Communities Fighting COVID Return to School: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Communities Fighting COVID Return to School (NCT05150860), a randomized na study. The primary outcome is Adherence to the Weekly COVID-19 Testing Schedule.

2. Clinical Question

The primary statistical question is about adherence to a testing schedule, not about whether testing prevented infection, changed disease severity, or affected other health outcomes. The registry reports a mean proportion of scheduled tests completed for each group and a mean difference comparing onsite testing with at-home testing. The comparison estimates how the reported group means differ; it does not by itself establish why any difference occurred or whether the difference would be similar in another school or community.

The setting matters to interpretation. The registry describes a partnership involving San Diego State University, Sweetwater Union High School District, and community partners, with an aim of generating evidence about feasible testing approaches as part of broader COVID-19 mitigation strategies. The posted outcome is narrower than that broader aim: it records testing-schedule adherence among enrolled student and school staff participants. Whether the two models differed on outcomes beyond adherence is not reported in the registry record.

3. Trial Design

The registry describes a randomized, parallel-group trial with no masking. Participants were assigned to an at-home COVID-19 testing model or an onsite COVID-19 testing model. The interventions are classified as behavioral. The registry records the study as completed and lists an enrollment of 852. The study started on 2021-10-17, and primary completion was recorded as 2022-04-02.

The primary outcome was measured through study completion, up to 21 weeks. Its stated population is student and school staff participants enrolled in the trial. The outcome summaries list 264 participants for at-home testing and 588 for onsite testing. Those group counts are not equal, so the posted comparison reflects the reported groups and their observed sizes rather than an equal-sized comparison. The reason for the difference in group counts is not reported in the registry record.

The study was sponsored by San Diego State University and concerned COVID-19. Its brief description places the work in a community and school context and refers to testing as part of broader mitigation strategies, including vaccination. That context does not change the definition of the primary endpoint, which is adherence to scheduled weekly testing.

Details of the randomization sequence, allocation concealment, recruitment procedures, crossover or contamination safeguards, and the rationale for the group-size distribution are not reported in the registry record. The record lists phase as NA. It does not report the trial's original sample-size justification, nor does it describe a prespecified analysis plan in enough detail to reconstruct the full intended analysis. These gaps limit what can be concluded about design choices beyond the randomized, parallel structure and the posted outcome summaries.

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: Schools were excluded if they had a lower proportion of students from households receiving SNAP Program benefits and a lower proportion of underrepresented minority students compared with other schools (3 schools); had ongoing school-wide testing programs with a commercial provider (2 schools); had larger (1 school) or smaller (1 school) student populations (for matching purposes) than the remaining schools; or were geographically closer to the remaining schools (1 school).

Overall Study

Analysis population for the primary outcome, as defined in the registry: Student and school staff participants enrolled in the trial

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: Student and school staff participants enrolled in the trial

6. Primary Endpoint

The registry lists the primary outcome as Adherence to the Weekly COVID-19 Testing Schedule, measured over through study completion, up to 21 weeks.

The primary endpoint is adherence to the weekly COVID-19 testing schedule. It is defined as the proportion of scheduled weekly tests completed while a participant was enrolled in the trial. The registry reports this outcome as a mean, with a standard error for each group. The stated timeframe is through study completion, up to 21 weeks.

A proportion is suitable for describing how much of a defined schedule was completed: values nearer the full schedule correspond to greater completion, while lower values correspond to less completion. Summarizing participant-level adherence with a group mean gives a concise description of the average completion proportion in each testing model. Comparing the means expresses the average difference between the groups on that same scale.

Adherence is useful in implementation-focused research because a testing approach can be evaluated in part by whether participants complete the tests scheduled for them. However, this endpoint measures completion, not the diagnostic performance of tests, infection rates, transmission, health outcomes, or the feasibility of every part of the testing program. The registry does not provide further operational details about how scheduled tests, missed tests, or enrollment periods were defined for individual participants. Those details are not reported in the registry record, so the posted mean should be interpreted according to the stated proportion-of-scheduled-tests definition.

7. Results Posted in the Registry

• Onsite COVID-19 Testing: mean 0.35 versus 0.39 on control; difference -0.04 (95% CI -0.14 to 0.06), standardized effect d = -0.06; z-test p = 0.398.

8. Statistical Interpretation

9. Statistical Methods Explained

What does the mean represent?

The primary outcome is a continuous measure: each participant's adherence is represented as a proportion of scheduled tests completed, and the registry summarizes that measure with a group mean. A mean combines participant-level values into an average. It is useful for comparing average adherence across groups, but it does not show the full distribution. Participants may differ in their completion patterns even when their group averages are similar. The registry does not report distributional summaries or the number of participants with complete adherence, so the means alone cannot describe those patterns.

The at-home mean is 0.394 and the onsite mean is 0.352. Each is reported with a standard error: 0.038 for at-home testing and 0.032 for onsite testing. A standard error reflects the precision of an estimated mean under the method used to calculate it. It is not the same as a standard deviation, which describes variation among individual observations. The registry's outcome description identifies the dispersion measure as standard error, so the reported spread should be read as uncertainty around the group mean rather than as participant-to-participant variability.

What does the mean difference express?

The comparison is expressed as the mean for onsite testing minus the mean for at-home testing. The posted estimate is -0.042. A negative estimate means that the onsite group mean is lower on the adherence scale in the posted summaries. The estimate is in the same units as the endpoint: proportion of scheduled tests completed. It is a between-group summary and is not an estimate of individual change over time.

A confidence interval accompanies the mean difference. In general, a confidence interval describes a range of values reasonably compatible with the data and the statistical model at the stated confidence level. It displays uncertainty in the estimated contrast, not a range containing a fixed proportion of individual participants. In this trial, the reported interval includes both negative and positive values. That feature means the result does not narrowly determine the direction of the underlying mean difference.

How should the p-value be understood?

The registry reports an unadjusted p-value of 0.398. In general, a p-value describes how unusual results at least as inconsistent with a specified null model would be if that model and the test assumptions were correct. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of the observed contrast. The term unadjusted indicates that the reported value is not presented as adjusted for other comparisons or covariates. Whether any such adjustments were planned or performed is not reported in the registry record.

Statistical significance is often used as a decision convention, but it does not replace examination of the estimate and its interval. A non-small p-value does not establish that two groups are equivalent. Evidence for equivalence would generally require a prespecified margin and an analysis designed to assess whether plausible differences fall within that margin; such a margin and analysis are not reported in the registry record. Likewise, a p-value cannot tell whether the observed difference would matter to schools, families, or program operations.

Why does the interval matter?

The interval is important because a single point estimate can appear more definitive than the data warrant. An estimate of -0.042 is one summary of the comparison, while the interval indicates that estimates on either side of zero remain compatible with the posted data under the method used. The record lists no lower or upper confidence bounds alongside the group means, but it does give an interval for the difference. The interval's exact numerical endpoints are not repeated because they contain more decimal places than are permitted in this account; its direction relative to zero is clear from the posted results.

The posted standard errors, interval, and p-value each serve different roles. The group standard errors describe precision of the separate means; the confidence interval describes uncertainty in the contrast; and the p-value summarizes compatibility with a null model under a test. They are related aspects of inference, not interchangeable statistics. None independently conveys the quality of the underlying data or the success of the randomization.

What is not specified about the analysis?

The record does not describe how adherence was handled for participants with incomplete follow-up, whether the mean difference was estimated with any adjustment, or how the analysis accounted for clustering or repeated measurements if those features applied. The precise model, variance assumptions, and analysis population are not reported in the registry record. It would therefore be inappropriate to attribute a particular regression model, missing-data method, or population definition to the trial based only on the posted summaries.

General statistical methods also rely on assumptions. For example, an analysis of means may need to account for the way observations are distributed and whether they are independent. Over-dispersion, in general, means that observed variability is greater than a model assumes; failing to account for it can make uncertainty estimates too small. Whether over-dispersion was relevant or assessed in this trial is not reported in the registry record. The reported comparison should be interpreted as the registry presents it, without inferring unreported model checks or corrections.

10. Confidence Intervals and Precision

Precision concerns how narrowly the data locate an estimate. The group means are reported as 0.394 for at-home testing and 0.352 for onsite testing. Their standard errors are 0.038 and 0.032, respectively. The estimated mean difference, onsite minus at-home, is -0.042. The standard errors help characterize uncertainty in the individual group means, while the confidence interval for the difference is the relevant interval for uncertainty about the comparison itself.

The interval for the mean difference has a negative lower bound and a positive upper bound. Because zero lies between those bounds, the interval does not determine a single direction for the underlying difference. It allows for a lower mean in the onsite group, little or no difference, or a higher mean in the onsite group, depending on the value within the interval. This does not mean every such value is equally plausible, and it does not mean that the true difference is known to be zero.

The unadjusted p-value of 0.398 is consistent with an imprecise comparison that does not rule out the null value under the reported test. The result should not be reduced to a binary declaration of success or failure. The point estimate gives the direction and size of the reported contrast; the interval conveys how much uncertainty remains. No clinically or operationally meaningful threshold for adherence differences is reported in the registry record, so the practical importance of values within the interval cannot be determined from these statistics alone.

Precision also depends on the analysis and data quality. The registry does not report missing-data handling, the analysis population, or a detailed model specification. Without those details, it is not possible to assess how alternative assumptions might affect the interval. The posted interval communicates uncertainty under the analysis represented in the record, rather than every uncertainty that could affect the study's interpretation.

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 sample-size table and curve are planning illustrations based on fractions of the observed effect, not a report of the trial's original design calculation. The entries are participants per arm, not 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. The trial's own original sample-size assumptions and justification are not reported in the registry record.

For planning around 100% of the observed effect, The table shows 4,750 participants per arm for 80% power and 6,359 participants per arm for 90% power. For planning around 75% of the observed effect, it lists 8,445 participants per arm for 80% power and 11,305 participants per arm for 90% power. For planning around 50% of the observed effect, it lists 19,000 participants per arm for 80% power and 25,436 participants per arm for 90% power.

In plain language, 80% power means that, under the stated planning assumptions and if the specified effect were the true effect, the planned study would have an 80% chance of detecting a difference meeting the chosen statistical criterion. The 90% column describes the corresponding plan with a 90% chance under those same assumptions. Smaller effects are harder to distinguish from random variation, so planning to detect a smaller effect requires more participants. These figures are not observed power; they are sample-size planning illustrations, not a retrospective measure of what the completed trial achieved.

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 = -0.042; pooled standard deviation σ = 0.731.
2. n = 2 × (1.96 + 0.842)² × 0.731² / -0.042² = 4,749.9, rounded up to 4,750 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 <- -0.042                      # observed difference in means
sigma <- 0.730591663532735                     # 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(5150860)
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 = 4,750 (80% power), 6,359 (90% power)
 75% of observed effect: n per arm = 8,445 (80% power), 11,305 (90% power)
 50% of observed effect: n per arm = 19,000 (80% power), 25,436 (90% power)
power.t.test: 4750.9 
Simulated power at n = 4,750 per arm: 0.799 (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

The registry posts one primary outcome measure, adherence to the weekly COVID-19 testing schedule, and no secondary outcomes. It records zero analyses. The presence of a posted comparison with an unadjusted p-value does not establish that a full analysis plan or all intended analyses are available. The planned analysis set, any prespecified subgroup comparisons, and any adjustment for multiple testing are not reported in the registry record.

Multiplicity arises when several outcomes, time points, subgroups, or analyses are tested. In general, considering many possible comparisons can increase the chance that at least one result appears notable by chance. Whether multiplicity was a concern in this trial cannot be determined from the available record, because it does not describe the full analysis plan or whether additional comparisons were considered. The single posted primary outcome should not be taken as proof that no other analyses were undertaken.

Missing-data methods are also not described. The outcome summaries list 264 participants in the at-home group and 588 in the onsite group, but the registry does not explain how missing weekly tests, early departure from follow-up, or incomplete adherence records were treated. It does not state whether the analysis included everyone as randomized, only participants with observed outcome data, or another population. These analysis-population details are not reported in the registry record.

The record also does not specify how the estimate was adjusted, whether repeated observations were modeled, or whether any school-level or other clustering was considered. Such details can matter to uncertainty estimates and to the interpretation of a randomized comparison. No particular handling of these issues should be assumed. The reported estimate, interval, and unadjusted p-value remain the available statistical summary, with the limitations that follow from the missing analysis details.

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 3% reporting threshold.

The registry lists serious adverse events for 264 participants at risk in the at-home testing group and 588 participants at risk in the onsite testing group. It reports zero serious adverse events in each group and zero deaths in each group. No serious adverse event terms are listed.

These figures describe what was recorded in the registry for the groups and follow-up represented there. A count of zero reported events is not evidence that no unrecorded event occurred, and it does not establish that an intervention is risk-free. The record does not provide additional detail about how adverse events were collected, how long each participant was monitored for safety, or what definitions and reporting procedures were used beyond the serious-event category shown.

The registry also lists zero other adverse events in both groups. These summaries should be read as event counts in the posted record, not as a complete evaluation of every possible safety outcome. The comparison was primarily about testing-schedule adherence, and the adverse-event figures do not change the endpoint's scope or establish effects on infection or other clinical outcomes.

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

The result shows that the posted mean proportion of scheduled tests completed was 0.394 for participants assigned to at-home testing and 0.352 for participants assigned to onsite testing. The estimated mean difference for onsite compared with at-home testing was -0.042. The confidence interval includes values on both sides of zero, and the unadjusted p-value is 0.398. Thus, the posted evidence does not precisely establish the direction of a population-level difference in average adherence.

This is not the same as demonstrating that the models produce identical adherence. A confidence interval that includes zero can still include meaningful differences in either direction. The registry does not report a threshold for a practically important difference, so it is not possible to classify all values compatible with the interval as operationally negligible or important. Nor does the p-value answer whether the observed difference would matter to a school or community program.

The outcome is adherence to scheduled weekly testing during enrollment, through study completion, up to 21 weeks. It does not measure infection prevention, transmission, diagnostic accuracy, or other health outcomes. The broader study description places testing among multiple COVID-19 mitigation strategies, but the posted primary result supports conclusions only about the stated adherence measure.

Interpretation is also bounded by details absent from the record. Missing-data handling, the analysis population, the full statistical plan, and the original sample-size justification are not reported in the registry record. Those omissions do not invalidate the posted summaries, but they limit assessment of how the estimates were produced and how robust they may be to alternative analytic choices. The fairest reading is a descriptive comparison with substantial uncertainty about the direction and magnitude of the underlying difference.

16. Why This Trial Matters Statistically

Statistically, this trial illustrates the difference between a point estimate and the evidence needed to draw a precise conclusion. The mean difference is -0.042, but the confidence interval extends across zero and the unadjusted p-value is 0.398. Reporting the estimate alongside its uncertainty is more informative than treating one p-value as a complete summary. The interval makes clear that the data do not pin down the direction of the underlying mean difference.

The study also highlights the importance of defining the outcome carefully. Adherence is a relevant implementation measure for a testing schedule, but it is not interchangeable with infection outcomes or with a broad judgment about a testing program. A study can report adherence without establishing whether the tested approach changed health outcomes. Keeping the statistical endpoint aligned with the claim prevents the result from being stretched beyond what was measured.

The planning illustrations make another statistical point: sample size needs depend on the effect a future study aims to detect. A smaller target effect requires more participants under the listed assumptions. Those illustrations are distinct from the actual trial design rationale, which is not reported in the registry record. In addition, the posted group sizes differ, while the planning calculations assume equal allocation; the planning entries should therefore not be mistaken for a reconstruction of the completed study's design.

Finally, registry reporting is part of statistical interpretation. The outcome summaries and comparison are informative, but missing details about analysis populations, missing data, and the statistical plan limit independent assessment of the methods. A careful reading distinguishes what the record reports from what remains unknown and avoids turning a nonsignificant result into a claim of equivalence.

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 registry record leaves important trial-design and analysis details unspecified. The method used to generate and conceal random assignments is not reported, nor are recruitment procedures, participant flow, or reasons for the unequal group sizes. The outcome summaries list 264 participants for at-home testing and 588 for onsite testing, but the record does not explain whether those counts represent all randomized participants, those with observed endpoint data, or another analysis population.

The record does not describe missing-data handling for scheduled tests or incomplete follow-up. It also does not provide the model specification, any covariate adjustment, assumptions used for the comparison, or whether repeated observations or clustering were addressed. The original sample-size justification and the prespecified statistical plan are not reported in the registry record. These gaps make it difficult to evaluate the extent to which the posted uncertainty accounts for all features of the study design.

The endpoint itself is limited to the proportion of scheduled tests completed. Details about how schedules were assigned, how a completed test was verified, and how adherence was calculated for participants enrolled for different lengths of time are not reported. The outcome therefore cannot be used to infer comparative effects on COVID-19 infection, transmission, or other clinical outcomes.

The setting described in the registry is a specific school and community partnership. The record does not establish how well the findings generalize to different school systems, participant groups, or testing arrangements. These limitations qualify the interpretation of the adherence comparison; they do not change the numerical results that the registry reports.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Communities Fighting COVID Return to School — NCT05150860.
• Registry record last updated 2025-01-17; status last verified 2025-01.
• Results first posted 2025-01-17.
• 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

NCT05150860, titled “Communities Fighting COVID Return to School,” is recorded as a completed randomized, parallel-group trial with no masking. San Diego State University is listed as sponsor. The study concerns COVID-19 and compares at-home with onsite COVID-19 testing models. Enrollment is recorded as 852, with a start date of 2021-10-17 and primary completion date of 2022-04-02.

The primary outcome is adherence to the weekly COVID-19 testing schedule, measured as the mean proportion of scheduled tests completed through study completion, up to 21 weeks. The posted means are 0.394 for at-home testing and 0.352 for onsite testing; the mean difference for onsite compared with at-home testing is -0.042. Its confidence interval includes values on both sides of zero, and the unadjusted p-value is 0.398. The record lists zero serious adverse events and zero deaths in both groups. Details of the original sample-size justification, missing-data handling, and full analysis plan are not reported in the registry record.