Neurocognitive Factors in Substance Use Treatment Response: The Ways of Rewarding Abstinence Project: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Neurocognitive Factors in Substance Use Treatment Response: The Ways of Rewarding Abstinence Project (NCT03799341), a randomized na study. The primary outcome is % Cocaine-Negative Urine Specimens Per Participant.

2. Clinical Question

This is a question about comparative outcomes in the groups studied, not a question answered by a single arm’s average in isolation. The registry reports group means and standard deviations for the primary measure and a Wilcoxon (Mann-Whitney) test for the comparison. It does not report a confidence interval or an estimated between-group difference for that primary comparison. Consequently, the posted record allows a description of the arm-level averages and the reported test, but does not establish the precision of a primary effect estimate. The Treatment As Usual results are descriptive only and do not furnish a reliable statistical comparison with either contingency-management group.

3. Trial Design

The study is registered as NCT03799341, with the title Neurocognitive Factors in Substance Use Treatment Response: The Ways of Rewarding Abstinence Project. Its official title is Electrophysiological Predictors and Indicators of Contingency Management Treatment Response. The registry describes a randomized, parallel-group design with no masking. Phase is listed as NA, and enrollment is 63. The sponsor is the VA Office of Research and Development. The study start is recorded as 2019-11-13, primary completion as 2023-10-31, and status as completed.

The listed behavioral interventions include Prize-Based Contingency Management and Treatment As Usual Outpatient Substance Use Treatment. The two contingency-management arms were TangiblePBCM and VoucherPBCM. The tangible approach linked prize draws to access to a cabinet containing financial incentive items; the voucher approach linked prize draws to VA Canteen vouchers. Treatment As Usual is the third named arm. The registry’s primary statistical comparison is between the two prize-based contingency-management groups, rather than a three-arm comparison.

For the primary endpoint, the registry reports 22.0 participants in TangiblePBCM, 19.0 in VoucherPBCM, and 4.0 in Treatment As Usual. The small Treatment As Usual result group was not included in statistical comparisons; the record states that 2 out of 4 participants withdrew during the 12-week treatment interval. Its mean and standard deviation remain descriptive information, not evidence from a formal comparison.

The study description emphasizes neurocognitive predictors and indicators of treatment response, but the registry record does not report enough detail to reconstruct the full intervention schedule, randomization procedure, allocation concealment, or analysis populations. These design details are not reported in the registry record. The presence of randomization identifies the broad allocation design; it does not by itself resolve limitations arising from small groups, missing outcome information, or incomplete reporting of the analysis plan.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: 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.

12-Week Treatment Interval

Reasons for not completing this period:

Post-Tx Evaluation Session

Reasons for not completing this period:

6-Month Post-Tx Follow-Up Period

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 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).

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as % Cocaine-Negative Urine Specimens Per Participant, measured over 12-Week Treatment Interval.

The primary endpoint is the percentage of cocaine-negative urine specimens per participant during the 12-Week Treatment Interval. For each participant, the measure expresses the number of negative specimens relative to the number of possible specimens. It is a continuous measure, and the registry summarizes it within each arm using a mean and standard deviation. A cocaine-negative specimen is a direct laboratory-based indicator for the target substance at the time represented by that specimen; it is not, by itself, a complete account of every participant’s substance use over the interval.

Using a per-participant percentage makes the endpoint a measure of the proportion of possible specimen results that were negative, rather than simply a total count of negative specimens across an arm. The registry says that denominators were adjusted for participants who could not receive a full course of treatment because face-to-face research activities were suspended during the COVID-19 pandemic. It also says denominators were adjusted for 2 Treatment As Usual participants who withdrew during the treatment interval. The exact individual-level missing-data procedures beyond those denominator adjustments are not reported in the registry record.

The endpoint is useful for describing treatment-period cocaine-negative results, but it does not capture every dimension of recovery, treatment engagement, or longer-term outcome. It should be interpreted alongside the group sizes, variability, and limitations in the reported comparison. The registry does not report a confidence interval or a between-group effect estimate for this primary endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is continuous: it is a participant-level percentage, and the registry reports a mean and standard deviation for each arm. A mean is the arithmetic average of participant values. A standard deviation describes how dispersed those values are around the mean; it is not a confidence interval and does not describe the precision of the mean itself. In general, a larger standard deviation means participant outcomes are more variable, while precision also depends on the number of participants and the analysis method.

The primary comparison reported by the registry uses the Wilcoxon (Mann-Whitney) method. In general, this is a rank-based method for comparing two independent groups. It evaluates how values from one group are positioned relative to values from the other group and does not require the comparison to be framed as a difference between means. It is not automatically a test of mean difference, and its result should not be interpreted as a direct estimate of how many percentage points separate the group means. The registry reports p = 0.177 but no corresponding primary confidence interval or effect estimate.

In general, a p-value measures how compatible the observed data are with a specified null model under the assumptions of a test. A larger p-value does not prove that there is no difference, and it does not give the probability that the null model is true. It also does not describe the magnitude of a difference. For this trial, interpretation is especially limited because the primary result has no reported interval estimate and the descriptive standard deviations are large relative to the reported group means.

The registry’s secondary results include recalculated unadjusted mean differences for selected continuous outcomes. Such a recalculation is a simple comparison of posted summary statistics; it is not the same as the registry’s reported rank-based test. For these secondary measures, the recalculated mean-difference estimates have confidence intervals, while the registry separately reports Wilcoxon (Mann-Whitney) p-values. Different methods can produce different numerical summaries because they target or use the data differently. The recalculated estimates are not substitutes for the sponsor’s own analysis, and they should be understood as exploratory descriptions of posted summaries.

In general, a confidence interval gives a range of values compatible with a statistical model and the observed data at a stated confidence level. A wide interval indicates that the estimate is uncertain; an interval that includes zero in a mean-difference analysis is compatible with no mean difference as well as with effects in either direction. The secondary recalculated intervals span zero. That observation does not establish that the outcomes are equal; it indicates that those estimates do not isolate a direction or size with precision under the stated calculation.

Planning illustrations in the statistical material use the observed effect as a starting point and consider smaller fractions of it. They are based on a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. In general, power is the probability that a planned study would detect a specified effect under a particular model and assumptions. The posted planning figures are not observed power and do not describe the probability that the completed trial detected an effect. They are illustrative sample-size calculations, not the trial’s documented design justification. The trial’s own sample-size justification is not reported in the registry record.

Over-dispersion is relevant when outcome variability is larger than a statistical model assumes. In general, extra variability can make standard errors and confidence intervals larger than a model that assumes less variation would suggest. The primary measure has sizable within-arm standard deviations, but the registry does not report a model-based over-dispersion assessment. It would therefore be inappropriate to claim that a particular adjustment was or was not made.

The record does not report the detailed statistical analysis plan, the exact analysis population, how missing observations were handled beyond the stated denominator adjustments, or whether interim analyses were conducted. The reported endpoint descriptions and methods identify what was analyzed at a high level, but they do not support assumptions about covariate adjustment, a prespecified hierarchy among outcomes, or other model features. Those details are not reported in the registry record.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Tangible Prize-Based Contingency Management (Tangi", "Voucher Prize-Based Contingency Management (Vouche", "Treatment As Usual (TAU)")
n   <- c(22, 19, 4)
m   <- c(42.2, 59.1, 51)
sp  <- c(37.4, 34.5, 18.8)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Tangible Prize-Based Contingency Management (Tangi  22  42.200  37.400  25.618  58.782
CARM  Voucher Prize-Based Contingency Management (Vouche  19  59.100  34.500  42.472  75.728
CARM  Treatment As Usual (TAU)  4  51.000  18.800  21.085  80.915

11. Analyses Reported by the Sponsor

The registry’s primary analysis is the sponsor-reported Wilcoxon (Mann-Whitney) comparison of TangiblePBCM and VoucherPBCM, with p = 0.177. No primary recalculated mean difference or confidence interval is reported in the statistical material, so there is no independent primary estimate available to set alongside the sponsor’s test result. The arm-level means and standard deviations are descriptive and should not be treated as a replacement for that analysis.

For the secondary measures with recalculated results, the unadjusted mean differences and confidence intervals are simple calculations from posted summary statistics. The sponsor’s reported analyses for those comparisons use Wilcoxon (Mann-Whitney) tests. Because these are different statistical summaries, their estimates and p-values need not match exactly or answer identical questions. The record does not flag a discrepancy between the sponsor’s results and the recalculations.

When a discrepancy is flagged, the sponsor’s own analysis takes precedence over a simple unadjusted recalculation. In this record, the central interpretive limitation is instead that the primary result lacks a reported effect estimate and confidence interval, while the secondary recalculations remain exploratory and unadjusted. Neither should be used to imply more precision than the registry reports.

12. Multiplicity, Missing Data and Analysis Populations

The registry identifies one primary endpoint and multiple secondary outcomes. The primary endpoint is the percentage of cocaine-negative urine specimens per participant during the 12-Week Treatment Interval. Secondary outcomes include attendance, treatment encounters, self-reported abstinence, and post-treatment abstinence measures. When researchers test multiple outcomes or comparisons, the probability of at least one apparently unusual result can increase. In general, multiplicity procedures or a prespecified testing hierarchy can help control how such evidence is interpreted. A multiplicity adjustment or hierarchy is not reported in the registry record.

The posted secondary recalculations are unadjusted. Their confidence intervals and any related p-values should consequently be read as exploratory, particularly when considered across several outcomes. The registry’s primary Wilcoxon test is a distinct reported analysis; the absence of reported multiplicity details does not change its numerical result, but it limits understanding of how the primary and secondary tests were planned and interpreted together. The statistical analysis plan is not reported in the registry record.

Missingness is particularly relevant to the primary endpoint. The record states that denominators were adjusted for participants who could not receive a full course of treatment because face-to-face research activities were suspended during the COVID-19 pandemic. It also describes denominator adjustments for 2 Treatment As Usual participants who withdrew during the 12-week treatment interval. These statements clarify that possible specimens were not treated identically for every participant, but they do not fully specify how missing specimen results were classified or handled, nor whether missingness differed across the two primary comparison arms.

The Treatment As Usual summary is especially limited: the primary result description reports n = 4, with 2 out of 4 withdrawn, and says that arm was omitted from statistical comparisons due to insufficient data. This is an explicit reason not to interpret its descriptive mean as a stable comparative estimate. The record does not report a complete account of withdrawals by reason and arm for all outcomes, an imputation strategy, or sensitivity analyses for missing data. Those details are not reported in the registry record.

The exact analysis populations are not reported in the registry record. It is therefore not possible to determine from the posted description whether analyses followed an intention-to-treat approach, a per-protocol approach, or another population definition. Interim analyses are also not reported. These omissions matter because they affect how the estimates should be understood, but they do not justify assuming that a particular method was used.

13. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: hospitalization due to cardiac symptoms (25.0%), initial or prolonged hospitalization due to stroke (25.0%), hospitalization for psychiatric symptoms (15.8%), hospitalization due to acute intoxication (10.5%), hospitalization due to neurological symptoms (5.3%).

The registry reports serious adverse events affecting 5 of 22 participants in TangiblePBCM, 5 of 19 participants in VoucherPBCM, and 1 of 4 participants in Treatment As Usual. It reports deaths of 1, 0, and 0 participants, respectively. The record reports 0 participants with other, non-serious adverse events affecting them in each of the three groups. These figures describe events recorded in this study; they are not a formal comparison of safety between groups.

The small Treatment As Usual group makes its event count particularly difficult to compare with the larger groups. The registry does not report a statistical test or confidence interval for the adverse-event comparisons, and it does not establish that differences in event counts are attributable to an intervention. Serious adverse events included hospitalizations for cardiac symptoms, stroke, psychiatric symptoms, acute intoxication, neurological symptoms, and skin infection. The record gives percentages for listed serious-event terms, but those percentages should not be interpreted as participant-level risks without a clearly specified denominator.

Adverse-event counts must be read in the context of the study’s enrolled participants and the registry’s reporting. The available figures do not show the timing, duration, relatedness assessment, or full clinical course of each event. Those details are not reported in the registry record. The figures therefore serve as a descriptive safety account, not as evidence for a causal safety conclusion or an overall benefit-risk judgment.

14. What the Result Does, and Does Not, Mean

The registry result describes outcomes among participants assigned to two prize-based contingency-management approaches and a small Treatment As Usual group. For the primary endpoint, the posted arm means differ, while the reported comparison is a Wilcoxon (Mann-Whitney) test with p = 0.177. The result does not provide a primary effect estimate or confidence interval. It therefore does not precisely quantify the size or range of a possible difference between the two contingency-management groups.

A p-value is not a probability that a treatment works, and a result that is not statistically compelling does not prove that the groups are equivalent. Conversely, a difference between descriptive means does not establish that one approach caused better outcomes. Interpreting the comparison requires attention to the randomized design, group sizes, outcome variability, missing-data adjustments, and the absence of an interval estimate for the primary result.

The endpoint captures cocaine-negative urine specimens over the treatment interval, not all aspects of substance use, recovery, or treatment response. Secondary measures add information about attendance, treatment encounters, and reported abstinence, but they measure different aspects of care and behavior. They are unadjusted and should not be treated as independent confirmation of the primary comparison. The Treatment As Usual results are descriptive and too limited for a reliable inferential comparison in this record.

The registry’s stated research aim concerns neurocognitive predictors and indicators of response, but the posted primary result alone does not answer whether neurocognitive measures predict response or which individual treatment would be best matched to a participant. Detailed findings for that broader aim are not reported in the registry record. Nor does this result establish what would happen in other populations or settings.

In short, the record documents a randomized study and reports a primary group comparison, but the available inferential information is limited. The findings should be described as the observed results and reported analysis, not as medical advice or a definitive statement that one treatment is superior, inferior, or equivalent.

15. Why This Trial Matters Statistically

This trial raises several important statistical issues for clinical researchers. First, it illustrates the difference between a descriptive contrast and a comparative estimate. Arm means and standard deviations describe the observed participants, while a test result addresses a particular statistical comparison. Without a primary effect estimate and confidence interval, the registry offers limited information about the magnitude and precision of the contrast that matters to interpretation.

Second, variability and small groups affect how confidently a result can be characterized. Participant-level percentages show substantial spread in the two primary comparison arms, and the Treatment As Usual result group is very small. A large standard deviation does not automatically invalidate a mean, but it signals that participants’ outcomes vary and that the average may conceal heterogeneous experiences. Precise inference depends on both variability and sample size, among other assumptions.

Third, pandemic-related suspension of face-to-face research activities affected the possible treatment course and led to denominator adjustments. This highlights why outcome definitions and missing-data procedures matter: a percentage based on a participant-specific denominator depends on how possible observations are defined. The registry states the adjustment but does not report the full missing-data strategy or sensitivity analyses.

Finally, the study includes a primary endpoint alongside multiple secondary outcomes. Clear separation of confirmatory and exploratory analyses is important because multiple unadjusted tests can create misleading impressions if individual results are overemphasized. The registry record does not report a multiplicity plan or full analysis plan. These limitations do not erase the study’s descriptive contribution, but they define how far its statistical conclusions can reasonably extend.

16. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several limitations are specific to this registry record. The primary statistical comparison is limited to TangiblePBCM versus VoucherPBCM, while the Treatment As Usual arm is omitted from inferential comparisons because of insufficient data. The record reports only n = 4 for that arm’s primary result group and notes that 2 out of 4 participants withdrew during the treatment interval. Its mean is consequently descriptive and should not be treated as a comparative estimate.

The primary analysis reports a Wilcoxon (Mann-Whitney) p-value but no effect estimate or confidence interval. The arm-level means and standard deviations do not fill that reporting gap, because they do not specify the uncertainty around a primary contrast or describe the rank-based test’s estimated effect. The large within-arm standard deviations also indicate varied participant-level results.

The record describes denominator adjustments for pandemic-related suspension of face-to-face activities and withdrawals, but does not fully report how missing specimens were handled beyond those adjustments. Detailed eligibility criteria, baseline characteristics, allocation procedures, the analysis population, the statistical analysis plan, and interim analyses are not reported in the registry record. Their absence limits assessment of representativeness, implementation, and the exact inferential framework.

Secondary results are unadjusted and cover different constructs and follow-up periods. The registry does not report whether multiplicity was addressed or how the secondary outcomes were prioritized. The neurocognitive research aim is broader than the primary outcome, but detailed results addressing predictors of response are not reported in the registry record. These constraints call for a narrow reading of the posted endpoint results rather than extrapolation to unreported analyses or populations.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Neurocognitive Factors in Substance Use Treatment Response: The Ways of Rewarding Abstinence Project — NCT03799341.
• Registry record last updated 2025-09-30; status last verified 2025-09.
• Results first posted 2025-09-30.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

20. Record Summary

NCT03799341 is a completed randomized, parallel-group study with no masking and enrollment of 63. It compares Tangible Prize-Based Contingency Management and Voucher Prize-Based Contingency Management, with Treatment As Usual also named as an arm. The primary endpoint is the percentage of cocaine-negative urine specimens per participant during the 12-Week Treatment Interval. The registry reports arm-level means and standard deviations and a Wilcoxon (Mann-Whitney) test for the two contingency-management groups, with p = 0.177.

The Treatment As Usual primary results are descriptive and excluded from statistical comparisons because of insufficient data. The registry reports secondary outcomes and serious adverse-event counts, but the available secondary comparisons are unadjusted and do not resolve the primary comparison. The primary effect estimate and confidence interval, detailed missing-data approach, analysis population, multiplicity plan, and full statistical analysis plan are not reported in the registry record. Interpretation is therefore limited to the posted summaries and reported test.