Computerized Cognitive Training for Diabetic Elderly Veterans: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Computerized Cognitive Training for Diabetic Elderly Veterans (NCT01736124), a randomized phase 4 study. The primary outcome is Diabetes Self-management.

2. Clinical Question

The registry describes a randomized, parallel-group comparison of the two approaches. Its broader research question was whether cognitive training could affect outcomes relevant to diabetes care as well as cognitive performance. Alongside diabetes self-management, the posted outcomes include medication adherence, memory, executive function and attention, hemoglobin A1c, systolic and diastolic blood pressure, and prospective memory. The primary endpoint focuses on self-management rather than cognition itself, so it addresses a practical behavioral outcome that may be relevant to diabetes care.

The registry record does not report a separate primary question about a particular subgroup, nor does it describe a prespecified threshold for a clinically important score difference. The posted comparison is therefore best read as an estimate of the difference between the randomized groups on the stated scale at the stated follow-up, rather than as a test of whether the intervention achieved a defined clinical target.

3. Trial Design

The registry identifies this as a Phase 4, randomized, triple-masked, parallel-group study. The reported enrollment is 199, and the study status is completed. The recorded start date is 2015-04-07 and the primary completion date is 2018-01-31. The sponsor is the VA Office of Research and Development.

Participants were assigned to adaptive computerized cognitive training or an active control. The training games were personalized to address cognitive deficits; the control games were intended to be engaging but were not designed to improve cognitive skills. The registry labels the masking as triple, but it does not report which groups of people were masked or how masking was maintained. The allocation method is described as randomized, while details such as sequence generation and concealment are not reported in the registry record.

The population description refers to non-demented older Veterans with type 2 diabetes who were considered at high risk for cognitive impairment. Recruitment was to take place through VA medical centers in Bronx and Ann Arbor. The registry does not report further eligibility criteria in the material summarized in the record, so characteristics such as the age threshold, diabetes duration, or specific cognitive screening criteria cannot be characterized from this information.

The primary outcome was measured six months after the intervention. The secondary outcomes include measures recorded at six months and medication adherence at 12 months post-intervention. Outcome-specific group sizes differ across these measures, so the number of observations contributing to one endpoint should not be assumed to apply to another. The registry record does not explain the reasons for those differences or specify the analysis population for each outcome. Although the study is marked completed and results are posted, the original statistical analysis plan and original design assumptions 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.

Before assignment: Review of demographic and medical records in the VA's Computerized Patient Record System, baseline assessment of diabetes self-management and cognitive functions, assessment of computer/internet availability, identification of a potential informant. 199 Veterans enrolled, but 19 were subsequently excluded prior to randomization

Baseline

Post-intervention Assessment

Reasons for not completing this period:

6 Mo. Post-Intervention Assessment

Reasons for not completing this period:

12 Mo. Post-Intervention Assessment

Reasons for not completing this period:

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: Participants included in data had evidence of less than optimal diabetes self-management

6. Primary Endpoint

The registry lists the primary outcome as Diabetes Self-management, measured over 6 months after the intervention.

The primary endpoint is Diabetes Self-Management Questionnaire score six months after the intervention. It is a continuous score on a scale from 0 to 20, with higher scores indicating better self-management. The registry states that the score has the same interpretation at baseline, allowing the scale to be understood consistently across those measurement occasions. The posted primary results are the group means and standard deviations at the six-month follow-up.

Self-management is a behavioral outcome: it captures diabetes-care practices rather than a laboratory measure or a measure of cognitive test performance. It is relevant to the study's broader question because cognitive difficulties could affect the ability to manage ongoing care, while the intervention itself is intended to train cognitive skills. The endpoint therefore asks whether the intervention groups differed on a diabetes-management measure after training.

The registry record does not report a minimum difference considered important to patients or clinicians, a threshold for a favorable response, or a primary-outcome analysis adjusted for baseline scores. Without such information, the score difference can be described statistically, but its clinical importance cannot be determined from the posted endpoint definition alone. The follow-up comparison also does not by itself establish whether any group difference arose from a change over time, because the posted primary summary is for the six-month measurement.

7. Results Posted in the Registry

• Adaptive CCT: mean 14.90 versus 15.10 on control; difference -0.20 (95% CI -1.10 to 0.70), standardized effect d = -0.08; z-test p = 0.662.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Medication Adherence

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Adaptive CCT: mean 0.11 versus 0.14 on control; difference -0.03 (95% CI -0.07 to 0.01), standardized effect d = -0.28; z-test p = 0.145.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Adaptive CCT and the control. The interval gives the range of effects that remain compatible with the data.

Memory

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Computerized Cognitive Training (CCT): mean 0.65 versus 0.46 on control; difference 0.19 (95% CI -0.06 to 0.44), standardized effect d = 0.29; z-test p = 0.128.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Computerized Cognitive Training (CCT) and the control. The interval gives the range of effects that remain compatible with the data.

Executive Function/Attention

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Adaptive CCT: mean 0.29 versus 0.33 on control; difference -0.04 (95% CI -0.25 to 0.18), standardized effect d = -0.07; z-test p = 0.721.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Adaptive CCT and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes extend beyond diabetes self-management to medication adherence, memory, executive function and attention, hemoglobin A1c, blood pressure, and prospective memory. They represent different domains and do not all measure the same underlying construct. Some ask about cognitive performance, some about health-related behavior, and others about clinical measurements. Taken together, they offer a broader description of the outcomes the study was designed to examine, but they should not be treated as interchangeable evidence.

The posted results show a mixed directional pattern rather than a uniform shift favoring adaptive training across all measured outcomes. The primary result likewise does not establish a clear group difference. The secondary results therefore do not provide a consistent basis for revising the primary interpretation. Their value is in describing possible patterns that may warrant careful consideration, not in establishing that the intervention changed every domain or that any one favorable estimate outweighs the primary finding.

The secondary comparisons are reported as unadjusted. When many outcomes are examined, the chance of seeing an apparently notable result by chance can increase. Multiplicity procedures are one way of controlling the interpretation of a family of tests, but the registry record does not report whether such a procedure was used. The secondary p-values should therefore carry less weight than a prespecified primary test, and should not be interpreted as conclusive evidence in isolation.

Differences in outcome-specific group sizes and measurement times also matter. Medication adherence is reported at 12 months post-intervention, while the cognitive outcomes are reported at six months after the intervention. A difference across these endpoints could reflect distinct underlying domains, different follow-up timing, or variation in the available observations. The record does not report enough information to distinguish those possibilities. The most defensible reading is descriptive: the secondary endpoints provide context, but do not overturn the uncertainty surrounding the primary comparison.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Medication Adherence
• Memory
• Executive Function/Attention
• Hemoglobin A1c
• Systolic Blood Pressure
• Diastolic Blood Pressure
• Prospective Memory

9. Statistical Interpretation

10. Statistical Methods Explained

What does a mean difference represent?

A mean is the average score among the observations contributing to a group summary. The reported primary comparison subtracts the active control mean from the adaptive cognitive training mean. A negative mean difference therefore indicates a lower average score in the adaptive training group. Because higher Diabetes Self-Management Questionnaire scores indicate better self-management, the sign can be translated into the direction of the outcome without changing the underlying estimate.

A mean difference stays in the original scale's units. That makes it directly interpretable against the score range, although the registry does not give a threshold for an important difference. The standardized effect size expresses a mean difference relative to the pooled variability, so it is unitless and can be useful for comparing outcomes measured on different scales. It is not a clinical threshold and should not be read as the proportion of participants who improved.

What does the standard deviation describe?

The standard deviation describes the spread of individual scores around their group mean. It is not the uncertainty in the mean itself and does not describe the range of every participant's score. Larger variation can make a group mean less precise when the group size is unchanged. The posted primary standard deviations are 2.4 for adaptive training and 2.5 for active control. They describe variability in scores, not variability in the treatment effect.

How should the confidence interval be read?

A confidence interval communicates uncertainty around an estimated group difference. In general statistical terms, an interval gives a range of effect sizes that are reasonably compatible with the data and the method used to construct it. A wide interval indicates less precision than a narrow interval. When an interval for a difference includes zero, the data do not rule out no group difference at the confidence level used to form that interval. This does not prove that the true difference is zero.

The registry posts an interval for the primary mean difference that includes zero. The unadjusted p-value is 0.662. A p-value describes how unusual data at least as incompatible with a specified no-difference model would be under that model and its assumptions. It is not the probability that the null hypothesis is true, nor does it quantify the probability that the intervention is beneficial. The registry does not report an adjusted p-value for this comparison.

What can be inferred from the posted calculations?

The comparison statistics are unadjusted recalculations from the posted group summaries. They do not incorporate baseline scores or other covariates. The registry record does not report a sponsor analysis for these posted outcomes, a model specification, or details of any adjustment. The discrepancy indicator is false, so no difference between the recalculated comparison and a sponsor analysis is flagged. That does not supply the missing analysis plan or establish that every modeling choice has been documented.

Unadjusted summary comparisons can provide a transparent description of the observed group means, but they do not address every design or data issue. For example, the registry record does not state whether outcome analyses were based on all randomized participants, how missing data were treated, or whether a baseline-adjusted analysis was performed. The calculated precision also depends on assumptions about the summary statistics and the comparison method. Statistical significance, if present, would not by itself establish clinical importance; conversely, an imprecise estimate does not establish that a meaningful effect is absent.

Why does variability matter?

In a continuous outcome, variability affects how well the group means distinguish an underlying difference from random fluctuation. Greater spread generally makes a mean difference harder to estimate precisely for a fixed amount of information. The reported standard deviations and group sizes therefore help describe the precision of the comparison. They do not, without a fuller analysis plan, reveal whether the score distribution was symmetric or whether unusual observations influenced the mean.

Over-dispersion is a general concern when observed variability is greater than a model assumes. In that situation, model-based standard errors can be too small and confidence intervals too narrow if the extra variation is ignored. This is a general statistical principle, not a finding reported for this trial. The registry presents continuous outcomes with means and standard deviations, but it does not report a diagnostic assessment of distributional assumptions or an alternative model for unusual variation.

What is the role of effect-size planning?

The planning illustrations use fractions of the observed effect and give the resulting participants per arm for specified power targets. These are not the trial's original design calculations. They show how required sample size can change when the effect to detect is smaller or larger, under the stated planning assumptions. Since observed effects are uncertain, using the observed estimate for planning can produce an illustration rather than a reliable estimate of the sample size needed for a future study. The original sample-size justification and its assumptions are not reported in the registry record.

The posted outcome summaries and these planning calculations answer different questions. The outcome summaries describe the trial's observed group data. Planning figures describe hypothetical future study sizes under a specified model and assumed effect. They should not be confused with evidence that the completed trial had a particular probability of detecting an effect.

11. Confidence Intervals and Precision

The primary mean difference is -0.2 points, with a confidence interval that extends from a negative value to a positive value and therefore includes zero. This means the data are compatible with no average difference as well as with differences in either direction. The unadjusted p-value of 0.662 is consistent with that uncertainty. The interval is more informative than the p-value alone because it conveys the range and direction of effects compatible with the estimate.

Precision is limited by the amount of outcome information and the variability in scores. The primary group summaries include 57 participants in adaptive training and 58 in active control, with standard deviations of 2.4 and 2.5, respectively. These are endpoint-specific observations; they should not be taken as the number contributing to every other outcome. The registry does not explain why the available numbers differ by endpoint.

The scale runs from 0 to 20, but a range alone does not establish how much difference matters. The registry record does not report a minimally important difference or another benchmark for judging the primary interval clinically. It is consequently possible to describe the estimate's direction and statistical uncertainty, but not to say from the registry alone whether every effect within the interval would be negligible or important.

The same caution applies to the secondary outcomes. Their precision depends on their own group sizes and variability, and their intervals should be interpreted separately rather than pooled into a single conclusion. A confidence interval does not account for missing-data bias, selection into the analyzed sample, or unreported multiplicity procedures. Those issues are distinct from sampling uncertainty and cannot be resolved by the interval alone.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm for 80% power and participants per arm for 90% power. They are not total participant counts or total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. These assumptions describe the planning illustration; the registry record does not report the original trial's sample-size assumptions or justification.

For the primary endpoint, planning for 100% of the observed effect gives 2,358 participants per arm for 80% power and 3,156 participants per arm for 90% power. Planning for 75% of the observed effect gives 4,192 participants per arm for 80% power and 5,611 participants per arm for 90% power. Planning for 50% of the observed effect gives 9,430 participants per arm for 80% power and 12,624 participants per arm for 90% power.

The 80% power column describes a planning scenario in which a study of the stated size would have an 80% probability of detecting the assumed effect under the calculation's assumptions and significance threshold. The 90% column uses a higher probability of detection and consequently calls for more participants under the same assumptions. The curve should be read as showing the general relationship between the assumed effect and the required sample size: smaller effects need more participants to distinguish them from random variation. These are planning calculations, not observed power for the completed trial.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.200; pooled standard deviation σ = 2.451.
2. n = 2 × (1.96 + 0.842)² × 2.451² / -0.200² = 2,357.5, rounded up to 2,358 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -0.199999999999999                      # observed difference in means
sigma <- 2.45095249680464                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(1736124)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 2,358 (80% power), 3,156 (90% power)
 75% of observed effect: n per arm = 4,192 (80% power), 5,611 (90% power)
 50% of observed effect: n per arm = 9,430 (80% power), 12,624 (90% power)
power.t.test: 2358.4 
Simulated power at n = 2,358 per arm: 0.802 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists several secondary outcomes in addition to the primary self-management endpoint. Testing multiple outcomes creates a multiplicity issue: the more comparisons considered, the greater the chance that at least one result may appear notable by chance alone. This is a general statistical concern, not evidence that any particular posted result is false. It affects how strongly unadjusted secondary findings should be interpreted.

The registry record does not report a multiplicity strategy, such as a hierarchy of hypotheses or an adjustment across outcomes. It also does not report an analysis plan describing which outcomes were confirmatory and which were exploratory. The posted secondary p-values are therefore best regarded as unadjusted evidence, not as results with documented family-wise error control. The record lists zero reported analyses, so a fuller inferential plan cannot be reconstructed from the registry entry.

Missing-data handling is also not reported in the registry record. The group sizes differ across outcomes, but the record does not state whether this reflects missing measurements, eligibility for particular analyses, or another reason. It does not specify imputation, complete-case analysis, or sensitivity analyses. Without that information, the direction or extent of any bias from unavailable outcomes cannot be assessed.

The analysis population is not reported in the registry record. In particular, it is not stated whether the primary analysis included all randomized participants according to assigned group, only participants with observed outcomes, or another population. The record also does not describe baseline adjustment, interim analyses, or the original statistical analysis plan. These omissions do not negate the posted descriptive summaries, but they limit assessment of whether the analyses followed prespecified methods and how robust the conclusions are to missingness or analytic choices.

15. 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: death (3.3%), surgery (2.2%), stroke (1.1%).

The registry reports serious adverse events for the computerized cognitive training and active control groups. The safety summaries list 90 at-risk participants in each group, with 4 serious adverse events in the training group and 6 in the control group. The registry also lists 0 other adverse events in each group. Deaths are listed as 3 in each group. These are descriptive registry counts and should be read in the context of the safety reporting categories used in the record.

The listed leading serious event terms are Death at 3.3%, Surgery at 2.2%, and Stroke at 1.1%. The record does not explain whether event terms overlap, how terms were selected for the leading list, or whether events were judged related to an intervention. The counts do not establish causation, and the summaries are not a substitute for an adjusted comparative safety analysis. The registry does not report such an analysis or a detailed account of the timing and clinical circumstances of these events.

Safety counts should also be kept distinct from the outcome-specific group sizes for efficacy measures. The registry reports the safety at-risk figures separately, and does not explain the relationship between those figures and the participants contributing to each efficacy endpoint.

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

The primary result is an estimate of the average difference in diabetes self-management scores between the randomized groups at the stated follow-up. The observed means were 14.9 for adaptive cognitive training and 15.1 for active control, and the estimated difference was -0.2. The confidence interval includes zero and the unadjusted p-value is 0.662. This is evidence that the posted data do not clearly distinguish the group means, not proof that the true effects are identical.

The result does not establish equivalence. An equivalence conclusion generally requires a prespecified margin defining how much difference would still count as practically similar, together with an analysis designed to assess that margin. The registry does not report such a margin or an equivalence analysis. Nor does the result establish that the intervention had no effect on any participant or outcome. It addresses the group average for the primary measure, with uncertainty.

It also does not establish whether a difference, if present, would matter in day-to-day diabetes care. The score has a defined range and direction, but the registry record does not report a minimally important difference. A statistical estimate can be interpreted in scale units; translating it into practical importance requires context that is not present in the record.

The secondary outcomes provide additional descriptive information but carry less inferential weight because they are multiple and unadjusted, and because their analysis procedures are not reported. Their varied directions do not support a single broad conclusion across cognitive and diabetes-related domains. Safety figures likewise describe reported events but do not establish that training caused or prevented an event. Taken as a whole, the record supports a cautious account of the observed results and their uncertainty, rather than a claim that the intervention works, does not work, is equivalent to the control, or is responsible for the listed safety outcomes.

17. Why This Trial Matters Statistically

This trial raises a useful statistical question about the distance between a plausible study rationale and the evidence that a particular endpoint can establish. Cognitive training is intended to affect cognitive skills, while the primary outcome is diabetes self-management. The primary comparison therefore tests a behavioral outcome that is relevant to the study's wider aim, but it is not itself a direct cognitive test. Keeping the endpoint and the intervention target distinct helps prevent a result in one domain from being misread as a result in another.

The active control is also important to interpretation. Participants in both groups engaged with computer games, but only the training games were designed to address personal cognitive deficits. This design can help separate the specific training approach from engagement with computer-based activities. The registry description does not provide enough detail to evaluate how closely the experience was matched between groups or which people were masked, so those features should not be assumed beyond the recorded design labels.

Statistically, the primary estimate is close to zero relative to the scale, but its interval includes effects in either direction. That distinction illustrates why an estimate, its precision, and clinical importance are separate questions. The result is not a demonstration of equivalence, and the registry does not provide a clinical-importance threshold with which to interpret the interval.

Finally, the outcomes span several domains and times, while the statistical plan, multiplicity approach, and missing-data methods are not reported. This makes transparent reporting especially important: readers can see the descriptive results, but cannot fully assess how the analysis was prespecified or how incomplete observations were handled. The study is statistically informative as a comparison with an active control, while the registry record leaves important interpretive details unresolved.

18. 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 specific to this trial follow from the information recorded. The population is described as non-demented older Veterans with type 2 diabetes at high risk for cognitive impairment, recruited through VA centers in Bronx and Ann Arbor. The registry record does not report the detailed eligibility rules or demographic characteristics of the analyzed groups. The results therefore cannot be assumed to apply to people outside the described population, and the record does not allow a detailed assessment of representativeness.

The primary result is a six-month group comparison. The posted endpoint summary does not show a baseline-adjusted estimate or a trajectory of change over time. The registry record does not report whether baseline scores were included in the analysis, how missing measurements were handled, or which analysis population was used. These omissions limit assessment of the robustness of the mean comparison and the potential for bias from unavailable outcomes.

The outcome-specific numbers differ, and the registry does not explain those differences. The control description and intervention description identify their intended distinction, but details about intervention exposure, adherence to the assigned activities, and masking procedures are not reported in the registry record. The triple-masked label alone does not identify who was masked or how masking was maintained.

The primary endpoint lacks a reported threshold for clinical importance. Secondary outcomes are numerous, and the record does not describe a multiplicity strategy or analysis plan. This makes secondary findings less conclusive and complicates interpretation of any isolated pattern. Finally, the serious adverse event summaries are descriptive; they do not include a reported causal assessment or detailed comparative safety model.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Computerized Cognitive Training for Diabetic Elderly Veterans — NCT01736124.
• Registry record last updated 2023-07-27; status last verified 2023-07.
• Results first posted 2021-02-23.
• 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.

22. Record Summary

The registry describes a completed, randomized, triple-masked, parallel-group Phase 4 study sponsored by the VA Office of Research and Development. It compared adaptive computerized cognitive training with an active computer-game control in non-demented older Veterans with type 2 diabetes. Recorded enrollment was 199.

The primary endpoint was Diabetes Self-Management Questionnaire score six months after the intervention. Mean scores were 14.9 with adaptive training and 15.1 with active control; the estimated mean difference was -0.2, with an unadjusted p-value of 0.662 and a confidence interval that includes zero. This does not establish a clear group difference or equivalence.

Secondary outcomes cover medication adherence, cognition, and clinical measures. Their results are unadjusted, and the registry record does not report multiplicity procedures, missing-data methods, or the analysis population. Serious adverse event counts are reported for both groups, but the record does not establish causal attribution.