Restoration of Cognitive Function With TDCS and Training in Serious Mental Illness: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Restoration of Cognitive Function With TDCS and Training in Serious Mental Illness (NCT03208036), a randomized na study. The primary outcome is Working Memory Capacity Composite Score.

2. Clinical Question

The comparison is designed to examine the incremental contribution of active stimulation when both groups receive cognitive remediation. It is not a comparison of cognitive remediation against no training, or tDCS against no treatment. The registry’s stated hypotheses are that pairing cognitive remediation with tDCS will be more efficacious than pairing it with sham stimulation, and that intervention-induced cognitive change will be sustainable. These are hypotheses, not findings.

The trial addresses a population described in the record as having serious mental illness, with diagnoses of schizophrenia, schizoaffective disorder, or bipolar disorder. The proposed treatment context is cognitive and functional impairment associated with serious mental illness. The endpoint focuses on working memory, one component of cognition, rather than functioning or symptoms as a whole. The record lists social functioning, other cognitive measures, and psychiatric symptom measures as secondary outcomes, but does not report their results. The registry record does not report a numerical primary-outcome contrast or its uncertainty, so the clinical question cannot be answered from the results information available in this record.

3. Trial Design

The registry describes a completed, randomized, parallel-group clinical trial with enrollment of 22. Participants were assigned in a 1:1 ratio to tDCS or sham stimulation, with both groups receiving working-memory-focused cognitive remediation. The study is listed as phase NA. The sponsor is the VA Office of Research and Development. The registry lists a start date of 2018-07-15 and a primary completion date of 2021-12-01.

The brief description characterizes the study as double-blind, double-baseline, and sham-controlled, while the registry’s masking classification is triple. These descriptions do not match exactly. The record does not specify which roles were masked or provide further detail to resolve the difference. The intervention is described as taking place in a small-group format. Training was planned to consist of 48 sessions, with 2–3 sessions scheduled in a week and each session lasting 2 hours. The description says 1 hour was spent on computer-based cognitive exercises requiring working-memory skills.

The primary outcome is the working memory capacity composite score at Baseline, 4 months, and 5.5 months. Its stated analysis population is participants who completed the intervention. The registry identifies schizophrenia, bipolar disorder, and schizoaffective disorder as conditions. It lists 10 outcome measures and 36 analyses, but does not report the statistical model or a prespecified primary contrast in the record.

The completed status indicates the trial is no longer listed as ongoing; it does not establish that the primary comparison was statistically conclusive or clinically important. Participant flow, numbers assessed at each time point, and reasons for non-completion are not reported in the registry record. The record also does not report a sample-size justification or explain how the enrollment target was selected.

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.

Baseline Assessment

Reasons for not completing this period:

2nd Baseline Assessment

Reasons for not completing this period:

Intervention

Reasons for not completing this period:

4-month Assessment

Analysis population for the primary outcome, as defined in the registry: Participants who completed the intervention.

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 Working Memory Capacity Composite Score, measured over Total score at Baseline, 4 months, and 5.5 months.

The primary endpoint is the Working Memory Capacity Composite Score, expressed as a proportion of correct responses. It averages performance from the Change Detection and Change Localization Tasks. The score ranges from 0.00 to 1.00, with higher scores indicating better performance. The registry describes it as a measure of one core component of working memory.

Working memory is relevant to the study’s focus because the cognitive remediation intervention specifically trains working-memory skills. A composite based on two tasks can represent performance across the task components named in the record, rather than relying on a single task alone. The registry does not report the rationale for combining the two task scores or details of how the average was constructed beyond the description given.

The outcome is recorded at Baseline, 4 months, and 5.5 months. Including later assessments is consistent with the stated interest in whether cognitive change is sustainable, but the record does not report a specific definition of sustained change or identify a single time point as the sole primary comparison. The stated primary population consists of participants who completed the intervention, which means the endpoint description does not by itself establish how participants with missing assessments were handled.

The measure is a performance outcome, not a direct measure of daily functioning, psychiatric symptoms, or overall clinical status. Although the registry lists those kinds of outcomes elsewhere, their results are not reported in the record. No primary endpoint group means, between-group estimate, confidence interval, or p-value are reported, so the magnitude and precision of any treatment contrast cannot be assessed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does randomization contribute?

Random assignment is intended to make the groups comparable, on average, with respect to factors that could affect the outcome. It helps support a causal interpretation of a between-group contrast, provided that assignment is implemented as planned and follow-up and analysis do not introduce important biases. In this trial, the record describes randomized allocation in a 1:1 ratio to tDCS or sham stimulation, while cognitive remediation is part of both arms. Randomization does not guarantee that all measured or unmeasured characteristics will be identical between groups, particularly in a small trial. Baseline characteristics and their group-wise balance are not reported in the registry record.

What would a between-group estimate mean?

A between-group estimate would quantify the difference in working-memory performance associated with assignment to tDCS rather than sham stimulation, in the setting where both groups receive cognitive remediation. For a continuous endpoint, an analysis might express the contrast in score units or as a difference in change over time. The registry record does not report the actual estimate or specify which form of contrast was used. It therefore does not support interpreting a numerical effect or judging its practical importance.

How should a confidence interval be read?

In general, a confidence interval describes uncertainty around an estimated effect under the assumptions of the statistical method. A narrower interval indicates greater precision, while a wider interval indicates less precision. Intervals can include values consistent with effects in more than one direction or with little difference. They do not, by themselves, establish whether an effect matters clinically. No confidence interval for the primary outcome is reported in the registry record, so the precision of the trial’s primary comparison cannot be assessed.

What role would a p-value play?

In general, a p-value measures how unusual the observed data would be under a specified null hypothesis and statistical model. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect. A p-value needs to be interpreted with the estimate, its uncertainty, the analysis plan, and the number of comparisons in mind. The registry record does not report a primary-outcome p-value or the details needed to identify the inferential test.

Why do repeated assessments matter?

The primary score is recorded at Baseline, 4 months, and 5.5 months. Repeated measurements can be analyzed in different ways, such as comparing follow-up values, comparing change from baseline, or modeling the pattern over time. Those approaches answer related but not identical questions and can handle within-person correlation differently. A model for repeated measurements may use information across time points, but its interpretation depends on the model specification and missing-data assumptions. The registry record does not report which approach was used.

What does the endpoint scale tell us?

The endpoint is a continuous measure defined as a proportion of correct responses, with a stated range from 0.00 to 1.00. The registry says higher values indicate better task performance. A mean summarizes the central tendency of a continuous outcome, while a standard deviation describes spread among observations around that mean. Reporting both can help readers understand typical performance and the variability of scores. Those summaries are not reported for the primary outcome in the registry information.

Why can over-dispersion matter?

In general, over-dispersion occurs when observed variability is greater than a statistical model expects. If a model assumes less variability than the data contain, its standard errors may be too small and its intervals or tests may be misleading. Methods that account for extra variability can improve inference when their assumptions fit the data. This trial’s primary endpoint is described as a continuous mean outcome, and the registry record does not report a model or an over-dispersion assessment. It would therefore be inappropriate to claim that over-dispersion was present or that a particular adjustment was used.

How do planning assumptions differ from trial results?

In general, sample-size calculations depend on assumptions about the effect to detect, variability, allocation, significance level, and statistical method. The record describes planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation, a normal approximation, and effects expressed as a stated fraction of the observed effect. For non-time-to-event outcomes, the illustrations are described as participants per arm; time-to-event illustrations use a total event count. These are planning calculations, not observed power and not evidence that the study had a particular probability of detecting an effect. The record does not report the trial’s own design justification, and the specific illustration values are not reported in the registry information used for this explanation.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 10 outcome measures and 36 analyses. Multiple outcomes, time points, and analyses create a multiplicity issue: when many statistical tests are conducted, the chance of at least one apparently notable result can increase. In general, a prespecified analysis plan can identify the principal comparison and distinguish confirmatory tests from exploratory ones. Adjustments for multiple testing are another possible approach, although the appropriate method depends on the questions and analysis structure. The registry record does not report a multiplicity strategy, identify a hierarchy of outcomes, or specify which of the 36 analyses should be treated as the primary inferential test.

The primary endpoint is measured at Baseline, 4 months, and 5.5 months, and the registry indicates that it has several categories or time points rather than reducing it to one comparison. The record does not report a single primary contrast, a procedure for comparing the time points, or a plan for interpreting multiple assessments. Without those details, readers cannot determine whether a reported test—if one were available—was selected in advance or how it relates to the broader set of analyses.

The stated primary population is participants who completed the intervention. This description does not explain how participants who did not complete the intervention were treated in the analysis, or whether all randomized participants were included. The registry record does not report numbers with outcome data at each time point, reasons for missing measurements, the method used to address missing data, or sensitivity analyses. Missing-data assumptions matter because results can depend on how unobserved outcomes are handled. No assumptions or procedures should be inferred from the completer-based population description alone.

The registry record also does not report a statistical analysis plan, the prespecified primary model, covariate adjustment, or any interim-analysis approach. Their absence from the record means these details are not reported in the registry record; it does not establish which procedures were or were not used outside the recorded information.

11. 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 0% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: hospitalization (11.1%).

The serious adverse event figures list 1 affected participant among 9 at risk in the tDCS group and 0 among 8 at risk in the sham group. The term reported as the top serious event is hospitalization, shown as 11.1%. No deaths are listed in either group. The record also lists 0 other adverse events among 9 at risk in the tDCS group and 0 among 8 at risk in the sham group.

These figures describe events recorded in the registry, not a complete estimate of comparative safety. The denominators are small, so a single event represents a substantial amount of the observed information. Counts alone do not establish whether an event was caused by an intervention, how severe or prolonged it was, or whether groups differed in underlying risk. The record does not report event timing, detailed event descriptions, or an attribution assessment beyond the listed categories.

The hospitalization term and the serious-event count should be read as safety reporting, not as an efficacy result. The registry record does not provide enough information to assess uncertainty around a between-group safety contrast. No conclusion about relative safety should be drawn from these figures alone.

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

The record describes a randomized comparison of active tDCS plus cognitive remediation against sham stimulation plus cognitive remediation, with working-memory capacity as the primary outcome. It identifies the score, its direction, the assessment schedule, and the population specified for the primary endpoint. This information clarifies what the trial was designed to study and what a primary result would need to address.

It does not report the primary outcome values or a treatment-effect estimate. Consequently, the direction and magnitude of any difference, its statistical uncertainty, and its practical importance are not reported in the registry record. The missing result cannot be interpreted as evidence that the groups were the same, nor as evidence that one intervention was superior.

Even with a reported score difference, interpretation would require attention to the analysis population, missing measurements, uncertainty, and the number of outcomes and analyses. A statistically detectable difference would not automatically establish a meaningful benefit, while an imprecise estimate would not establish that no effect exists. The registry information does not provide the results or analysis details needed for those judgments.

The trial’s two arms both include cognitive remediation. Any contrast between them would concern the incremental contribution of active stimulation relative to sham in that training context. It would not isolate the effect of cognitive remediation compared with no cognitive remediation, and it would not answer every question about cognitive or functional outcomes in serious mental illness. The registry lists additional social, cognitive, and symptom measures, but does not report their results.

Accordingly, the most defensible conclusion is limited: the registry describes the intervention comparison and working-memory endpoint, but does not report the primary efficacy comparison. The safety counts are also limited descriptive information and do not establish comparative risk. Further claims about effectiveness, durability, or safety are not supported by the information reported in the registry record.

13. Why This Trial Matters Statistically

This trial raises a useful statistical question about an add-on component within a combined intervention. Because both groups receive working-memory-focused cognitive remediation, randomization is intended to isolate the contrast between active and sham stimulation while holding the training context broadly in common. This makes the target question more specific than asking whether the full program differs from usual care or no intervention.

The endpoint also draws attention to the importance of defining a clinical outcome precisely. The score combines performance on Change Detection and Change Localization Tasks and is expressed as a proportion correct. Its scale and direction are stated, and assessments are scheduled over time. Those details help identify what an eventual estimate would mean, while also making clear that a working-memory score is not interchangeable with social functioning, psychiatric symptoms, or overall clinical status.

The repeated assessments create analytical choices. A comparison at a particular follow-up, a change-from-baseline analysis, and a model of the trajectory can target different estimands. With several outcomes and analyses listed, prespecification and multiplicity handling are important for interpreting statistical evidence. The registry record does not report those analysis details or a primary result, so the design’s statistical question is clearer than its evidential answer.

The record also illustrates the limits of drawing strong conclusions from a small enrolled group and sparse safety counts. Estimate precision depends on the observed data and analysis, not enrollment alone. Since the primary estimate and its uncertainty are not reported, readers cannot judge how well the data distinguish among plausible effects. The trial is statistically informative as a design example, but the available registry information does not establish the size or reliability of an outcome difference.

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 central limitation for interpretation is that no primary-outcome result is reported in the registry information. Group means, an effect estimate, uncertainty intervals, and a p-value are not available. Thus, the principal efficacy question remains unanswered by the recorded results. Secondary outcome results are also not reported in the information used for this explanation.

The registry lists enrollment of 22, but does not report how many participants completed the intervention, how many contributed outcome data at each assessment, or why any data may be missing. The primary endpoint is defined for participants who completed the intervention. Without participant-flow details or missing-data procedures, it is not possible to assess the effect of attrition or determine the analysis population used for any comparison.

The registry does not report the statistical model, the primary contrast, a multiplicity plan, covariate adjustment, or interim-analysis procedures. It also does not report the rationale for the enrollment target or the trial’s own sample-size justification. The planning assumptions described in the record do not substitute for a reported design justification or observed treatment-effect analysis.

There is an unresolved difference in the masking descriptions: the registry classifies masking as triple, while the brief description calls the trial double-blind. The record does not identify the masked roles or explain the difference. The reported serious adverse event information is limited to counts and a listed term, without timing, detailed descriptions, or a reported assessment of attribution. These limitations constrain interpretation but do not establish that any specific bias or analytic problem occurred.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Restoration of Cognitive Function With TDCS and Training in Serious Mental Illness — NCT03208036.
• Registry record last updated 2022-12-21; status last verified 2022-11.
• Results first posted 2022-12-21.
• 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

NCT03208036 is listed as a completed, randomized, parallel-group study sponsored by the VA Office of Research and Development. It enrolled 22 participants and compared tDCS with sham stimulation, with working-memory-focused cognitive remediation in both arms. The primary outcome is a working-memory capacity composite score based on two task types, measured at Baseline, 4 months, and 5.5 months. Higher scores indicate better performance.

The registry lists 10 outcome measures and 36 analyses. It describes the primary endpoint population as intervention completers, but does not report participant flow or primary-outcome results in the information available for this analysis. Serious adverse event figures list 1 affected participant in the tDCS group and 0 in the sham group; no deaths are listed. These counts are descriptive and do not establish a comparative safety conclusion.