Clinical Validation of New Injection Molded Flocked Nasopharyngeal Swabs in Response to the COVID-19 Pandemic: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Clinical Validation of New Injection Molded Flocked Nasopharyngeal Swabs in Response to the COVID-19 Pandemic (NCT04368260), a non-randomized na study. The primary outcome is Number of Subjects for Which the Prototype Swab and the Control Swab Provide the Same COVID-19 PCR Result.

2. Clinical Question

3. Trial Design

The registry describes a clinical validation study of prototype nasopharyngeal swabs developed in response to a shortage of swabs used for virological testing during the COVID-19 pandemic. The sponsor is the University of Virginia. The condition is COVID19, the phase is listed as not applicable, and the study status is completed. The registry lists the start date as 2020-04-24, the primary completion date as 2020-08-01, and enrollment as 40.

The study is classified as parallel, non-randomized, and unmasked. Those labels describe the registry's overall design fields, but the primary outcome description says that each subject would have both a prototype and a control swab used. That makes the intended primary comparison a within-subject comparison of the two PCR results. The registry also lists the swabs under separate intervention groups, but it cautions that the allocation was non-randomized and that the arms are not a like-for-like control comparison. Group-level comparisons should therefore not be treated as if they arose from randomized assignment.

The prototype is an injection molded polypropylene flocked nylon nasopharyngeal swab. The control is described as an FDA-cleared swab. The study summary says the nasopharynx would be swabbed with both devices in patients with COVID-19 and patients under investigation for COVID-19, with each sample transported in validated FDA-cleared viral transport medium under the medical center's standard operating procedure. The record does not describe the order in which swabs were collected or whether collection order was controlled.

Although the registry marks the study completed, that status does not itself establish that the primary outcome was analyzed or that results were posted. The primary outcome value, the number of evaluable paired results, and the details of the final analysis are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

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 Number of Subjects for Which the Prototype Swab and the Control Swab Provide the Same COVID-19 PCR Result, measured over at time of enrollment.

The primary endpoint is the number of subjects for whom the prototype swab and the control swab yield the same COVID-19 PCR result at enrollment. It is a count of matched results, rather than a direct measure of how many infections were correctly detected. The underlying comparison is paired: both devices are used in each subject, so the key observation is whether the two results for that subject agree.

Agreement is relevant to a swab validation question because it describes whether the prototype and the established control produce the same reported test classification under the study's collection and transport procedures. A count alone, however, is difficult to interpret without the number of subjects with usable paired results. A rate of agreement would relate the number of matching pairs to the number of evaluable pairs, and the registry does not report either the observed match count or the evaluable denominator.

Agreement also does not establish correctness. If both swabs produce the same result, that does not show that the result is accurate against an independent reference standard. Nor does agreement by itself demonstrate non-inferiority. A non-inferiority conclusion generally depends on a prespecified margin and an appropriate estimate with uncertainty. The registry record does not report the margin, the observed endpoint, or an analysis establishing whether any non-inferiority criterion was met.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The central statistical feature is that both swabs are used in each subject. The primary outcome is therefore based on paired observations: each subject contributes a prototype result and a control result, and the endpoint counts subjects whose results match. A paired design differs from comparing outcomes in unrelated groups because the two measurements from one subject are likely to share subject-level characteristics. As a general principle, an analysis should respect that pairing rather than treat the two measurements as independent observations.

What does agreement measure?

Agreement describes how often two procedures give the same classification or value. For this study, the registry defines the primary endpoint as a count of subjects with the same COVID-19 PCR result from both swabs. A count is an absolute quantity. To describe the frequency of agreement, it would ordinarily be interpreted against the number of subjects with complete results from both devices. The record does not report the match count or the number of evaluable pairs, so an agreement rate cannot be read from the registry information.

Agreement is distinct from accuracy. A comparison can show that two methods agree without establishing which result is correct. Accuracy ordinarily requires a reference standard or another basis for determining the true status. The registry's primary outcome description is a comparison between prototype and control results; it does not report an independent reference standard for adjudicating discrepancies. Thus the stated endpoint addresses concordance between swabs, not sensitivity, specificity, or correctness against a separate truth standard.

How does non-inferiority differ from agreement?

As a general concept, a non-inferiority analysis asks whether a new approach is not unacceptably worse than a comparator according to a prespecified margin. That margin gives statistical meaning to the phrase not unacceptably worse. A study also needs a defined estimand, an analysis method, and a decision rule that connects the estimated difference and its uncertainty to the margin. The registry summary states a non-inferiority objective, but the margin, estimand, and decision rule are not reported in the registry record. The objective alone cannot establish that the criterion was met.

Agreement and non-inferiority are related only insofar as an agreement measure might be used to evaluate a validation objective. They are not interchangeable. A count of matching pairs does not by itself state how much disagreement is acceptable, how uncertainty was handled, or how results were compared with a non-inferiority threshold. Those details would be needed to interpret a formal non-inferiority conclusion.

Why does pairing matter?

With paired measurements, the result from one device is naturally linked to the result from the other device for the same subject. General statistical methods for paired binary data use the pattern of matched and mismatched pairs, rather than only separate totals for each device. The mismatches can be particularly informative about whether one device tends to yield a different classification from the other. The registry record does not post the paired-result pattern or identify a method used to analyze it. It is therefore not possible to assess the statistical treatment of within-subject dependence.

Collection order could also matter as a general design consideration if one collection affects the other, but the order and any order-control procedure are not reported in the registry record. This does not show that order affected results; it identifies a design detail that cannot be evaluated from the posted information.

What can be inferred from the planning assumptions?

The registry includes assumptions for a planning illustration: a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. These assumptions describe a planning calculation, not the trial's own design or an observed analysis. The registry record does not report the observed primary effect needed to evaluate such an illustration, and a planning sample size is not observed power. A sample-size calculation under hypothetical assumptions cannot replace a reported analysis of the study's data.

The registry's separate arm labels should also be interpreted in light of its design note. It classifies allocation as non-randomized and cautions that the arms are not a like-for-like control comparison. That warning prevents a simple interpretation of group-specific counts as randomized treatment effects. The primary endpoint instead describes both swabs being used in each subject, so paired data are central to the stated question.

What statistical evidence is absent?

The registry lists one outcome measure and zero analysis records. It does not report the primary result, an analysis population, a missing-data method, a confidence interval, a p-value, or a non-inferiority margin. These omissions do not establish that no analyses were performed outside the registry. They mean that the statistical analysis and its results are not reported in the registry record and cannot be evaluated from that record.

In general, an estimate describes the magnitude of an observed result, while its uncertainty describes how precisely that result has been estimated under an analysis model. For paired binary data, precision depends in part on the observed pattern of agreement and disagreement and on how many pairs can be analyzed. Neither the paired outcome pattern nor the evaluable paired sample is reported. No quantitative conclusion about the size or precision of agreement can therefore be drawn from the registry entry.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and zero analysis records. The named primary outcome is the number of subjects for whom both swabs provide the same COVID-19 PCR result. The record does not report a statistical analysis for that outcome, nor does it describe additional outcome comparisons or a multiplicity adjustment. A multiplicity plan is not reported in the registry record; the presence of one listed outcome does not establish what analyses may have been considered outside the record.

Missing-data handling is also not reported. The registry does not state how invalid, unavailable, or incomplete PCR results would be handled, whether both swabs had to yield an interpretable result for a subject to enter the primary analysis, or how many subjects had complete paired results. The listed enrollment of 40 should not be read as the number analyzed for the endpoint.

The analysis population is not defined in the posted results. The record does not say whether the primary analysis included all enrolled subjects, only subjects with paired results, or another population. This matters because exclusions or missing pairs can change both the denominator and the interpretation of the agreement estimate.

No interim analysis plan, analysis-set definitions, or statistical analysis plan is reported in the registry record. These absences limit reproducibility and make it impossible to determine how outcome availability, multiplicity, or analysis population choices were addressed.

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 registry's safety entries list the Control Swab group and the Prototype Swab group separately. For each group, the entry lists 40 at risk for serious adverse events and 0 affected; it also lists 40 at risk for other adverse events and 0 affected. The deaths entry is 0 for each group, and no top serious terms are listed.

These figures describe what is recorded in the registry, not proof that no adverse event occurred in any setting or that all possible events were captured. The record does not provide further detail about how events were defined, collected, assessed, or attributed to a swab, nor does it explain the relationship between the registry's separate group entries and the paired use of both swabs in each subject. The two group entries should not be added together to infer a number of distinct subjects, because the primary outcome description indicates that both devices were used in each subject.

Zero reported events means that no affected participants are listed in these safety fields. It does not quantify the precision of safety estimates, establish the absence of risk, or resolve whether monitoring captured events completely. The entries are best read as registry-reported safety counts with limited supporting detail.

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

The record establishes that the University of Virginia registered a completed clinical validation study of an injection molded flocked nasopharyngeal swab. It specifies the intended comparison with an FDA-cleared control, the use of both swabs in each subject, and a primary endpoint based on whether the PCR results match. It also identifies patients with COVID-19 and patients under investigation as the intended population.

The registry does not report the primary result. It is therefore not possible to say how often the swabs agreed, how many paired samples were evaluable, or how precisely agreement was estimated. The stated non-inferiority aim is not itself evidence that non-inferiority was demonstrated. The non-inferiority margin and decision rule are not reported, and no analysis or confidence interval is posted.

The endpoint is also narrower than overall diagnostic performance. Matching results indicate concordance between the two swabs, but do not by themselves establish that either result is correct. The registry does not report an independent reference standard or diagnostic-accuracy results. It also does not provide enough information to assess whether the swab collection order, missing results, or other design details influenced the endpoint.

The registry's completed status and enrollment listing provide study-record context, not a substitute for results. The safety entries list zero serious adverse events, zero other adverse events, and zero deaths in each listed group, but the record provides limited detail about event collection and interpretation. These entries should be understood as reported counts, not as proof that risk is absent.

In short, the registry communicates the question the study intended to address, but the posted information does not answer it quantitatively. It cannot support a conclusion that the prototype performed better, worse, or similarly to the control, or that it met a non-inferiority criterion.

13. Why This Trial Matters Statistically

This study illustrates why a clear endpoint definition and a posted result are both needed for a useful statistical interpretation. The registry defines the primary endpoint as a count of subjects whose prototype and control PCR results match. That makes the intended outcome understandable, but without the count and the number of evaluable pairs, the degree of agreement cannot be quantified.

The paired structure is also important. Both devices are described as being used in each subject, so the primary comparison is not simply a comparison of two independent groups. Analyses of paired measurements should account for the link between observations from the same subject. At the same time, the registry classifies allocation as non-randomized and cautions that the arms are not a like-for-like control comparison. These design details make it especially important to distinguish the paired endpoint from any separate arm summaries.

The study's stated non-inferiority objective highlights another statistical distinction. A validation aim is not a statistical conclusion: a prespecified margin, a defined analysis, and an estimate with its uncertainty are needed to interpret non-inferiority. The registry record does not report those elements or the primary outcome result. This means readers can assess the question and broad design, but not whether the objective was met.

The record also shows how safety counts need context. Zero events are the recorded counts in the listed groups; event definitions, ascertainment details, and fuller safety analysis are not reported. Statistical interpretation depends not only on the displayed count, but also on what was collected and how completely.

In the context of the shortage described in the study summary, a validation comparison could be relevant to testing operations. The registry entry alone, however, does not establish whether the prototype had adequate agreement or whether it resolved a supply constraint. Those claims require results and methods that are not reported in the record.

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 specific study is that the completed registry record does not report the primary endpoint result or an analysis. The outcome is defined as the number of subjects with matching PCR results from both swabs, but neither the match count nor the number of complete paired results is reported. The registry's enrollment field lists 40, but it does not identify the number included in the endpoint analysis.

The summary states a non-inferiority objective, but the non-inferiority margin, the statistical decision rule, and the analysis method are not reported in the registry record. Without those details and an estimate with uncertainty, the objective cannot be evaluated. No confidence interval, p-value, or other inferential result is posted.

The study is classified as non-randomized and unmasked. The registry specifically cautions that the intervention groups are not a like-for-like control comparison. Although the endpoint description says both swabs would be used in each subject, the record does not report the order of collection or how any within-subject dependence was handled analytically. These limitations prevent assessment of the precise comparison and its potential sources of bias.

The registry does not describe the handling of missing, invalid, or discordant results, define the primary analysis population, or report whether all enrolled subjects contributed paired data. It also does not provide an independent reference standard for evaluating accuracy, so agreement between swabs should not be interpreted as diagnostic correctness.

Safety interpretation is limited by the absence of event definitions and collection details. The listed zero counts are registry entries, but cannot show how completely events were ascertained. Finally, the planning assumptions in the record are illustrations, not a report of observed power or evidence of the study's findings.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Clinical Validation of New Injection Molded Flocked Nasopharyngeal Swabs in Response to the COVID-19 Pandemic — NCT04368260.
• Registry record last updated 2022-02-17; status last verified 2022-02.
• Results first posted 2022-02-17.
• 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

NCT04368260 is a University of Virginia-sponsored clinical validation study of injection molded flocked nasopharyngeal swabs for COVID-19 PCR testing. The registry lists the condition as COVID19, phase as not applicable, allocation as non-randomized, masking as none, and status as completed. It lists enrollment as 40, a start date of 2020-04-24, and a primary completion date of 2020-08-01.

The prototype was compared with an FDA-cleared control swab. The primary outcome was the number of subjects for whom both swabs produced the same COVID-19 PCR result at enrollment. The summary describes a non-inferiority aim and says both devices would be used in each subject.

The registry does not report the primary outcome value, an analysis, or the details needed to assess non-inferiority. Its safety entries list zero serious adverse events, zero other adverse events, and zero deaths in each intervention group. The record therefore describes the study question and design, but does not provide a quantitative answer to the primary question.