Cognitive Enhancement Through Computerized Training: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Cognitive Enhancement Through Computerized Training (NCT04244032), a randomized na study. The primary outcome is Change in Working Memory Performance.

2. Clinical Question

3. Trial Design

The registry record describes a randomized, parallel-group behavioral study of computerized cognitive training for alcohol use disorder. The listed phase is NA. The study began on 2021-04-28, had a primary completion date of 2023-09-22, and is marked terminated. The record attributes early termination to low enrollment.

Four groups are described. Working Memory Training used a working-memory task with individually adapted difficulty levels. Inhibitory Control Training used an inhibitory-control task with individually adapted difficulty levels. Bias Modification Training was an active comparator task intended to weaken approach responses to alcohol-associated cues. The Non-Trained Control received typical care without an additional active intervention. The summary frames computerized training as an adjunct to inpatient treatment, but does not give details of the care received in any group.

Allocation is recorded as randomized, the model as parallel, and masking as none. The registry record does not report how the random sequence was generated, whether allocation was concealed, or which people involved in the study knew group assignment. It also does not report recruitment procedures, planned group allocation, eligibility criteria, or intervention duration.

Total enrollment was 22. The record says no participants in the three training groups were analyzed for the primary working-memory outcome because the study ended early due to low enrollment. It reports outcome data for 10 participants in the non-trained control group. Thus, the posted primary outcome does not support a between-group estimate. The registry record lists three outcome measures, but no secondary outcome results are posted. The study summary also mentions interest in neurocognitive recovery and post-discharge alcohol consumption; results for those topics are not reported in the registry record.

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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: No participants were analyzed for Working Memory Training (WM), Inhibitory Control Training (IC), and Bias Modification Training (BM) due to early termination of the study resulting from low enrollment. Outcome data are reported for Non-Trained control (NTC) only.

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.

Population: Baseline includes screen failures with available demographic data for descriptive purposes only. Participant Flow includes only those randomized. No participants were analyzed due to insufficient enrollment. The study did not meet recruitment targets, and no data were available for analysis.

6. Primary Endpoint

The registry lists the primary outcome as Change in Working Memory Performance, measured over Baseline to Post-Test, Up to 3 months.

The primary endpoint, Change in Working Memory Performance, measures change in percent accuracy on neuropsychological measures indexing working-memory capacity. The recorded timeframe is baseline to post-test, up to 3 months. Scores range from 0% to 100%, with higher percentages indicating better performance. The endpoint is expressed in percent points, and the reported summary measure is a mean with a standard deviation.

Working memory is relevant to the study’s stated interest in neurocognitive impairment and recovery in alcohol use disorder. In general, a standardized task-based measure can make performance comparable across assessments, while a baseline-to-post-test change measure focuses on within-person movement over the study period rather than the post-test score alone. The registry record does not describe the specific neuropsychological tests, how multiple measures were combined, or how the change score was calculated.

For the non-trained control group, the record reports a mean change of 1.61 percent points with a standard deviation of 0.07 among 10 participants. No primary endpoint values are reported for the three training groups. Without their results, the control-group summary alone cannot estimate the difference in change between training and control, which is the comparison needed to assess the randomized group contrast.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous measure: percent accuracy change in working-memory performance. The registry record specifies the mean as the summary measure and standard deviation as its dispersion measure. It reports these statistics for the non-trained control group, but no outcome values for the three training groups. No inferential analysis of a group contrast is reported.

What does a mean describe?

In general, a mean is the arithmetic average of measured values in a group. For a change endpoint, each value represents a participant’s change over the specified assessment period, and the group mean summarizes the average change among participants with outcome data. Here, the posted control-group mean is 1.61 percent points. It is a descriptive summary for that group, not an estimate of the difference between a training group and control.

In general, the standard deviation describes how dispersed individual observations are around their group mean. The control-group standard deviation is 0.07. It should not be read as the uncertainty around a comparison, nor does it convey the range of individual outcomes. The registry record does not report a confidence interval for the mean or for a between-group effect.

Why is a group contrast needed?

For a randomized trial, the central comparative question is generally whether outcome changes differ between assigned groups. A contrast would require outcome information from both the training group and its comparator, and an estimate of uncertainty would help show how compatible the observed data are with a range of possible effects. Because primary outcome data are reported only for the Non-Trained Control group, a training-versus-control contrast is unavailable. The posted control mean and standard deviation cannot substitute for missing training-arm results.

The registry record identifies the study as randomized and parallel, but it does not describe the randomization method, allocation concealment, or statistical model. Those details are not reported in the registry record. It would therefore be inappropriate to assume an adjusted analysis, a particular test, or a prespecified method for handling repeated measurements. The record’s designation of the endpoint as a mean with standard deviation does not establish which inferential procedure was intended.

What does early termination imply for interpretation?

The study is marked terminated, and the record says early termination resulted from low enrollment. It reports 22 participants enrolled overall, but only 10 participants contribute to the posted primary outcome, all in the non-trained control group. This prevents a direct comparison for that outcome. General statistical explanation: small or incomplete datasets can yield imprecise estimates, but the more immediate issue in this record is that the active-group outcome estimates are absent. No estimate of their uncertainty can be interpreted because no such results are posted.

The registry record lists zero analyses. It does not report an analysis population definition, a missing-data procedure, or any adjustment for baseline measurements or other variables. General statistical explanation: an analysis plan matters because choices about which participants and observations are included can affect the interpretation of a result. Without the relevant documentation and active-arm outcome data, no assumptions about those choices should be made.

How should planning calculations be read?

General statistical explanation: sample-size planning illustrations estimate how many participants might be needed under stated assumptions to detect a specified effect with a chosen significance level and power. They are not observed power and do not show what the trial actually achieved. The registry record describes a planning approach based on the observed effect, two-sided significance level 0.05, equal allocation, and a normal approximation; for a continuous endpoint, the stated convention is participants per arm. However, the posted primary outcome lacks an observed training-versus-control effect. No defensible comparative effect for this endpoint is available from these results, and planning assumptions should not be treated as trial findings.

General statistical explanation: when variability is greater than expected, a fixed sample can provide less precision for a group comparison. In this record, the control-group standard deviation is reported, but the active-arm spread and a between-group effect are not. The available data therefore do not support interpreting comparative precision, statistical significance, or the magnitude of a training effect.

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("Non-Trained Control (NTC)")
n   <- c(10)
m   <- c(1.61)
sp  <- c(0.07)
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  Non-Trained Control (NTC)  10  1.610  0.070  1.560  1.660

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists three outcome measures, reports no secondary outcome results, and lists zero analyses. The primary working-memory endpoint has outcome data only for the Non-Trained Control group. This is an absence of comparative outcome information, not a set of null findings across the training groups.

Multiplicity concerns arise in general when a study makes several statistical comparisons or tests several outcomes: the chance of at least one apparently notable result can increase as the number of tests grows. The registry record does not report a statistical analysis plan, a hierarchy of outcomes, or any method for addressing multiplicity. Because no analyses are listed, there is no reported multiplicity procedure to interpret. The number of outcome measures alone does not establish how many tests were planned or performed.

Missing outcome data are especially consequential because none of the three training groups has analyzed primary endpoint data, while the control group has data for 10 participants. The record attributes the absence of training-group analysis to early termination due to low enrollment. It does not report participant-level reasons for missing measurements, timing of withdrawals, whether post-test assessments were attempted, or methods used to handle missingness. Those details are not reported in the registry record.

In general, missing-data methods rely on assumptions about why observations are unavailable. No imputation, sensitivity analysis, or other missing-data method is reported, and none should be inferred. The registry record also does not define an intention-to-treat or per-protocol analysis population. Although the randomized design is stated, the posted outcome set cannot provide a randomized comparison when active-arm outcome data are absent.

The registry record reports no participants analyzed for the training groups and no secondary outcome results. It does not clarify whether additional participants had partial assessments or whether data were collected but not posted. Interpretation should therefore remain bounded to the outcome information actually recorded: a control-group summary and no reported comparative analysis.

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

The primary endpoint describes working-memory accuracy change from baseline to post-test, over a timeframe of up to 3 months. The registry record reports a mean change of 1.61 percent points and a standard deviation of 0.07 for 10 participants in the Non-Trained Control group. This is the extent of the posted primary outcome result that can be interpreted numerically.

The result does not show whether working-memory training, inhibitory-control training, or bias-modification training changed performance relative to control. No analyzed primary outcome data are reported for those groups. It also does not establish whether any observed change is clinically important, whether the endpoint reflects broader cognitive recovery, or whether cognitive outcomes relate to post-discharge alcohol consumption. Such conclusions would require data and analyses not reported in the registry record.

In general, a single group’s mean change can describe the observed group but cannot separate changes associated with an intervention from other changes over time. In a randomized study, the group contrast is important for that purpose. The absence of training-arm outcome estimates means no such contrast is available. The control result should not be framed as an estimate of what would have happened in the training groups.

The standard deviation describes variability among the reported control observations; it is not a measure of the precision of a training effect. No confidence interval or inferential test for a group difference is reported. Nor does the record’s zero count for participants at risk in adverse-event listings establish that the interventions were free of adverse events: the record lists zero participants at risk in each group, as well as zero events. Those entries do not provide an interpretable safety comparison.

Overall, the registry record documents a randomized study that ended early because of low enrollment and a limited set of primary outcome observations. It supports description of the reported control-group result, but not a conclusion about comparative effectiveness, safety, or the broader aims described in the study summary.

13. Why This Trial Matters Statistically

This study is statistically instructive because it illustrates the difference between enrolling participants and obtaining analyzable outcomes for a randomized comparison. The registry record lists 22 participants enrolled, but the primary working-memory result is reported for 10 control participants only. No training-group participants were analyzed for that endpoint after early termination due to low enrollment. As a result, the primary comparison that the randomized design was intended to support is not available in the posted results.

In general, randomization helps make groups comparable at the start of a study, but it does not eliminate the need for outcome data from each group. A control-group mean, even when paired with a standard deviation, cannot stand in for the missing training-group results. This distinction is central to reading trial summaries: descriptive statistics for one group are not equivalent to an estimate of an intervention’s effect.

The record also highlights the role of precision and analysis transparency. No confidence interval, p-value, or analysis is reported for the primary endpoint. The registry record does not describe the intended statistical model or the handling of missing outcomes. Without those results and details, readers cannot assess the precision of a comparative estimate, the analysis population, or the robustness of findings.

Finally, the record’s study aims were broader than the posted primary result. It describes neurocognitive recovery and post-discharge alcohol consumption as interests, yet no secondary outcome results are posted. Statistically, a carefully bounded interpretation matters: the available data describe one group’s reported outcome, while the central comparative and broader questions remain unanswered in the registry record.

14. 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.

The principal limitation is that early termination due to low enrollment left primary working-memory outcome data only for the Non-Trained Control group. The three training groups have no analyzed participants for this endpoint. As a result, no randomized comparison or estimate of a training-versus-control difference is reported.

The total enrollment of 22 should be distinguished from the 10 participants with reported control-group outcome data. The registry record does not provide a full participant flow, group-specific enrollment counts beyond the endpoint listings, reasons for individual missing observations, or information about whether participants completed the planned assessments. It also does not report how the primary change score was derived from the neuropsychological measures.

Important design and analysis details are absent. The record does not report eligibility criteria, random sequence generation, allocation concealment, intervention duration, the specific assessment instruments, a statistical analysis plan, a missing-data method, or an analysis-population definition. No confidence interval or inferential test is posted. These omissions limit assessment of the planned and conducted analyses.

There is also ambiguity in the registry record’s control labeling: the control identifier in the arm listing is associated with Inhibitory Control Training, whereas the primary outcome description says outcome data are reported for Non-Trained Control only. This inconsistency warrants caution when interpreting the group coding. The reported outcome description identifies the group with data, but the record does not explain the discrepancy.

Adverse-event listings show zero participants at risk and zero events in each group. Because no participants are listed as at risk, these entries cannot establish that no adverse events occurred among a defined population. The record does not provide enough information to interpret those counts as a comparative safety result.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Cognitive Enhancement Through Computerized Training — NCT04244032.
• Registry record last updated 2026-09-16; status last verified 2024-12.
• Results first posted 2026-09-16.
• 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.

18. Record Summary

NCT04244032, titled Cognitive Enhancement Through Computerized Training, is a randomized, unmasked, parallel-group behavioral study concerning alcohol use disorder. It planned working-memory training, inhibitory-control training, and bias-modification training, with a non-trained control receiving typical care. Total enrollment was 22. The study began on 2021-04-28, had a primary completion date of 2023-09-22, and is marked terminated.

The primary outcome was change in working-memory performance from baseline to post-test, up to 3 months. The registry record reports a mean change of 1.61 percent points and standard deviation of 0.07 for 10 non-trained control participants. It reports no analyzed primary outcome participants in the three training groups, citing early termination due to low enrollment. No comparative analysis or secondary outcome results are posted, so the record does not estimate a training effect.