Improving Cognitive Health in COVID-19 Survivors: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Improving Cognitive Health in COVID-19 Survivors (NCT04843930), a randomized phase 2 study. The primary outcome is Change in Cognitive Function, as Measured by Number Correct on the Digit Symbol Matching Task.

2. Clinical Question

3. Trial Design

This was a randomized, double-masked, parallel-group phase 2 trial sponsored by Weill Medical College of Cornell University. The registry lists an enrollment of 110 and a completed status. The recorded study start was 2021-07-13, and the primary completion date was 2023-02-06. These registry entries describe the design and timeline, but do not establish how well masking was maintained or how participants were allocated in practice.

The experimental group was assigned AKL-T01, a digital intervention delivered through an iPad application. The description specifies a 6-week intervention period and a target daily play schedule. Participants in the waitlist group did not receive an active control condition during the initial waitlist period; they could continue an existing cognitive intervention or receive no intervention. The registry also describes control participants who later started AKL-T01 and those who declined it. These categories matter when reading safety results, but the record does not give enough detail to reconstruct participant flow or determine the analysis population for each efficacy outcome.

The primary outcome was change in score on the Digit Symbol Matching Task between baseline and post-treatment at 6 weeks. The record identifies a least-squares mean and a 95% confidence interval as the reported statistic, but gives no estimate or interval. The statistical model used to obtain the least-squares mean, adjustment variables, handling of repeated measurements, and primary analysis population beyond the stated exclusions are not reported in the registry record. The randomization method, allocation concealment procedures, and specific masking roles are also not reported.

Eight secondary outcomes are named, covering self-reported daily functioning and several cognitive performance tasks. The registry lists 10 outcome measures and 10 analyses overall. These counts indicate that multiple outcomes were registered, but do not establish which results were analyzed together or whether the analysis plan changed.

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: Note that the final participant was consented and enrolled on 12/7/2022, but did not have their first study visit until 12/20/2022 due to scheduling constraints. The final 6-week primary endpoint for this participant was collected on 2/6/2023.

Randomization

Reasons for not completing this period:

Baseline Assessment

Week 6 Assessment

Reasons for not completing this period:

Week 10 Follow-Up

Analysis population for the primary outcome, as defined in the registry: Analysis of outcome measure excludes participants who did not complete this task / timepoint and performance-based outliers.

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 analysis includes the sample who initiated the trial and were included in the final analysis.

6. Primary Endpoint

The registry lists the primary outcome as Change in Cognitive Function, as Measured by Number Correct on the Digit Symbol Matching Task, measured over Baseline and Post Treatment (6 weeks).

The primary endpoint was change in cognitive function measured by the number of correct matches on the Digit Symbol Matching Task. Participants were asked to match pairs of shapes and numbers. The registry describes the task as a timed measure of attention and processing speed. The measurement was scheduled at baseline and post-treatment, with the post-treatment assessment at 6 weeks.

A task score is useful as a defined, repeatable measure of performance on a particular cognitive activity. It can make change in task performance more concrete than a broad description of cognitive functioning. The registry also reports a normative mean number correct of 65.79: scores above that reference mean indicate better task performance relative to the stated norm, while scores below it indicate poorer performance. That reference point is not a treatment comparison and does not show whether either trial group changed.

The recorded primary statistic is a least-squares mean, accompanied by a 95% confidence interval. In general, a least-squares mean is a model-based estimate of a group mean, often adjusted for variables included in the model. The specific model and any adjustment variables are not reported in the registry record, so the meaning of adjustment for this analysis cannot be specified further. The record states that participants missing the task or timepoint and performance-based outliers were excluded from the outcome analysis. The number excluded and the resulting analysis sample are not reported.

The primary endpoint is therefore clearly named, but its estimated change and comparison between groups are not reported in the registry record. Without those values, the direction, magnitude, and precision of the primary result cannot be assessed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registered design was randomized, double-masked, and parallel-group. In general, random allocation is intended to make the groups comparable, on average, with respect to measured and unmeasured characteristics at the start of a trial. Masking is intended to limit the influence of knowledge about assigned treatment on care, behavior, outcome assessment, or analysis. The registry does not report the randomization sequence method, allocation concealment, who was masked, or whether masking was successful. The design labels alone cannot resolve those details.

What does a least-squares mean represent?

A least-squares mean is a model-based group mean. In general, it can account for covariates or other structure included in the statistical model, unlike a simple arithmetic average that uses only the observed values. The primary outcome record names a least-squares mean and a 95% confidence interval, but does not report the model formula or its included variables. It is therefore not possible to determine which adjustments were used or how they affect interpretation. The estimate itself is not reported in the registry record.

How should the confidence interval be read?

In general, a confidence interval describes uncertainty in an estimate under the assumptions of the statistical procedure. It is useful to consider both the estimated effect and the range of values compatible with the data and model. The interval does not describe the range of individual participant outcomes, and it does not by itself establish clinical importance. For this trial, the registry names a 95% confidence interval but does not report its endpoints. Precision and compatibility with different effect directions therefore cannot be assessed.

Why does the randomized comparison matter?

The primary question concerns the contrast between groups, not simply whether scores changed in either group. In general, a group may change over time for reasons other than the intervention, including time, repeated testing, or concurrent activities. A randomized between-group comparison is intended to distinguish the intervention contrast from changes that may also occur in the comparator group. Here, the comparator was a waitlist without an active control condition during the initial waitlist period, and participants could continue an existing intervention or have no intervention. The record does not report the numerical between-group contrast.

What do sample-size planning assumptions mean?

The registry’s planning description refers to a sample-size illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. These are planning assumptions, not evidence that the trial had a particular probability of detecting an effect. The registry record does not report the numerical planning results or the trial’s own sample-size justification. Planning calculations are not observed power, and they cannot replace the actual effect estimate and its uncertainty.

Why can over-dispersion matter?

In general, over-dispersion means that the variability in observed data is greater than a model assumes. When a model understates variability, standard errors and confidence intervals may be too small, making results appear more precise than they are. The primary endpoint is continuous, and the registry record does not describe an over-dispersion assessment or a model for this outcome. This general issue should not be taken as evidence that over-dispersion occurred in this trial.

What is known about the analysis population?

The primary outcome description says that participants who did not complete the task or timepoint and performance-based outliers were excluded. In general, an analysis population based on completed assessments can differ from the population originally randomized. If missingness or exclusion is related to outcome performance or group assignment, the resulting estimate may be affected. The registry does not report the number excluded, details of the outlier rule, how missing values were handled, or analyses testing the sensitivity of the result to these choices. It also does not specify whether the main efficacy analysis followed an intention-to-treat principle.

The record names eight secondary outcomes, lists 10 outcome measures and 10 analyses, and does not report a multiplicity adjustment plan. In general, examining multiple outcomes can increase the chance of at least one apparently noteworthy result arising by chance if results are interpreted without accounting for the number of tests. Whether and how that concern was addressed is not reported in the registry record. Since outcome estimates and tests are not included in the information available for this analysis, their statistical interpretation cannot be reconstructed.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists the Digit Symbol Matching Task as the primary endpoint and names eight secondary outcomes. These include a daily-functioning measure and cognitive performance measures involving object tracking, digit span, reaction time, switching, sustained attention, and visual paired associates. It also reports 10 outcome measures and 10 analyses. The record does not state how the listed outcomes map to the total counts, whether all were tested in the same analysis population, or whether additional analyses were planned.

In general, testing multiple outcomes creates a multiplicity issue: the chance of at least one apparently unusual result can increase when many hypotheses are examined. Statistical plans may address this by designating a primary outcome, adjusting for multiple comparisons, or treating secondary results as exploratory. The registry record does not report whether any adjustment was used or how secondary findings were intended to be interpreted. The primary outcome result is also not reported, limiting interpretation of the overall efficacy evidence.

For the primary endpoint, the registry states that participants who did not complete the task or timepoint and performance-based outliers were excluded. It does not report the number of missing observations or exclusions, their reasons, the outlier definition, or any sensitivity analysis. It also does not report the intended handling of missing data or identify the efficacy analysis population beyond those exclusions. Whether the analysis retained participants according to randomized assignment is not reported in the registry record.

In general, missing-data methods and analysis populations matter because the conclusions depend on which participants contribute data and on assumptions about unavailable outcomes. The registry information does not permit evaluation of those assumptions or the robustness of the primary comparison to alternative choices.

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 non-serious adverse event terms

The registry lists serious adverse events for four named groups. For the AKL-T01 Intervention group, 0 of 55 participants at risk were recorded as affected; for the Waitlist Control group, 0 of 55 were recorded as affected. The registry also lists 0 of 46 among control participants who started AKL-T01 and 0 of 2 among control participants who declined it. Deaths were recorded as 0 in each of these groups. The record does not list serious adverse event terms.

For other adverse events, the registry records 5 of 55 participants affected in the AKL-T01 Intervention group and 2 of 55 in the Waitlist Control group. Among control participants who later started AKL-T01, 2 of 46 were recorded as affected; among those who declined it, 0 of 2 were recorded as affected. These are registry counts and should be read within the group definitions shown. The later-start and declined categories are subgroups of control participants, not additional randomized arms to combine with the original groups.

In general, a zero count in a registry record means that no affected participants were recorded for that group; it does not establish that an event is impossible. These figures do not include enough detail to describe event types, timing, severity beyond the serious/non-serious distinction, or attribution to intervention. Such details are not reported in the registry record.

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

The trial asks a focused question about change in performance on a timed cognitive task after a 6-week digital intervention compared with a waitlist control. The registry provides the intervention description, the primary endpoint definition, and selected safety counts. Those elements help identify what was studied and how the primary outcome was intended to be interpreted.

However, the primary result is not reported in the registry record. There is no reported between-group estimate, 95% confidence interval, or p-value for change in Digit Symbol Matching Task score. Consequently, the registry information does not establish whether the groups differed, the direction or size of any difference, or the precision of an estimated effect. It also does not allow assessment of whether any observed difference would be important beyond the task measure itself.

The stated normative mean of 65.79 provides a reference for task performance, but it is not a randomized treatment effect. A score relative to a norm and a difference in change between trial groups answer different questions. The registry also notes exclusions for incomplete assessments and performance-based outliers, while leaving the number excluded and the analysis approach unspecified. This further limits assessment of how the primary estimate would apply to all enrolled participants.

The safety record reports no serious adverse events or deaths in the listed groups and records other adverse events in the intervention, waitlist, and post-waitlist control categories. These counts do not determine efficacy and do not provide detailed event descriptions or attribution.

In general, absence of a posted efficacy estimate should not be mistaken for evidence of no effect, just as a trial design or sample-size planning assumption is not evidence that an effect was observed. The registry record supports a description of the question and methods, but not a quantitative conclusion about the primary cognitive outcome.

13. Why This Trial Matters Statistically

This trial is statistically notable for its combination of a defined performance-based cognitive endpoint and a waitlist comparison in COVID-19 survivors. The Digit Symbol Matching Task measures correct matches in a timed task, providing a specific outcome rather than an undefined judgment of cognitive improvement. The registry describes a 6-week digital intervention and assessments at baseline and post-treatment, establishing a clear time frame for the primary question.

At the same time, careful statistical interpretation depends on the reported estimate and uncertainty, not only on the endpoint name or design. The registry identifies a least-squares mean and 95% confidence interval as the primary summary but does not report their values. This omission prevents readers from evaluating magnitude, precision, or direction. It illustrates why transparent reporting of the estimate, interval, analysis sample, and model is central to interpreting a randomized trial.

The study also includes multiple secondary cognitive and functional outcomes. Such breadth can provide more than one perspective on performance, but general statistical practice requires clarity about outcome hierarchy and multiplicity. The registry does not report the multiplicity plan or detailed analysis procedures. Similarly, the stated exclusions for incomplete assessments and performance-based outliers make the analysis population important, while the number excluded and sensitivity analyses are not reported.

The sample-size description distinguishes planning illustrations from the original trial design. In general, that distinction matters: a calculation using an observed effect and stated assumptions is a planning exercise, not a measure of the probability that the completed trial detected an effect. The registry does not report numerical planning results or its own sample-size justification.

These features make the record useful for understanding the trial question and the reporting needed for interpretation, while leaving the main efficacy result unquantified.

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 main limitation for interpreting this trial from the registry record is that the primary outcome estimate and its 95% confidence interval are not reported. A p-value for the primary comparison is also not reported. Without these, the magnitude, direction, precision, and statistical compatibility of the primary result cannot be assessed.

The record states that participants who did not complete the task or timepoint and performance-based outliers were excluded from the primary outcome analysis. It does not report the numbers or reasons for those exclusions, the criteria used to identify outliers, or whether sensitivity analyses were conducted. The resulting analysis population and the possible effect of exclusions are therefore unclear.

Further statistical details are not reported in the registry record, including the specific model for the least-squares mean, adjustment variables, missing-data method, and whether the efficacy analysis followed randomized assignment. The randomization method, allocation concealment procedures, and masking roles are likewise not described. The presence of a double-masked design label does not establish how masking was implemented or maintained.

The waitlist control did not receive an active control condition during the initial 6-week period, and participants could continue a cognitive intervention or have no intervention. This comparator arrangement may be relevant to interpreting the contrast, but the registry does not describe how such concurrent activity was measured or handled in the analysis. The intervention description mentions weekly check-ins without further details.

Finally, the record lists multiple secondary outcomes and analyses but does not report a multiplicity strategy. Safety counts are reported without event terms or details of attribution. These limitations concern the registry information available for interpretation and do not establish that the trial investigators failed to conduct additional analyses.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Improving Cognitive Health in COVID-19 Survivors — NCT04843930.
• Registry record last updated 2024-02-29; status last verified 2024-02.
• Results first posted 2024-02-29.
• 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

NCT04843930, titled Improving Cognitive Health in COVID-19 Survivors, is recorded as a completed, randomized, double-masked, parallel-group phase 2 trial sponsored by Weill Medical College of Cornell University. The registry lists an enrollment of 110 and dates the study start to 2021-07-13, with primary completion on 2023-02-06.

The trial compared the remotely delivered AKL-T01 digital cognitive intervention with a waitlist control. The primary endpoint was change in number correct on the Digit Symbol Matching Task between baseline and post-treatment at 6 weeks. The registry names a least-squares mean and 95% confidence interval but does not report the primary estimate, interval, or p-value.

Eight secondary outcomes are named. The safety record lists no serious adverse events and no deaths in the reported groups, alongside counts for other adverse events. The primary analysis exclusions are described, but the analysis sample, missing-data handling, model details, and multiplicity plan are not reported in the registry record.