ReSET Aim 1b: Restarting Safe Education and Testing for Children With Medical Complexity - COVID-19 Testing in School With Children and Staff: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for ReSET Aim 1b: Restarting Safe Education and Testing for Children With Medical Complexity - COVID-19 Testing in School With Children and Staff (NCT04899245), a non-randomized na study. The primary outcome is Change in Symptomatic Test Rate: Number of Symptomatic Tests Completed.

2. Clinical Question

3. Trial Design

The registry lists this study as completed, non-randomized, parallel, and unmasked. Its phase is recorded as not applicable. The sponsor is the University of Wisconsin, Madison. The study start date is 2021-05-04, and the primary completion date is 2023-06-15. Enrollment is recorded as 112. The listed intervention is the BinaxNOW Rapid Antigen System.

The study focused on SARS-CoV-2 testing in a school setting involving children with medical complexity and school staff. The registry describes the broader motivation as supporting safe return to school while recognizing that school attendance is important for children's academic, social, and health-related services. The brief study description states an objective of increasing safe return to school and begins to describe evaluating feasibility; the remainder of that objective is not reported in the registry record.

The arm descriptions give recruitment targets of 65 children and their parents and 50 school staff. These are described as recruitment plans, not as the numbers observed in the safety record. The safety record lists 47 children with medical complexity and 24 school staff at risk. The record does not explain the relationship between the planned recruitment numbers and those safety denominators, or identify the number completing the primary outcome assessment.

Because allocation was non-randomized and the two groups differ in who they are, their outcomes should not be interpreted as a direct treatment-versus-control contrast. The registry record does not report a control group, random allocation, masking, a protocol-level statistical analysis plan, or a detailed schedule of testing. Those design details limit what can be concluded about comparative effects, even if group-level counts are available elsewhere.

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: The protocol enrollment, started, and completed values reflect caregiver/child dyads. There were n = 41 caregivers, n = 47 children.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The number analyzed must include those who were approached and not consented in addition to those consented.

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 Symptomatic Test Rate: Number of Symptomatic Tests Completed, measured over Study duration (up to 21 months).

The primary outcome is titled “Change in Symptomatic Test Rate: Number of Symptomatic Tests Completed.” It is measured as a count of symptomatic BinaxNOW rapid antigen tests, with information retrieved from the study log. The stated observation period is the study duration, up to 21 months.

A count captures the volume of tests completed, rather than the proportion of participants testing positive or the accuracy of the test. A rate, in general, relates an event count to a defined amount of time or another exposure denominator. The registry outcome description specifies a number of tests but does not state the denominator or method used to calculate a rate, nor how change would be assessed. The registry record also does not report a numerical result for this endpoint.

Counting completed tests can describe how much symptomatic testing took place in the study setting. It is therefore relevant to questions about the implementation or feasibility of testing. However, a count alone does not establish why the number of tests took a particular value. It may reflect the number of participants, the time they were observed, opportunities to test, symptoms, or completion of testing. These are general considerations for count outcomes; the registry record does not report how each was handled in this study.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a count outcome represent?

The primary measure is the number of symptomatic rapid antigen tests completed. In general, count data record how many times an event occurs within a defined observation period. Counts are non-negative whole-number quantities and can take different values across participants or settings. For this study, the registry states that data would come from a study log and that the observation period could extend up to 21 months. It does not report whether the primary analysis used the total count, a rate, or another summary of the recorded tests.

When participants have different observation times or different opportunities for an event, a count can be difficult to interpret without context. In general, an analysis may account for exposure time or another relevant denominator when estimating a rate. The outcome title refers to a test rate and change, but the description names the number of tests completed; the rate denominator and the definition of change are not reported in the registry record. No particular rate model or time comparison can therefore be attributed to this study.

Why can count data require special attention?

In general, count outcomes are often analyzed with methods designed for discrete data. A basic count model may assume that the mean and variance follow a specified relationship. Over-dispersion means that the observed variability is greater than that model expects. If present, over-dispersion can make standard errors too small and confidence intervals too narrow when ignored. This is a general statistical concern, not a reported feature of the study data. The registry record does not report a count model, a variance estimate, a check for over-dispersion, or a method for addressing it.

Counts may also vary with exposure duration, participant characteristics, or the number of opportunities to record an event. In general, analysts need to define the unit of analysis and account for dependence when observations are clustered or repeated. This study took place in a school setting and the outcome concerns tests over a study period, but the registry record does not specify the unit used in the primary analysis, whether repeated observations were modeled, or how clustering was handled. Those methods should not be inferred from the setting or endpoint title.

What does the non-randomized design imply?

In a randomized comparison, chance assignment is used to create groups intended to be comparable on average. This study is registered as non-randomized and parallel. Its named groups are children with medical complexity and school staff. Since the groups differ in population, they do not form a like-for-like control comparison. In general, an observed difference between unlike groups may reflect characteristics of the groups as well as the intervention or context. Without randomization or a clearly defined comparison strategy, a causal interpretation requires assumptions and methods that the registry record does not report.

The registry also lists no masking. In general, masking means that some people involved in a study do not know an assigned intervention or group. The record does not say which participants or personnel, if any, could have been masked in practice, nor how the absence of masking affected outcome recording. It is therefore appropriate to describe the registered masking status without speculating about implementation.

How should estimates and uncertainty be read?

For an outcome analysis, an estimate describes the quantity calculated from the observed data, while uncertainty measures indicate how much precision the analysis supports. A confidence interval is a range constructed by a statistical procedure to express uncertainty about an estimate under that procedure's assumptions. A p-value, in general, measures how unusual data at least as incompatible with a specified null model would be, if that model and its assumptions held. Neither measure establishes clinical importance by itself.

The registry record does not report a primary-outcome estimate, confidence interval, p-value, or analysis model. It lists zero analyses. Consequently, there is no reported inferential result to interpret using these concepts. The description of an endpoint should not be mistaken for an estimated effect, and the absence of an analysis result in the registry record is not evidence that the true outcome equals zero.

What is the role of the planning illustration?

The stated planning illustration is based on an observed effect and specifies a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It describes the effect to detect as a stated fraction of the observed effect. For time-to-event outcomes, the illustration is described as giving a total number of events; for other outcomes, it gives participants per arm. These are planning assumptions, not a description of the study's registered non-randomized design. No numerical planning estimates or observed effect are reported in the registry record discussed in these trial details.

In general, a sample-size calculation asks how much information a future or planned study would need to detect an assumed effect under specified conditions. It is not an observed power calculation and does not establish whether the study found an effect. The illustration's equal-allocation assumption does not convert the distinct school staff and child groups into comparable arms. It also cannot substitute for a trial-specific sample-size justification, which is not reported in the registry record.

Overall, the registry record supports describing the outcome, setting, design labels, and reported safety counts. It does not support naming a specific statistical model, reconstructing an analysis, or drawing a numerical conclusion about the primary endpoint.

10. Multiplicity, Missing Data and Analysis Populations

The record lists 30 outcome measures and zero analyses. It identifies one primary outcome, the number of symptomatic tests completed, but does not report results for that outcome or numerical findings for the other outcomes. A complete analysis plan describing outcome hierarchy, planned comparisons, or the handling of multiple tests is not reported in the registry record.

Multiplicity arises in general when multiple outcomes, time points, groups, or statistical comparisons are examined. The more opportunities there are to identify a nominally unusual result, the more important it becomes to distinguish planned analyses from exploratory ones and to consider appropriate methods for controlling error or interpreting findings. No multiplicity adjustment or strategy is reported for this study. With no listed analyses or outcome estimates, it is not possible to assess whether multiplicity affected any reported inference.

The registry record does not describe missing-data amounts, reasons for missingness, exclusions, or imputation methods for the primary outcome. It also does not report an analysis population definition, such as which enrolled participants were included in each analysis. The primary outcome is based on study-log data, but the record does not state whether every participant had complete logs, how incomplete records were treated, or whether participants with no recorded symptomatic tests were counted as having zero tests.

These omissions matter because missing observations and analysis-population choices can influence both the estimated outcome and its precision. In general, methods for missing data depend on why information is absent and on the assumptions made about that process. No missing-data assumption or sensitivity analysis is reported in the registry record. Accordingly, the available record does not permit an assessment of whether the primary outcome analysis, if conducted, was robust to incomplete data or analytic 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.

The safety record lists the children with medical complexity group with 47 participants at risk and zero serious adverse events, zero other adverse events, and zero deaths. It lists the school staff group with 24 participants at risk and zero serious adverse events, zero other adverse events, and zero deaths. These are the recorded counts in the registry record for the groups and categories shown.

A zero event count is not the same as proof that risk was absent. The meaning of a safety count depends on how events were defined, how and when they were collected, and which participants were observed. The registry record does not report the event definitions, surveillance procedures, duration of safety follow-up, or whether events were judged related to testing. It also does not report event-level details beyond the zero counts.

The safety denominators differ from the recruitment targets and overall enrollment figure. The record does not explain those differences, so the safety counts should be read with their listed denominators rather than assumed to represent all enrolled participants. These figures describe reported events; they do not establish that testing caused or prevented any event.

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

The registry record describes a school-based study using the BinaxNOW Rapid Antigen System with children with medical complexity and school staff. Its primary outcome was the count of symptomatic tests completed during the study, using study-log data over a period of up to 21 months. This provides a clear description of the activity the primary endpoint was intended to capture.

The record does not report the observed primary outcome, its change, a rate estimate, or uncertainty around an estimate. It therefore does not establish how many symptomatic tests were completed, whether testing changed over time, or how precisely any such quantity was measured. An endpoint definition is not itself a result.

The non-randomized design and distinct participant groups also constrain comparative interpretation. Children with medical complexity and school staff are not interchangeable groups or a randomized intervention and control pair. Even if their outcomes were reported, a contrast between them would not, on its own, identify an effect of testing. The registry record does not report a suitable control group or a causal analysis.

The safety record reports zero serious adverse events, zero other adverse events, and zero deaths in each listed group, with 47 children with medical complexity and 24 school staff at risk. Those entries are limited to the recorded categories and denominators; they do not prove the absence of all possible harms or determine whether an event was caused by testing.

In general, feasibility and implementation studies can inform whether procedures are being carried out in a particular setting. The study's stated objective concerns increasing safe return to school, but the registry record does not provide enough outcome results to determine whether that objective was achieved. Nor does the record establish diagnostic accuracy, changes in infection risk, or effects on school attendance. Such conclusions would require outcomes and analyses not reported in the registry record.

13. Why This Trial Matters Statistically

This study raises statistical issues that are common in research conducted in real-world settings. Its primary outcome is a count of completed symptomatic tests, yet its title also refers to a rate and change. Counts and rates answer different questions: a count describes how many events were recorded, while a rate relates events to a denominator such as time or another exposure. Clear definitions are essential for readers to know what the estimate represents. The registry record does not state the rate denominator or how change was calculated.

The setting also makes the design important. The groups are children with medical complexity and school staff, and allocation is non-randomized. These are meaningful populations for describing school testing, but they are not naturally comparable control arms. The distinction helps prevent a descriptive comparison from being mistaken for a causal estimate.

Reporting the outcome definition alongside its actual estimate and uncertainty would be necessary to assess the amount and precision of testing. The registry record identifies the planned measurement source and observation period but does not report a numerical primary result or analysis. This illustrates why endpoint registration, outcome reporting, and statistical inference are separate parts of a study record.

The safety entries also show why denominators and event-collection details matter. Zero recorded events are interpretable only in relation to who was observed and how events were defined and captured. Here, the record provides group-specific at-risk counts but not the procedures needed to characterize the scope of surveillance.

These features matter statistically because the design, endpoint definition, analysis, and reporting jointly determine what conclusions are warranted. The available record supports a description of the study's focus and registered measures, but not an estimate of testing activity or a comparative treatment effect.

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 registry record does not report numerical results for the primary outcome, despite specifying that symptomatic tests were to be counted from a study log. It also does not give a rate denominator or define how change in the test rate would be assessed. As a result, the principal endpoint cannot be interpreted quantitatively from the record.

The design is non-randomized, and the named groups comprise children with medical complexity and school staff. The record explicitly cautions that these are not a like-for-like control comparison. Any between-group contrast would therefore require careful interpretation and could not be treated as a randomized estimate of a testing effect.

The registry lists enrollment of 112, while the arm descriptions give recruitment targets of 65 children and parents and 50 staff, and the safety record lists 47 children with medical complexity and 24 staff at risk. The record does not explain how these figures relate to one another or provide primary-outcome analysis denominators. The number contributing to the primary endpoint is not reported in the registry record.

The statistical analysis plan, count model, handling of repeated observations or clustering, missing-data methods, and multiplicity strategy are not reported in the registry record. Nor are outcome estimates, confidence intervals, or p-values listed. These gaps prevent assessment of precision, analytic assumptions, and robustness.

Finally, safety-event definitions, ascertainment procedures, follow-up details, and attribution are not reported. The recorded zero event counts should therefore be understood as limited registry entries, not as comprehensive evidence that no safety events occurred. The record's description of the study objective is also incomplete, so additional objectives should not be inferred.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: ReSET Aim 1b: Restarting Safe Education and Testing for Children With Medical Complexity - COVID-19 Testing in School With Children and Staff — NCT04899245.
• Registry record last updated 2024-07-17; status last verified 2024-06.
• Results first posted 2024-07-17.
• 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.

18. Record Summary

ReSET Aim 1b (NCT04899245) is a completed, non-randomized, parallel, unmasked study sponsored by the University of Wisconsin, Madison. It concerns SARS-CoV-2 testing in a school setting among children with medical complexity and school staff. The listed intervention is the BinaxNOW Rapid Antigen System. Enrollment is recorded as 112; the study start date is 2021-05-04, and the primary completion date is 2023-06-15.

The primary outcome is the number of symptomatic rapid antigen tests completed, recorded from a study log during the study duration, up to 21 months. The registry record does not report a numerical primary-outcome result or its statistical analysis. The safety record lists zero serious adverse events, zero other adverse events, and zero deaths among 47 children with medical complexity and 24 school staff at risk. These counts do not establish a causal effect or the absence of all possible harms.