants | 4 Participants | 45 Participants |
| % Self-Reported Pre-Treatment Cocaine-Abstinent Days | 84.0 % cocaine-abstinent days (past month) STANDARD_DEVIATION 18 | 91.7 % cocaine-abstinent days (past month) STANDARD_DEVIATION 10.9 | 83.7 % cocaine-abstinent days (past month) STANDARD_DEVIATION 12.7 | 87.2 % cocaine-abstinent days (past month) STANDARD_DEVIATION 15.1 |
| % Self-Reported Pre-Treatment Drug- and Alcohol-Abstinent Days | 70.8 % fully abstinent days (past month) STANDARD_DEVIATION 30.2 | 71.2 % fully abstinent days (past month) STANDARD_DEVIATION 36.7 | 60.7 % fully abstinent days (past month) STANDARD_DEVIATION 41.7 | 70.1 % fully abstinent days (past month) STANDARD_DEVIATION 33.4 |
| Sex: Female, Male Female | 2 Participants | 1 Participants | 0 Participants | 3 Participants |
| Sex: Female, Male Male | 20 Participants | 18 Participants | 4 Participants | 42 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk |
|---|---|---|---|
| deaths Total, all-cause mortality | 1 / 22 | 0 / 19 | 0 / 4 |
| other Total, other adverse events | 0 / 22 | 0 / 19 | 0 / 4 |
| serious Total, serious adverse events | 5 / 22 | 5 / 19 | 1 / 4 |
Outcome results
Primary
% Cocaine-Negative Urine Specimens Per Participant
Percentage of cocaine-negative urine specimens during the 12-week treatment interval out of the total number possible, computed on a per-participant basis. The denominator was adjusted for participants for whom a full course of treatment could not be delivered due to suspension of face-to-face research activities during the COVID-19 pandemic. Denominators were similarly adjusted for 2 TAU participants who withdrew from the study during the 12-week treatment interval. Descriptive statistics represent the mean and standard deviation of the per-participant percentage of cocaine-negative urine specimens, computed across participants within each arm.
Time frame: 12-Week Treatment Interval
Population: Results for the TAU arm are summarized using descriptive statistics but are omitted from statistical comparisons due to insufficient data (n = 4, with 2 out of 4 withdrawn from the study).
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Tangible Prize-Based Contingency Management (TangiblePBCM) | % Cocaine-Negative Urine Specimens Per Participant | 42.2 % negative specimens per participant | Standard Deviation 37.4 |
| Voucher Prize-Based Contingency Management (VoucherPBCM) | % Cocaine-Negative Urine Specimens Per Participant | 59.1 % negative specimens per participant | Standard Deviation 34.5 |
| Treatment As Usual (TAU) | % Cocaine-Negative Urine Specimens Per Participant | 51.0 % negative specimens per participant | Standard Deviation 18.8 |
p-value: 0.177Wilcoxon (Mann-Whitney)
Primary
Longest Duration of Cocaine Abstinence (LDCA)
Longest period of objectively-verified abstinence from cocaine during treatment. It is noted that two TAU participants withdrew during the 12-week treatment interval. For these participants, the LDCA was computed for the pre-withdrawal period only.
Time frame: 12-Week Treatment Interval
Population: Results for the TAU arm are summarized using descriptive statistics but are omitted from statistical comparisons due to insufficient data (n = 4, with 2 out of 4 withdrawn from the study).
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Longest Duration of Cocaine Abstinence (LDCA) | 21.36 LDCA in days | Standard Deviation 27.87 |
| Voucher Prize-Based Contingency Management (VoucherPBCM) | Longest Duration of Cocaine Abstinence (LDCA) | 34.95 LDCA in days | Standard Deviation 27.55 |
| Treatment As Usual (TAU) | Longest Duration of Cocaine Abstinence (LDCA) | 11.00 LDCA in days | Standard Deviation 19.34 |
p-value: 0.033Wilcoxon (Mann-Whitney)
Secondary
% Contingency Management (CM) Sessions Attended Per Participant (CM Groups Only)
Percentage of CM treatment sessions attended out of the total number possible, computed on a per-participant basis. The denominator was adjusted for participants for whom a full course of treatment could not be delivered due to suspension of face-to-face research activities during the COVID-19 pandemic. Descriptive statistics represent the mean and standard deviation of the per-participant percentage of CM sessions attended, computed across participants within each arm.
Time frame: 12-Week Treatment Interval
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Tangible Prize-Based Contingency Management (TangiblePBCM) | % Contingency Management (CM) Sessions Attended Per Participant (CM Groups Only) | 51.7 % CM sessions attended per participant | Standard Deviation 33.2 |
| Voucher Prize-Based Contingency Management (VoucherPBCM) | % Contingency Management (CM) Sessions Attended Per Participant (CM Groups Only) | 68.0 % CM sessions attended per participant | Standard Deviation 31.2 |
p-value: 0.126Wilcoxon (Mann-Whitney)
Secondary
% Self-Reported Cocaine-Abstinent Days During Treatment
Percentage of self-reported cocaine-abstinent days during the 12-week treatment interval, computed on a per-participant basis. Only participants with complete self-report data for the full treatment interval were included.
Time frame: 12-Week Treatment Interval
Population: Two participants were excluded from each arm (TangiblePBCM, VoucherPBCM, TAU) due to missing self-reported substance use data (collected via timeline follow-back procedure) during the 12-week treatment interval. All participants included had complete self-report data for the full interval. Results for the TAU arm are summarized using descriptive statistics but are omitted from statistical comparisons due to insufficient data (n = 4, with 2 out of 4 withdrawn from the study).
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Tangible Prize-Based Contingency Management (TangiblePBCM) | % Self-Reported Cocaine-Abstinent Days During Treatment | 86.0 % cocaine abstinent days | Standard Deviation 26.5 |
| Voucher Prize-Based Contingency Management (VoucherPBCM) | % Self-Reported Cocaine-Abstinent Days During Treatment | 86.8 % cocaine abstinent days | Standard Deviation 24.4 |
| Treatment As Usual (TAU) | % Self-Reported Cocaine-Abstinent Days During Treatment | 81.7 % cocaine abstinent days | Standard Deviation 25.9 |
p-value: 0.308Wilcoxon (Mann-Whitney)
Secondary
% Self-Reported Drug- and Alcohol-Abstinent Days at Post-Treatment (CM Groups Only)
Percentage of self-reported drug- and alcohol-abstinent days during the 6 month post-treatment interval. For participants without complete data for the full 6-month post-treatment interval, the percentage of self-reported stimulant-abstinent days was computed based on the total number of days for which self-report data were available.
Time frame: 6 Month Post-Treatment Interval
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Tangible Prize-Based Contingency Management (TangiblePBCM) | % Self-Reported Drug- and Alcohol-Abstinent Days at Post-Treatment (CM Groups Only) | 70.9 % drug and alcohol abstinent days | Standard Deviation 37.6 |
| Voucher Prize-Based Contingency Management (VoucherPBCM) | % Self-Reported Drug- and Alcohol-Abstinent Days at Post-Treatment (CM Groups Only) | 72.7 % drug and alcohol abstinent days | Standard Deviation 34.4 |
p-value: 0.873Wilcoxon (Mann-Whitney)
Secondary
% Self-Reported Drug- and Alcohol-Abstinent Days During Treatment
Percentage of self-reported drug and alcohol-abstinent days during the 12-week treatment interval, computed on a per-participant basis. Only participants with complete self-report data for the full treatment interval were included.
Time frame: 12-Week Treatment Interval
Population: Two participants were excluded from each arm (TangiblePBCM, VoucherPBCM, TAU) due to missing self-reported substance use data (collected via timeline follow-back procedure) during the 12-week treatment interval. All participants included had complete self-report data for the full interval. Results for the TAU arm are summarized using descriptive statistics but are omitted from statistical comparisons due to insufficient data (n = 4, with 2 out of 4 withdrawn from the study).
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Tangible Prize-Based Contingency Management (TangiblePBCM) | % Self-Reported Drug- and Alcohol-Abstinent Days During Treatment | 72.6 % drug and alcohol abstinent days | Standard Deviation 32.2 |
| Voucher Prize-Based Contingency Management (VoucherPBCM) | % Self-Reported Drug- and Alcohol-Abstinent Days During Treatment | 65.3 % drug and alcohol abstinent days | Standard Deviation 39.6 |
| Treatment As Usual (TAU) | % Self-Reported Drug- and Alcohol-Abstinent Days During Treatment | 56.4 % drug and alcohol abstinent days | Standard Deviation 9.9 |
p-value: 0.976Wilcoxon (Mann-Whitney)
Secondary
% Self-Reported Stimulant-Abstinent Days at Post-Treatment (CM Groups Only)
Percentage of self-reported stimulant-abstinent days during the 6 month post-treatment interval. For participants without complete data for the full 6-month post-treatment interval, the percentage of self-reported stimulant-abstinent days was computed based on the total number of days for which self-report data were available.
Time frame: 6 Month Post-Treatment Interval
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Tangible Prize-Based Contingency Management (TangiblePBCM) | % Self-Reported Stimulant-Abstinent Days at Post-Treatment (CM Groups Only) | 86.1 % stimulant-abstinent days | Standard Deviation 27.8 |
| Voucher Prize-Based Contingency Management (VoucherPBCM) | % Self-Reported Stimulant-Abstinent Days at Post-Treatment (CM Groups Only) | 92.1 % stimulant-abstinent days | Standard Deviation 11.5 |
p-value: 0.92Wilcoxon (Mann-Whitney)
Secondary
Total Non-CM Treatment Encounters Per Participant
Number of non-CM, recovery-oriented treatment encounters during the 12-week treatment interval, computed on a per-participant basis. Encounters were documented in the electronic health record (VA encounters) and/or self-reported (non-VA encounters, including interactions with mutual support communities). Because this measure reflects the absolute number of non-CM treatment encounters per participant, only individuals with complete data for the 12-week period were included.
Time frame: 12-Week Treatment Interval
Population: Only participants with complete data for the 12-week treatment period were included in this measure. Accordingly, two participants who withdrew early were excluded from the TAU arm, and one participant was excluded from the TangiblePBCM arm due to missing data. Results for the TAU arm are summarized using descriptive statistics but are omitted from statistical comparisons due to insufficient data (n = 4, with 2 out of 4 withdrawn from the study).
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Total Non-CM Treatment Encounters Per Participant | 10.7 count of encounters per participant | Standard Deviation 15.8 |
| Voucher Prize-Based Contingency Management (VoucherPBCM) | Total Non-CM Treatment Encounters Per Participant | 12.5 count of encounters per participant | Standard Deviation 12.9 |
| Treatment As Usual (TAU) | Total Non-CM Treatment Encounters Per Participant | 20.0 count of encounters per participant | Standard Deviation 24 |
p-value: 0.384Wilcoxon (Mann-Whitney)
Other Pre-specified
Difference in Control-related Theta Synchronization From Baseline to Post-Treatment
Theta synchronization between anterior cingulate cortex (Cz) and lateral prefrontal cortex (F3, F4, FC5, FC6) was measured to assess differences in functional connectivity between baseline and post-treatment (i.e., follow-up assessment conducted after the conclusion of the 12-week treatment interval) for high versus low conflict events. At each timepoint, phase locking value (PLV) was computed for 4 electrode pairs (Cz-F3, Cz-F4, Cz-FC5, Cz-FC6) across 3 stimulus conditions ranging from low to high conflict (Congruent, Intermediate-Incongruent, High-Incongruent) and for high-conflict error versus low-conflict correct responses in the Parametric Conflict Flankers task. A wavelet transformation isolated theta activity, and PLV was derived from normalized complex wavelet phases. PLV quantifies trial-to-trial phase synchrony (0-1), with 1 representing perfect synchrony. Descriptive data reported below represent the average PLV across the 4 electrode pairs.
Time frame: Baseline and Post-Treatment Follow-up
Population: To accommodate the smaller final sample relative to the original target, analyses of neural and cognitive-behavioral outcomes were combined across arms. A revised analytic plan, approved prior to analysis, examined longitudinal differences by treatment engagement (attendance) and abstinence across arms.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Control-related Theta Synchronization From Baseline to Post-Treatment | Baseline Congruent | 0.163 Phase Locking Value | Standard Deviation 0.096 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Control-related Theta Synchronization From Baseline to Post-Treatment | Baseline Intermediate Incongruent | 0.173 Phase Locking Value | Standard Deviation 0.1 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Control-related Theta Synchronization From Baseline to Post-Treatment | Baseline High Incongruent | 0.205 Phase Locking Value | Standard Deviation 0.098 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Control-related Theta Synchronization From Baseline to Post-Treatment | Follow-Up Congruent | 0.158 Phase Locking Value | Standard Deviation 0.084 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Control-related Theta Synchronization From Baseline to Post-Treatment | Follow-Up Intermediate Incongruent | 0.165 Phase Locking Value | Standard Deviation 0.083 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Control-related Theta Synchronization From Baseline to Post-Treatment | Follow-Up High Incongruent | 0.200 Phase Locking Value | Standard Deviation 0.088 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Control-related Theta Synchronization From Baseline to Post-Treatment | Baseline Correct | 0.169 Phase Locking Value | Standard Deviation 0.099 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Control-related Theta Synchronization From Baseline to Post-Treatment | Baseline Error | 0.273 Phase Locking Value | Standard Deviation 0.146 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Control-related Theta Synchronization From Baseline to Post-Treatment | Follow-Up Correct | 0.158 Phase Locking Value | Standard Deviation 0.086 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Control-related Theta Synchronization From Baseline to Post-Treatment | Follow-Up Error | 0.245 Phase Locking Value | Standard Deviation 0.104 |
Comparison: Due to violations of normality assumptions, a non-parametric Aligned Rank Transform ANOVA was conducted, which does not permit inclusion of covariates. A median split approach was therefore used to examine the influence of Contingency Management session attendance. The design included 4 factors: time (2 levels), stimulus condition (3 levels), electrode pair (4 levels), and median split group membership (2 levels). Reported results reflect the interaction between time and median split group.p-value: <0.001ANOVA
Comparison: Due to violations of normality assumptions, a non-parametric Aligned Rank Transform ANOVA was conducted, which does not permit covariates. A median split approach was therefore used to examine the influence of self-reported cocaine abstinence during treatment. The design included 4 factors: time (2 levels), stimulus condition (3 levels), electrode pair (4 levels), and median split group membership (2 levels). Reported results reflect the interaction between time and median split group.p-value: <0.001ANOVA
Comparison: Due to violations of normality assumptions, a non-parametric Aligned Rank Transform ANOVA was conducted, which does not permit inclusion of covariates. A median split approach was therefore used to examine the influence of Contingency Management session attendance. The design included 4 factors: time (2 levels), response accuracy (2 levels), electrode pair (4 levels), and median split group membership (2 levels). Reported results reflect the interaction between time and median split group.p-value: 0.022ANOVA
Comparison: Due to violations of normality assumptions, a non-parametric Aligned Rank Transform ANOVA was conducted, which does not permit covariates. A median split approach was therefore used to examine the influence of self-reported cocaine abstinence during treatment. The design included 4 factors: time (2 levels), response accuracy (2 levels), electrode pair (4 levels), and median split group membership (2 levels). Reported results reflect the interaction between time and median split group.p-value: 0.024ANOVA
Other Pre-specified
Difference in Episodic Future Thinking Effect on Delay Discounting From Baseline to Post-Treatment
Delay discounting was quantified using the hyperbolic discounting parameter (k), estimated separately for the Episodic Future Thinking (EFT) condition-with personally meaningful event tags anchoring delayed rewards-and the Standard condition without such tags. Values of k were natural log-transformed (ln(k)), with higher ln(k) indicating steeper discounting of delayed rewards, reflecting greater devaluation of future outcomes and a stronger preference for immediate gratification.
Time frame: Baseline and Post-Treatment Follow-up
Population: To accommodate the smaller final sample relative to the original target, analyses of neural and cognitive-behavioral outcomes were combined across arms. A revised analytic plan, approved prior to analysis, examined longitudinal differences by treatment engagement (attendance) and abstinence across arms.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Episodic Future Thinking Effect on Delay Discounting From Baseline to Post-Treatment | EFT ln(k) Baseline | -3.62 ln(k) delay discounting parameter | Standard Deviation 2.15 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Episodic Future Thinking Effect on Delay Discounting From Baseline to Post-Treatment | Standard ln(k) Baseline | -2.42 ln(k) delay discounting parameter | Standard Deviation 3.28 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Episodic Future Thinking Effect on Delay Discounting From Baseline to Post-Treatment | EFT ln(k) Follow-up | -3.36 ln(k) delay discounting parameter | Standard Deviation 2.47 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Episodic Future Thinking Effect on Delay Discounting From Baseline to Post-Treatment | Standard ln(k) Follow-up | -2.77 ln(k) delay discounting parameter | Standard Deviation 3.3 |
Comparison: Due to violations of normality assumptions, a non-parametric Aligned Rank Transform ANOVA was conducted, which does not permit inclusion of covariates. A median split approach was therefore used to examine the potential influence of Contingency Management session attendance. The design included 3 factors: time (2 levels), condition (2 levels), and median split group membership (2 levels). Reported results reflect the interaction between time, condition, and median split group.p-value: 0.878ANOVA
Comparison: Due to violations of normality assumptions, a non-parametric Aligned Rank Transform ANOVA was conducted, which does not permit inclusion of covariates. A median split approach was therefore used to examine the potential influence of self-reported cocaine abstinence during treatment. The design included 3 factors: time (2 levels), condition (2 levels), and median split group membership (2 levels). Reported results reflect the interaction between time, condition, and median split group.p-value: 0.645ANOVA
Other Pre-specified
Difference in Executive Working Memory From Baseline to Post-Treatment
Difference in Brown-Peterson working memory scores between baseline and post-treatment (i.e., follow-up assessment conducted after the conclusion of the 12-week treatment interval). We will specifically use a modified Brown-Peterson test (Auditory Consonant Trigrams) for which both age- and Veteran-specific norms exist. Summary scores for this measure (including 9-, 18-, and 36-second delay conditions) can range from 0-45, with higher scores indicating improved executive working memory performance.
Time frame: Baseline and Post-Treatment Follow-up
Population: To accommodate the smaller final sample relative to the original target, analyses of neural and cognitive-behavioral outcomes were combined across arms. A revised analytic plan, approved prior to analysis, examined longitudinal differences by treatment engagement (attendance) and abstinence across arms.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Executive Working Memory From Baseline to Post-Treatment | Score at Baseline | 24.37 total score | Standard Deviation 7.82 |
| Tangible Prize-Based Contingency Management (TangiblePBCM) | Difference in Executive Working Memory From Baseline to Post-Treatment | Score at Follow-up | 24.78 total score | Standard Deviation 7.9 |
Comparison: No violations of normality assumptions were identified. A repeated measures ANOVA was conducted. To maintain consistency with other outcome measures, a median split approach was used to examine the potential influence of Contingency Management session attendance. The design included two factors: time (2 levels) and median split group membership (2 levels). Reported results reflect the interaction between time and median split group.p-value: 0.973ANOVA
Comparison: No violations of normality assumptions were identified. A repeated measures ANOVA was conducted. To maintain consistency with other outcome measures, a median split approach was used to examine the potential influence of self-reported cocaine abstinence during the 12-week treatment interval. The design included two factors: time (2 levels) and median split group membership (2 levels). Reported results reflect the interaction between time and median split group.p-value: 0.662ANOVA