This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Brain Stimulation and Cognitive Training (NCT03338673), a randomized na study. The primary outcome is Participant Retention in Combination Treatment.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03338673 |
| Phase | NA |
| Condition | Bipolar Disorder, Schizo Affective Disorder, Schizophrenia |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | None (open label) |
| Enrollment | 12 |
| Arms in the primary outcome | 2 |
| Lead sponsor | University of Michigan |
| Results posted | Yes |
2. Clinical Question
Population
The registry lists people with bipolar disorder, schizo affective disorder, or schizophrenia. The study description focuses on people with serious mental illness and memory deficits. Enrollment was 12. Further eligibility details and participant characteristics are not reported in the registry record.
Intervention
The combined treatment paired non-invasive brain stimulation, identified as tDCS, with computerized cognitive exercises using BrainHQ. The dual-therapy-first sequence included 10 stimulation sessions and 10 hours of exercises, followed by exercises alone. The registry describes stimulation as clinic-based and exercises as usable at home.
Comparator
The other sequence began with BrainHQ exercises alone and then switched to the combined treatment. Thus, the comparison concerns treatment order as well as treatment period: participants in both sequences were intended to receive both approaches. The trial was randomized, unmasked, and crossover.
Primary question
The primary endpoint counted participants who completed every session of the combined-treatment phase, with a stated timeframe of 4 weeks. The registry reports counts by randomized sequence: 4 participants in Dual Therapy First and 5 in Mono Therapy First. These counts describe completion; they do not by themselves establish that one treatment sequence improves memory or is more acceptable.
3. Trial Design
The registry describes a randomized, unmasked crossover study sponsored by the University of Michigan. The enrollment field reports 12 participants. The two assigned sequences were Dual Therapy First and Mono Therapy First, with 6 participants listed in each arm. In the first sequence, participants received tDCS alongside BrainHQ exercises before continuing with BrainHQ alone. In the other sequence, participants began with BrainHQ alone and then received the combined treatment.
The study description says that the treatment conditions switch after three weeks. It describes 10 clinic visits for brain stimulation over three weeks and 10 hours of computerized exercises over three weeks. Participants were also to complete paper-and-pencil assessments at the beginning, middle, and end of treatment. The overall study duration is described as about 7-8 weeks. These are features of the study description; the posted primary endpoint is specifically a count of participants completing every session of the combined phase, with a 4-week timeframe.
In a crossover design, each participant is intended to experience more than one treatment condition. That can make within-person comparisons useful, but the order of conditions and the timing of measurements matter. For this endpoint, the combined phase occurs at a different point in the sequence for the two arms. The registry record does not report a washout period, a period-specific endpoint analysis, or an adjustment for sequence or period effects. It also does not report whether every participant completed both phases.
The masking field is NONE, so the registry identifies the study as unmasked. The record lists the study status as COMPLETED, with start date 2018-10-11 and primary completion date 2019-09-27. A statistical analysis plan, allocation-concealment details, and further operational information 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
| Stage | Dual Therapy First | Mono Therapy First |
|---|---|---|
| STARTED | 6 | 6 |
| COMPLETED | 5 | 5 |
| NOT COMPLETED | 1 | 1 |
Reasons for not completing this period:
| Reason | Dual Therapy First | Mono Therapy First |
|---|---|---|
| Lost to Follow-up | 1 | 1 |
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.
| Characteristic | Unit | Dual Therapy First | Mono Therapy First | Total |
|---|---|---|---|---|
| Participants analysed | 6 | 6 | 12 | |
| Age, Continuous | years | 48.67 (SD 13.98) | 48.83 (SD 13.41) | 48.75 (SD 13.06) |
| Sex: Female, Male: Female | Participants | 5 | 5 | 10 |
| Sex: Female, Male: Male | Participants | 1 | 1 | 2 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Asian | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Black or African American | Participants | 3 | 1 | 4 |
| Race (NIH/OMB): White | Participants | 3 | 5 | 8 |
| Race (NIH/OMB): More than one race | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
6. Primary Endpoint
The registry lists the primary outcome as Participant Retention in Combination Treatment, measured over 4 weeks.
| Element | Registry entry |
|---|---|
| Result type | Count Of Participants |
| Measure of spread | not applicable |
| Unit | Participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint is titled Participant Retention in Combination Treatment. It is a count of participants completing every session of the combined phase of treatment, with a stated timeframe of 4 weeks. In the registry, the count is 4 for Dual Therapy First and 5 for Mono Therapy First. The endpoint therefore concerns completion of the combined treatment phase, rather than a change in working-memory performance.
Completion counts can be useful in a feasibility study because they describe whether participants reached the end of a treatment regimen. In general, completion may inform an assessment of treatment burden or acceptability, but it is not a direct measure of satisfaction, tolerability, cognitive benefit, or the reason a participant did or did not complete sessions. The registry description says that the study was designed to examine acceptability as well as a specific memory skill; this posted primary endpoint addresses completion only.
The count does not indicate the amount of cognitive change, and it cannot distinguish among reasons for incomplete treatment. The record does not report a definition of completion beyond completing every session, session-level attendance, participant-reported acceptability, or a separate primary cognitive-performance result. The endpoint can therefore support a descriptive account of combined-phase completion in the two assigned sequences, but it does not alone answer whether memory improved or whether the combined treatment was preferable.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry reports 4 participants completing every session of the combined phase in Dual Therapy First and 5 in Mono Therapy First. Each sequence has 6 participants listed in the arm record. These are raw counts, not estimates adjusted for the crossover structure. They describe completion within the assigned sequence and should not be read as a direct randomized comparison of the combined treatment against exercise alone: participants in both sequences were intended to receive both treatment conditions, but in a different order.
The record identifies the endpoint as a count and marks values as missing or not estimable for a two-arm binary comparison. It lists no posted analyses. Consequently, the record does not provide a treatment-effect estimate, confidence interval, or p-value for this endpoint. No inferential claim about a difference between sequences can be drawn from the counts alone. The counts also do not establish that any observed distinction is attributable to treatment order; timing, completion patterns, and other possible influences are not evaluated in a reported analysis.
A confidence interval, in general, expresses the uncertainty around an estimated quantity under a specified statistical method. Without a reported estimate and interval for a defined comparison, readers cannot assess the precision of a treatment contrast from the registry results. The absence of an inferential analysis is especially relevant because crossover data may require methods that account for paired observations, treatment period, and sequence. The registry does not report such a model or its assumptions.
The outcome is completion of the combined phase, not working-memory improvement. The study description mentions working memory as a motivating goal, but no primary cognitive score or secondary outcome result is posted in the record. The listed primary result therefore cannot be used to conclude whether cognitive performance changed. Nor can completion counts alone establish acceptability: they record whether every session was completed, not participants' reasons, preferences, or experiences.
The adverse-event record lists zero serious adverse events and zero other adverse events in each sequence, with zero deaths in each. Those are descriptive registry entries for the groups shown. They do not establish that no adverse effects occurred outside the record's reporting definitions or observation process, and they do not quantify uncertainty about uncommon events. The small enrollment and lack of a reported comparative safety analysis limit what can be inferred from these entries.
Overall, the posted results offer a narrow descriptive account of completion and adverse-event counts by sequence. They do not supply a statistical comparison, an estimate of cognitive benefit, or a complete account of treatment experience. Any interpretation should preserve that distinction between observed counts and conclusions that would require additional outcome data and an analysis suited to the crossover design.
9. Statistical Methods Explained
The registry records a binary endpoint expressed as a count of participants with an event, where the event is completing every session of the combined phase. The posted count is 4 in Dual Therapy First and 5 in Mono Therapy First. No measure of dispersion is listed, which is expected for a simple count, but the registry also marks a two-arm binary comparison as not estimable. No analysis is posted. The counts should therefore be treated as descriptive observations rather than a modeled treatment effect.
What does a count endpoint show?
A count states how many participants met a specified outcome definition. Here, the definition is complete attendance at every combined-phase session. A count alone does not describe the reasons other participants did not meet that definition. It also does not indicate whether participants completed most sessions, how completion varied over time, or whether missing records were treated as non-completion. Those details are not reported in the registry record.
For a binary outcome, an analysis may compare the chance of meeting the event definition across groups or conditions. In general, such a comparison needs a clearly defined denominator, a prespecified contrast, and a method that reflects the study design. The registry has counts and arm sizes, but marks the two-arm comparison as not estimable. No risk difference, relative measure, confidence interval, or p-value is reported. None should be inferred from the descriptive counts.
Why does the crossover design matter?
In general, a crossover study measures participants under more than one condition, so observations from the same person are related rather than independent. An analysis may use that pairing, while also considering the sequence in which treatments were received and the period when each outcome was measured. This differs from treating the assigned sequences as unrelated parallel groups. The primary completion count is organized by sequence, but the record does not describe a paired analysis or an analysis separating treatment effects from period and sequence effects.
The combined phase takes place first in Dual Therapy First and after exercise-only treatment in Mono Therapy First. That timing means that a difference in completion counts between sequences would not, by itself, isolate the effect of adding stimulation. It could also be associated with order, period, or other unmeasured factors. In general, crossover interpretation can depend on whether effects of an earlier period persist into a later period. Whether the study included a washout, or how it addressed carryover, is not reported in the registry record.
What is not reported about the statistical analysis?
The registry lists no analyses. It does not report a statistical model, a treatment contrast, adjustment variables, handling of incomplete outcome data, an analysis population, or assumptions used for inference. It also does not report whether a planned analysis was changed or whether interim analyses were conducted. These omissions mean that a reader cannot reproduce an inferential analysis or evaluate its assumptions from the registry record.
In general, a confidence interval communicates the uncertainty associated with an estimated effect, conditional on the analysis method and its assumptions. A p-value addresses compatibility between data and a specified null hypothesis under a model; it does not measure the size or importance of an effect. Neither type of result is posted for the primary endpoint. Their absence is not evidence of no effect; it means that the record does not report an inferential result for this comparison.
How should planning illustrations be understood?
General sample-size illustrations are planning tools, not trial findings. The registry's stated planning assumptions describe a two-sided significance level of 0.05, equal allocation, a normal approximation, and an effect defined as a stated fraction of the observed effect. Such an illustration does not measure the power actually achieved by this study. The record does not report a usable two-arm effect estimate for the binary endpoint, and no specific planning result is stated in the registry record. A planning calculation cannot replace the missing outcome analysis or turn these completion counts into an adjusted treatment estimate.
For general context, over-dispersion means that observed variability is greater than a statistical model expects. It can matter for outcomes such as counts when a model assumes less variability than the data exhibit, leading to uncertainty estimates that may be too narrow. This endpoint is reported as a simple participant count, and the registry does not report a dispersion estimate or a model requiring an over-dispersion adjustment. Thus, over-dispersion is a general methodological consideration, not a finding from this trial.
The record also lists two outcome measures but no posted secondary results. The named primary endpoint is completion of the combined phase. A complete statistical interpretation would ordinarily clarify which outcomes were designated for confirmatory testing and how multiple comparisons were handled. Neither a multiplicity strategy nor details of outcome selection are reported. In the absence of analysis details, the descriptive counts remain the most that can be stated about the primary endpoint from the registry results.
10. Exact Estimates for Each Arm
No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.
| Arm | Participants | Events | Observed % | Clopper-Pearson 95% CI | Wilson 95% CI | Bayesian 95% credible interval |
|---|---|---|---|---|---|---|
| Dual Therapy First | 6 | 4 | 66.7% | 22.3% to 95.7% | 30.0% to 90.3% | 29.0% to 90.1% |
| Mono Therapy First | 6 | 5 | 83.3% | 35.9% to 99.6% | 43.6% to 97.0% | 42.1% to 96.3% |
R code and output
arm <- c("Dual Therapy First", "Mono Therapy First")
n <- c(6, 6)
events <- c(4, 5)
for (i in seq_along(n)) {
ci <- binom.test(events[i], n[i])$conf.int # Clopper-Pearson exact interval
cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i]) # Bayesian, uniform Beta(1,1) prior
wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}
ARM Dual Therapy First 6 4 66.7 22.3 95.7 30.0 90.3 29.0 90.1
ARM Mono Therapy First 6 5 83.3 35.9 99.6 43.6 97.0 42.1 96.311. Multiplicity, Missing Data and Analysis Populations
The registry lists two outcome measures and identifies Participant Retention in Combination Treatment as the primary endpoint. It reports no posted secondary outcome results and lists no analyses. A statistical strategy for multiplicity—how the chance of false-positive findings across multiple outcomes or tests would be addressed—is not reported in the registry record. Because no inferential tests are posted, there is no reported family of p-values or adjustment procedure to interpret.
The primary endpoint result is reported as counts by sequence. The record does not state how participants with incomplete or unavailable endpoint data were handled, whether missingness occurred, or whether missing outcome records were distinguished from failure to complete treatment. Reasons for non-completion are not reported. The count definition requires completion of every combined-phase session, but does not describe a separate missing-data rule.
An analysis population, such as an intention-to-treat or per-protocol population, is not reported in the registry record. The arm counts identify the groups and their listed sizes, but do not establish which participants entered any particular analysis or whether all randomized participants contributed data. The record also does not describe exclusions, protocol deviations, or sensitivity analyses.
These gaps matter because missing outcomes and analysis populations can affect how completion counts are interpreted. In a crossover design, missing data may also differ by period or sequence. Without details on missingness and the analysis plan, no assumption about the handling of incomplete records should be made. The posted counts are best read as registry-reported descriptive results, not as a complete account of a prespecified statistical analysis.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Dual Therapy First | 0 | 6 | 0.0% |
| Mono Therapy First | 0 | 6 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Dual Therapy First | 0 of 6 |
| Mono Therapy First | 0 of 6 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Dual Therapy First | 0 of 6 |
| Mono Therapy First | 0 of 6 |
The adverse-event record lists 6 participants at risk for serious adverse events in each sequence and reports zero serious adverse events in both. It also lists 6 at risk for other adverse events in each sequence and reports zero other adverse events in both. Deaths are reported as zero in each sequence. No serious adverse-event terms are listed.
These figures describe what is recorded for the two assigned sequences. They are not a statistical comparison of safety between treatment conditions, and the registry does not report a comparative analysis. Because this is a crossover study in which participants were intended to receive both conditions, sequence-level group entries should not be mistaken for a complete, period-specific account of risk under each treatment.
In general, observing zero recorded events does not prove that an event is impossible or that risk is absent. A small study may provide limited information about uncommon events, and interpretation also depends on how events were defined, collected, and attributed. Those reporting details are not reported in the registry record. The entries support only the statement that the posted record lists no serious or other adverse events and no deaths in these groups.
13. What the Result Does, and Does Not, Mean
The result shows that the registry recorded completion of every combined-phase session for 4 participants in Dual Therapy First and 5 in Mono Therapy First. This is a descriptive finding about completion by sequence. It does not estimate the effect of adding brain stimulation to cognitive exercises, because both sequences were intended to receive the combined treatment and the registry does not report an estimable comparison.
The endpoint is not a memory score. Although the study description identifies working memory as a central motivation, the primary result posted concerns completion of treatment. The registry reports no secondary outcome results, so it does not establish whether working-memory performance changed, whether any change differed between conditions, or how precise such an estimate would be.
The counts also do not explain why every session was or was not completed. Completion can be relevant to feasibility, but it is not a full measure of acceptability or treatment burden. The record does not report participant preferences, reasons for missed sessions, or a separate satisfaction measure. No conclusion about participants' experiences should be inferred from the count alone.
Likewise, the zero adverse-event entries are registry-reported counts, not proof that risks are absent. They offer limited descriptive information and are not a substitute for a detailed safety analysis. The record does not report event definitions or a period-specific comparison of the two treatment conditions.
The study therefore contributes a narrow account of treatment completion in a small randomized crossover trial. It cannot, on the posted results, establish cognitive benefit, comparative efficacy, or a definitive safety profile. The missing inferential analysis and the lack of posted cognitive results are central to interpreting what the findings do and do not support.
14. Why This Trial Matters Statistically
This trial illustrates why a result must be interpreted in light of both its endpoint and its design. The study description concerns brain stimulation combined with cognitive exercises and working memory, but the posted primary endpoint counts completion of the combined-treatment phase. Keeping those questions separate prevents a feasibility outcome from being mistaken for evidence of cognitive change.
The crossover structure also matters. Participants were assigned to different treatment orders, and both sequences were intended to receive both conditions. A simple comparison of sequence-level counts is not equivalent to a direct, period-aware comparison of treatments. The record's indication that a two-arm binary comparison is not estimable is therefore an important limit, not a detail that can be overcome by interpreting the raw counts as an effect estimate.
The trial also shows the importance of transparent statistical reporting. No analysis is posted, and the registry does not report an analysis population, missing-data approach, model, or multiplicity strategy. Without those elements, readers cannot assess the precision of a contrast or determine how the crossover structure was incorporated. A clear descriptive result can still be useful, but it answers a narrower question than a model-based treatment comparison.
For clinical researchers and biostatisticians, the central lesson is to match claims to the information actually reported. Completion counts can describe participation in a treatment phase; they cannot establish a memory benefit or explain adherence. Reporting both the intended design and the limits of the available analysis helps readers distinguish feasibility observations from comparative evidence.
15. 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 enrollment was 12, and the posted primary outcome is a small set of descriptive counts. The record does not provide an estimable two-arm comparison or an inferential analysis. As a result, the precision of any treatment-sequence contrast cannot be evaluated from the reported results. The study's modest enrollment also limits the information available about uncommon adverse events.
The crossover design creates interpretive considerations that the registry results do not resolve. The combined phase occurs at a different point in the two sequences, while the endpoint is reported by sequence. A washout period, treatment-period analysis, carryover assessment, and sequence-adjusted model are not reported. The registry also does not establish whether all participants completed both treatment conditions.
The primary endpoint measures completion rather than cognitive performance, and no secondary results are posted. The record does not report reasons for non-completion, session-level attendance, participant-reported acceptability, or detailed treatment-experience outcomes. Thus, it cannot explain the completion counts or provide a full assessment of feasibility.
Further limitations include the absence of a reported statistical analysis plan, analysis population, missing-data method, multiplicity strategy, and outcome-model details. The registry lists bipolar disorder, schizo affective disorder, and schizophrenia, but further participant characteristics and eligibility details are not reported. These limitations restrict interpretation of both generalizability and the strength of any conclusions beyond the descriptive results.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Brain Stimulation and Cognitive Training — NCT03338673.
- Registry record last updated 2020-09-02; status last verified 2020-08.
- Results first posted 2020-09-02.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
NCT03338673, titled Brain Stimulation and Cognitive Training, is listed as a completed randomized, unmasked crossover study sponsored by the University of Michigan. Its official title describes neuromodulation plus cognitive training to improve working memory among individuals with serious mental illness. Enrollment is reported as 12, with 6 participants listed in each treatment sequence.
The study description outlines tDCS combined with BrainHQ exercises and a switch between combined treatment and exercises alone. The primary endpoint is the count completing every session of the combined phase, with a stated timeframe of 4 weeks. The registry reports counts of 4 for Dual Therapy First and 5 for Mono Therapy First, but no estimable two-arm comparison or posted analysis.
The record lists zero serious adverse events, zero other adverse events, and zero deaths in each sequence. It does not report secondary outcome results. The listed study start date is 2018-10-11, and the primary completion date is 2019-09-27.