COVID-19, Chronic Adaptation and Response to Exercise (COVID-CARE): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for COVID-19, Chronic Adaptation and Response to Exercise (COVID-CARE) (NCT04595773), a randomized na study. The primary outcome is Change in 6-minute Walk Test Distance (6MWD).

2. Clinical Question

The six-minute walk test is used as a measure of physical capacity. It asks participants to walk as far as possible around a measured course during six minutes. The registry states that greater distance represents better physical capacity, but the outcome analyzed is the change in distance rather than the post-intervention distance alone.

The trial is randomized and has a crossover model. The first-period comparison is useful for interpreting the reported exercise-versus-education result because the control group later crosses over to exercise. The registry record does not report enough detail to establish how the crossover period was incorporated into this primary comparison or what other outcomes may have been analyzed.

3. Trial Design

The registry describes COVID-CARE as a randomized controlled trial with single masking and a crossover model. The listed enrollment is 48. The registry status is active, not recruiting. The listed primary completion date is 2025-05-05. These registry entries describe the study record; they do not by themselves establish that every planned assessment or follow-up has been completed.

Two groups are described. The Aerobic Exercise Intervention group performed aerobic exercise three times a week and received weekly education for 10 weeks. The Control Group received weekly education for 10 weeks and then crossed over to aerobic exercise three times a week for 10 weeks. The registry does not report the identity of the person or people masked, so the meaning of single masking cannot be specified further.

The primary endpoint is defined for participants who completed the first 10 weeks. The posted group summaries include 20 participants in the aerobic exercise group and 20 in the control group. Their mean changes in six-minute walk distance were reported as 57.7 meters and 18.4 meters, respectively. These outcome counts are not the same as the listed enrollment, and the registry record does not explain the difference.

Participants were assessed using a range of procedures described in the registry, including heart and lung function tests, blood and urine samples, movement and balance tests, surveys, and activity monitoring. The primary result is specifically the change in six-minute walk distance. The registry lists one outcome measure and does not list secondary outcomes in the posted results.

The record does not report the trial’s original sample-size justification, detailed randomization procedures, masking procedures, or a full statistical analysis plan. In particular, the registry does not explain the planned handling of the crossover data, the choice of analysis population beyond the stated completion criterion, or the approach to missing outcome measurements.

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: 48 participants who consented to the study * 40 participants were randomized to study interventions * 6 participants were screen failure * One participant withdrew before randomization * One participant withdrew after randomization and prior to start of intervention

Period 1 - Week 1-10

Period 2 - Week 11- 20

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All participants who completed the first 10 weeks of the study 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 Change in 6-minute Walk Test Distance (6MWD), measured over Before and after 10 weeks.

The primary endpoint is change in six-minute walk test distance, abbreviated 6MWD, measured in meters. Participants walk around a pre-measured course for six minutes, and the total distance is recorded. The registry describes this test as a measure of physical capacity and states that a greater distance indicates better physical capacity.

The reported outcome is not simply the distance at the end of treatment. It is the difference between the measurement after the first 10 weeks and the measurement before those weeks: post minus pre. A positive change therefore describes an increase in walking distance over that interval. The registry reports the endpoint as a mean and standard deviation for each group.

Using a functional test as an endpoint can provide a standardized way to describe performance over a defined task. In general, an endpoint of this kind can be easier to interpret than a collection of separate symptoms, but it captures only the specific capacity tested. It does not represent every aspect of recovery, daily function, or health. The registry also describes surveys and other assessments, but no secondary outcome results are posted in the record.

The outcome population consists of participants who completed the first 10 weeks of the study intervention. The registry does not report whether participants who did not complete that period were included in another analysis, or how missing pre- or post-intervention measurements were handled.

7. Results Posted in the Registry

• Aerobic Exercise Intervention (AET+): mean 57.70 versus 18.40 on control; difference 39.30 (95% CI 16.72 to 61.88), standardized effect d = 1.08; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean change represent?

The registry reports a continuous outcome: each participant’s change in walking distance is measured in meters, and the group result is summarized by a mean. The mean is the arithmetic average of the measured changes within a group. Here, the aerobic exercise group’s reported mean change was 57.7 meters, while the control group’s was 18.4 meters. These are changes from before to after the first 10 weeks, not endpoint values at a single visit.

A group mean is a compact summary, not a description of every participant. The standard deviations reported for the two groups, 44.67 meters and 25.67 meters, describe how spread out the participant changes were around their respective group means. A larger standard deviation indicates more variation in the observed changes. It does not tell which participants changed more or explain the reasons for differences between participants.

How is the between-group difference interpreted?

The comparison focuses on the difference between the group mean changes. The reported mean difference is 39.3 meters, with the exercise group’s mean change above the control group’s. This comparison is more informative about the contrast between the assigned groups than reading either group’s change in isolation. A within-group increase alone cannot show whether the change differed from what occurred in the comparison group over the same period.

The registry also reports a pooled standard deviation of 36.43 meters and a standardized effect size of 1.079. In general, a standardized effect places a difference on a scale relative to variability, which can help compare effects measured in different units. It is not a measure of the outcome in meters and should not replace the original-unit estimate when describing the walking-distance result.

What does the confidence interval add?

The 95% confidence interval for the mean difference runs from 16.72 to 61.88 meters. In general, a confidence interval combines an estimate with a range reflecting statistical uncertainty under the method’s assumptions. If the same sampling process and interval method were repeated many times, the stated confidence procedure would capture the underlying parameter at its nominal rate. The interval is not a statement that there is a 95% probability that this particular fixed difference lies within these endpoints.

The interval describes uncertainty in the estimated group mean difference; it is not an interval containing 95% of individual participant results. Its width is affected by sample size and variability, among other factors. The registry record does not report a prespecified clinically important threshold, so the interval cannot be classified against such a threshold using the posted information alone.

How should the p-value be read?

The posted unadjusted p-value is <0.001. In general, a p-value is calculated under a null hypothesis and a statistical model. It describes how unusual data at least as incompatible with that null would be under the model, if the null and model assumptions held. It is not the probability that the observed difference arose by chance, nor is it a measure of the size or practical importance of the difference.

The registry labels this result as unadjusted. The record does not report a multiplicity adjustment or enough detail to reconstruct every aspect of the test. Because only one primary outcome is listed, the number of planned or exploratory statistical tests cannot be established from the posted outcome count alone. The result should be understood as the unadjusted comparison reported for the primary outcome.

Why can variability affect inference?

General explanation: when participant outcomes vary widely, the group mean difference is typically estimated less precisely for a given amount of data. The reported standard deviations show that participant changes were not identical within either group. A standard error summarizes uncertainty in an estimated mean or contrast, while a confidence interval uses that uncertainty to express a plausible range under the selected method. The registry reports standard errors of 5.74 and 9.989 for the control and exercise group means, respectively.

General explanation: some continuous outcomes can have more variability than a simple normal model captures, a feature often called over-dispersion. If over-dispersion is present, methods that assume less variability can produce uncertainty estimates that are too small. The registry record does not report an over-dispersion assessment or indicate that over-dispersion was identified for this endpoint; this is a general consideration, not a finding about this trial.

What is and is not reported about the analysis?

The primary outcome is a mean change, and the posted comparison gives a mean difference, interval, standardized effect, and unadjusted p-value. The comparison is based on the first 10 weeks, when one group received aerobic exercise and the other received education. The crossover design means that the control group later received exercise, but the registry record does not describe how later measurements were incorporated into a model or whether period and sequence were analyzed.

The record does not report the statistical model, assumptions checked, covariates, handling of missing observations, or an analysis population beyond participants who completed the first 10 weeks. It also does not report whether the analysis was prespecified or whether any alternative analyses were performed. These details affect how fully the reported result can be independently evaluated; they should not be inferred from the numerical summary alone.

10. Confidence Intervals and Precision

The reported mean difference in change in six-minute walk distance is 39.3 meters, and its 95% confidence interval extends from 16.72 to 61.88 meters. The interval gives a measure of the precision of the estimated average contrast. It is centered around the reported estimate, but it still allows a range of possible values consistent with the data and the interval method.

Precision is distinct from statistical significance. The interval does not include zero, and the reported unadjusted p-value is <0.001, but neither fact tells whether the estimated difference is important to participants. The registry record does not report a threshold for clinical importance, so the interval cannot be interpreted against such a benchmark.

The interval also does not describe the spread of individual responses. The group standard deviations, 44.67 meters for exercise and 25.67 meters for control, describe participant-level variation in change. The interval instead concerns the group-level mean difference. It should not be read as the range within which most individuals’ changes fall.

The registry reports 20 participants in each group for the primary outcome. The listed enrollment is 48, while the primary outcome population is restricted to those who completed the first 10 weeks. The registry record does not explain the difference between enrollment and the posted outcome counts, or how missing measurements or non-completion affected the estimate. That limits assessment of precision and potential selection effects beyond the interval itself.

In general, confidence intervals depend on the analysis method and its assumptions. The record does not give enough detail to evaluate all those assumptions independently. The reported interval is therefore the best available summary of the statistical uncertainty in the posted comparison, rather than a guarantee about the range of results that would occur in another study.

11. 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 illustrations are participant counts per arm, not total event counts. They are based on a planning calculation rather than the trial’s original design justification. The registry record does not report the assumptions used for the original design or its sample-size justification.

The listed 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 set to a stated fraction of the observed effect. The rows therefore explore planning for 100%, 75%, and 50% of the observed effect, rather than asserting that any of those effects is certain to occur in a future study.

For planning for 100% of the observed effect, the figures are 14 participants per arm for 80% power and 19 participants per arm for 90% power. For planning for 75% of the observed effect, they are 24 participants per arm for 80% power and 33 participants per arm for 90% power. For planning for 50% of the observed effect, they are 54 participants per arm for 80% power and 73 participants per arm for 90% power.

In plain terms, the 80% power column gives the estimated participants per arm under which the calculation would have an 80% chance of detecting the chosen effect at the stated significance level, if the assumptions held. The 90% column describes the corresponding planning target with a higher chance of detection under those assumptions. Smaller effects need more participants because they are harder to distinguish from random variation.

These figures are planning illustrations, not observed power. They use the effect estimated in this trial, which may be imprecise and may not recur at the same size in another study. The curve should be read as a visual representation of how the planning requirement changes with the effect assumed; it does not add a separate observed trial result.

12. 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 = 39.300; pooled standard deviation σ = 36.430.
2. n = 2 × (1.96 + 0.842)² × 36.430² / 39.300² = 13.5, rounded up to 14 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 <- 39.3                      # observed difference in means
sigma <- 36.430466645378                     # 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(4595773)
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 = 14 (80% power), 19 (90% power)
 75% of observed effect: n per arm = 24 (80% power), 33 (90% power)
 50% of observed effect: n per arm = 54 (80% power), 73 (90% power)
power.t.test: 14.5 
Simulated power at n = 14 per arm: 0.785 (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.

13. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several statistical questions are tested. The more tests performed, the greater the chance that at least one result will appear statistically notable even if none of the underlying effects is real. General approaches to multiplicity may adjust significance thresholds or otherwise account for a family of tests. The registry record does not report a multiplicity plan or adjustment for the posted primary comparison.

The registry lists one outcome measure and no secondary outcomes in the posted results. That does not establish how many analyses were originally planned, whether exploratory analyses were conducted, or whether outcomes were changed. The record does not report a full analysis plan, so the scope of any testing family cannot be determined. The p-value is identified as unadjusted and should be read accordingly.

Missing data are another concern because the primary outcome population is defined as all participants who completed the first 10 weeks. The registry does not report how many randomized participants lacked measurements, why any data were missing, or how missing values were handled. It also does not report whether an intention-to-treat analysis was performed. The posted mean changes therefore describe the stated completion population, and the available information is insufficient to determine how results might differ under another analysis population or missing-data approach.

Analysis populations are important because estimates can change depending on whether a study includes all randomized participants, only those who received an intervention, or only those with completed assessments. The registry’s wording identifies completers of the first 10 weeks for the primary outcome, but does not provide further detail about eligibility for the reported comparison. A reader should not assume an analysis approach that the record does not describe.

The crossover design raises additional questions about the role of period, treatment sequence, and possible carryover when data from both periods are analyzed. The registry describes the control group’s later switch to exercise but does not report the methods used to account for these features. The posted first-period result is interpretable as the reported comparison for that interval; the record does not establish how later-period data were handled.

14. 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

For serious adverse events, the registry reports zero events among 20 participants at risk in the Aerobic Exercise Intervention group, zero among 20 at risk during the Control Group education period, and zero among 20 at risk during the Control Group exercise period. Deaths are reported as zero in each of these groups. These are the posted counts for the listed groups and periods.

A zero count means that no serious adverse events were recorded in that group in the registry results. It does not establish that the risk is absent, particularly when the number of participants at risk is limited. The registry record does not list serious adverse event terms, and it does not provide enough detail to assess duration of follow-up or ascertainment methods.

The registry also reports 16 participants with other adverse events in the aerobic exercise group, 11 in the control education period, and 11 in the control exercise period. These are separate from the serious-event counts and should not be conflated with them. The record does not provide event terms in the posted safety summary. Counts should be read as registry reporting, not as evidence that groups had identical safety profiles or that no unreported events occurred.

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

The posted result shows a greater average increase in six-minute walk distance during the first 10 weeks in the aerobic exercise group than in the education control group. The estimated difference in mean change was 39.3 meters, with a 95% confidence interval from 16.72 to 61.88 meters. The outcome was measured among participants who completed the first 10 weeks, and the reported p-value was unadjusted.

This result addresses one defined functional measure. It does not summarize every symptom or daily activity, even though the registry describes additional tests and surveys. Nor does the mean difference describe what happened to every participant. The reported standard deviations indicate variation around group means, and the record does not provide individual-level outcome distributions.

The result should not be interpreted as an estimate for every person recovering from COVID-19. Eligibility was limited to adults ages 18-80 with laboratory-confirmed infection and ongoing symptoms, and the primary outcome population was limited to first-period completers. The registry does not provide enough detail to establish how representative those participants were of people outside the study or of those who did not complete the intervention period.

A statistically notable result does not by itself define clinical importance. The registry does not report a threshold for a meaningful change in walking distance, and the posted confidence interval cannot answer that question without such a reference. The result also does not establish that the same effect would occur in a different setting, under a different exercise program, or with a different comparison.

The crossover structure means the control group later received exercise. The reported primary comparison concerns the first 10 weeks, and the registry does not explain the analysis of the later period in enough detail to make broader claims about the crossover sequence. The result is best understood as the posted first-period comparison, with its stated outcome population and uncertainty, rather than as a conclusion about every possible outcome or analysis.

16. Why This Trial Matters Statistically

This trial’s statistical question is a comparison of changes, rather than a comparison of post-treatment scores alone. That distinction matters because participants may begin at different levels. Examining change from before to after the first 10 weeks focuses the reported endpoint on how much the measure shifted over the intervention period. The between-group contrast then asks whether those average shifts differed.

The outcome is expressed in meters, which keeps the estimated contrast tied to the walking test. The reported standardized effect size adds a scale-free summary relative to variability, but it does not replace the original-unit estimate. Looking at both the mean difference and the standard deviations helps readers distinguish the estimated group contrast from variation among participants.

Precision is also central. The confidence interval communicates uncertainty around the mean difference, while the p-value addresses compatibility with a null hypothesis under a statistical model. Neither statistic answers whether the difference matters to participants. The absence of a registry-reported clinically important threshold is therefore important when considering the practical meaning of the estimate.

The study’s crossover structure makes timing relevant to interpretation. The control group received education first and crossed over to exercise later, while the primary outcome is defined after the first 10 weeks. This makes the first-period comparison distinct from any analysis that might combine both periods. The registry record does not report the detailed analysis of the crossover data, so that distinction should remain explicit.

Finally, the planning illustrations show how sample-size requirements depend on the effect chosen for a future study. Smaller assumed effects require larger participant counts under the stated planning assumptions. Because the calculations use the observed effect, they are planning aids rather than evidence about the probability that this trial’s result would be significant again.

17. 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 reported outcome population includes participants who completed the first 10 weeks, while the listed enrollment is 48 and the posted group counts are 20 per arm. The registry record does not explain why these counts differ or how participants without complete outcome data were treated. This restricts assessment of possible effects of non-completion or missing measurements.

The registry describes randomization, single masking, and a crossover model, but does not give detailed methods for randomization or masking. It also does not identify who was masked. The record does not report how the later crossover period was analyzed, including whether period, sequence, or carryover were considered. The posted primary result should therefore be read as the stated first-10-week comparison, without assuming a particular model for the full study.

The statistical model, covariate adjustments, assumptions, and procedures for handling missing data are not reported in the registry record. Nor is the trial’s original sample-size justification reported. The planning illustrations are based on the observed effect and should not be mistaken for the assumptions that governed the original design.

The registry lists one outcome measure and no secondary outcomes in the posted results. It does not report a full statistical analysis plan, multiplicity procedures, or whether the analysis was prespecified. This limits evaluation of how the reported comparison fits within all analyses that may have been planned.

The primary endpoint is a walking-distance measure. Although the registry describes additional assessments, the posted primary result does not establish effects on every symptom, daily activity, or other measure. The registry record also does not report a clinically important threshold for the endpoint. These limits concern interpretation of the available result, not evidence that the study did or did not assess other aspects of recovery.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: COVID-19, Chronic Adaptation and Response to Exercise (COVID-CARE) — NCT04595773.
• Registry record last updated 2026-08-12; status last verified 2026-07.
• Results first posted 2026-05-12.
• 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.

21. Record Summary

COVID-19, Chronic Adaptation and Response to Exercise (COVID-CARE) is listed as a randomized, single-masked, crossover trial sponsored by the National Institutes of Health Clinical Center. The record lists enrollment of 48, a status of active, not recruiting, and a primary completion date of 2025-05-05.

The primary outcome is change in six-minute walk distance after the first 10 weeks, among participants who completed that period. The posted group means were 57.7 meters for aerobic exercise and 18.4 meters for control, with a mean difference of 39.3 meters and a 95% confidence interval from 16.72 to 61.88 meters. The unadjusted p-value was <0.001.

The registry reports zero serious adverse events and zero deaths in each listed safety group. It does not report secondary outcome results or the detailed statistical plan, including missing-data, multiplicity, and crossover-analysis methods.