Examining How a Text Message-based Educational Intervention Influences COVID-19 Testing: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Examining How a Text Message-based Educational Intervention Influences COVID-19 Testing (NCT07074171), a non-randomized na study. The primary outcome is Number of Participants Who Self-Reported Use of a COVID-19 Antigen or PCR Test.

2. Clinical Question

The study concerns whether a text message-based COVID-19 education intervention, supported by community health workers, is associated with COVID-19 testing and related behaviors among people recruited from intervention priority census block groups. Its primary endpoint is the number of participants who self-reported using a COVID-19 antigen test or PCR test in the past 60 days. The endpoint is specified at baseline, so it describes reported testing before the intervention can be evaluated as a change over time.

The record also describes a historical control group drawn from participants in the prior Phase II embedded study. Those participants received CDC-designed testing brochures and completed pre- and post-intervention surveys. The registry states that new participants were recruited from the same priority block groups using convenience sampling and had not participated in Phase I or Phase II. The central interpretive question is therefore not simply whether the intervention group differs from a concurrent randomized control. It is whether outcomes in a newly recruited intervention group can be meaningfully compared with outcomes collected earlier from a different group under a different study phase and intervention context.

3. Trial Design

The study is listed as Phase III, completed, non-randomized, unmasked, and with a single-group model. The registry record describes a text message-based educational intervention supported by community health workers and a historical control group drawn from the control arm of the Phase II embedded study. The control group is therefore historical rather than a concurrent randomized group. Although an intervention and control are described, the non-randomized allocation means that the groups cannot be assumed to be alike apart from intervention exposure.

The investigators planned to recruit participants who had not taken part in the earlier phases from intervention priority census block groups used in Phase II. Recruitment is described as convenience sampling. The stated enrollment is 384. The adverse-event record separately lists 384 participants at risk in the intervention group and 397 in the historical control group; those safety denominators should not be treated automatically as the endpoint analysis populations.

The study start date is 2023-04-03, and the primary completion date is 2024-12-12. The record says the study is completed. Its broad purpose includes examination of COVID-19 testing and vaccination, using community health worker interactions alongside digitally delivered intervention content. However, the primary endpoint recorded for this registry entry is a baseline self-reported testing count, not a reported post-intervention contrast.

The detailed protocol, the rationale for the chosen enrollment, and any prespecified statistical analysis plan are not reported in the registry record. In particular, it is not reported in the registry record how the historical data were aligned with the newly recruited intervention participants for a formal comparison, or how potential differences between the study phases were addressed. The non-randomized design and historical control structure are important context for interpreting any comparison; they do not by themselves establish whether an intervention effect exists.

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: Under the protocol, historical controls were not considered enrolled in this study, so they are not included in the enrollment number.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Data were only collected from participants who answered the question.

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 Participants Who Self-Reported Use of a COVID-19 Antigen or PCR Test, measured over Baseline.

The primary endpoint is the number of participants who self-reported using a COVID-19 antigen test or PCR test in the past 60 days. It is expressed as a count of participants, and the listed timeframe is baseline. The registry says the information was collected only from participants who answered the question. This describes self-reported testing behavior, rather than a laboratory-confirmed testing event or a measure of testing frequency.

A count can summarize how many respondents reported the behavior, which is a direct and accessible way to describe uptake in a study focused on testing behavior. It does not, by itself, show what proportion of all enrolled participants tested, because the endpoint population is restricted to respondents and the number who answered is not reported in the registry record. Nor does a baseline count establish a change after education or a difference from the historical control group.

The endpoint may be relevant to the study’s stated interest in COVID-19 testing, but its timing matters. A baseline measurement precedes the intervention’s potential influence on participants. To evaluate change or a comparative effect, analysis would need to connect outcome measurements at suitable timepoints across groups, with a clear account of who contributed data and how the groups were compared. Such comparative results are not reported in the registry record. The endpoint’s measurement details beyond the question description, including any validation of self-report, are also not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry classification identifies the study as non-randomized, unmasked, and single-group, while also describing a historical control group. These labels convey different aspects of the design. Non-randomized means participants were not assigned by random allocation as described in the record. Unmasked means masking is listed as none. A historical control is a comparator whose data come from an earlier study period, rather than a concurrently recruited randomized group. These features affect how well observed differences can be attributed specifically to the intervention.

What does a count endpoint represent?

The primary outcome is a count of participants self-reporting use of an antigen or PCR test within the past 60 days. A count describes the number of respondents with the reported behavior. It is not interchangeable with a rate or proportion unless the relevant denominator is also specified. Here, the record says that data were collected only from participants who answered the question, but does not report the number of respondents or the outcome count. Thus the endpoint definition is available, while its observed value and the completeness of responses are not reported in the registry record.

For a count outcome, the analysis approach depends on how the data are distributed and on the exact question being estimated. A simple count can be described directly; a comparison may require a model or a clearly defined difference in counts, potentially accounting for the number of people who could contribute data. No statistical model or inferential method for this endpoint is reported in the registry record. The registry’s classification of the endpoint as counts does not establish which model was used.

Why does the historical comparison matter?

Random allocation helps balance measured and unmeasured characteristics between study groups on average. That protection is absent in a non-randomized design. With historical controls, differences may reflect changes over time, recruitment, participant characteristics, survey administration, or other contextual factors, as well as any intervention-associated difference. These are general design concerns, not findings about this particular study. The registry record does not report whether such factors were measured or adjusted for.

Unmasked participation can also matter for self-reported outcomes. Participants know whether they received the educational content, and their awareness could influence reporting or behavior. That possibility does not show that reporting was biased; it is a general consideration when interpreting self-reported outcomes in an unmasked study. Details about outcome validation and procedures intended to limit reporting bias are not reported in the registry record.

How should confidence intervals and p-values be read?

In general, a confidence interval describes the range of estimates compatible with the data and the statistical method under its assumptions. A narrower interval indicates greater precision; a wider interval indicates greater uncertainty. A p-value, in general, describes how compatible the observed data are with a specified null model. Neither measure establishes that an effect is important, nor does a p-value prove that a result is true or false. No confidence intervals or p-values for the primary endpoint are reported in the registry record, so precision and statistical evidence cannot be assessed.

The record includes a planning illustration framework that uses a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It describes planning for a stated fraction of an observed effect. These are planning assumptions, not evidence that the completed study used that analysis plan. The sample-size justification for the trial and the statistical plan actually used are not reported in the registry record. A planning calculation based on an observed effect is not observed power and does not substitute for an estimate with its uncertainty.

What is missing from the analysis record?

The registry lists 0 analyses and reports no primary outcome estimate. It therefore does not identify an analysis population, an adjustment strategy, a treatment contrast, or the handling of missing endpoint responses. The planned enrollment and the safety denominators cannot fill these gaps. Any valid interpretation of a reported analysis would require clarity about participants included, data available at each timepoint, and how the historic comparator was made comparable. Those details are not reported in the registry record.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Text Message-based Education Intervention", "Historical Control Group")
n      <- c(348, 395)
events <- c(31, 22)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Text Message-based Education Intervention  348  31  8.9  6.1  12.4  6.3  12.4  6.4  12.4
ARM  Historical Control Group  395  22  5.6  3.5  8.3  3.7  8.3  3.7  8.3

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 5 outcome measures and names secondary outcomes that include self-reported receipt of COVID-19 vaccination and use of mitigation measures after a positive test. The vaccination outcome appears twice in the listed secondary outcomes. The record does not explain whether this repetition represents distinct measures, separate timepoints, or a duplicate listing. No results for these secondary outcomes are reported in the registry record.

When a study evaluates multiple outcomes or multiple comparisons, the chance of at least one apparently notable result can increase. General approaches to multiplicity include prioritizing a primary outcome, prespecifying a testing sequence, or adjusting statistical thresholds. The statistical plan for this study, including any approach to multiplicity, is not reported in the registry record. It is therefore not possible to determine how the listed outcomes were prioritized or whether any adjustment was planned or used.

The primary outcome population consists only of participants who answered the question. The number of answers, the number of missing responses, reasons for nonresponse, and methods for handling missing data are not reported in the registry record. Missingness can affect interpretation if the people who respond differ systematically from those who do not. This is a general concern; the record does not establish whether such differences occurred in this study.

The analysis populations are also not described. Enrollment is reported as 384, and the safety record lists group denominators, but the record does not state how many participants were included in each endpoint analysis. It does not report whether all eligible participants were analyzed, whether any exclusions occurred, or how historical records were matched or aligned with the new participants. Without these details, the endpoint’s denominator and the handling of incomplete data cannot be assessed.

12. 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 384 participants at risk in this discussion message-based education intervention group and 397 in the historical control group. It reports 0 serious adverse events in each group, 0 other adverse events in each group, and 0 deaths in each group. The registry does not list serious adverse event terms.

These figures should be read as the adverse-event entries recorded for the groups, not as proof that no adverse experiences occurred in any broader sense. The record does not report the procedures used to collect events, the period over which participants were monitored, or whether adverse-event reporting was comparable between the new intervention group and historical controls. Those details matter when interpreting zero recorded events, especially when groups come from different study periods and contexts.

The safety denominators are not necessarily the same as the primary endpoint analysis population. The primary outcome is collected only from people who answered the testing question, and the number of respondents is not reported in the registry record. The adverse-event counts also do not provide information about intervention effectiveness or testing behavior. They are a separate set of descriptive safety entries and should not be used as a substitute for the missing primary outcome analysis.

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

The registry record establishes that investigators studied a text message-based education intervention supported by community health workers and described a Phase II historical control group. It identifies self-reported antigen or PCR testing in the past 60 days at baseline as the primary endpoint. It also records enrollment of 384 and lists the study as completed. These details describe the study’s setting, intervention concept, and outcome definition.

The record does not establish whether the intervention changed testing behavior. It provides no primary endpoint count, comparative estimate, confidence interval, or p-value, and lists 0 analyses. Because the endpoint is measured at baseline, it is not itself a post-intervention result. A baseline measure cannot show a response to the intervention without later outcomes and a suitable analysis.

The historical control design further limits what could be concluded from a simple difference between groups. The groups were not assigned randomly, and the comparator comes from an earlier phase. Any observed difference could reflect group composition, timing, measurement, or other contextual differences, in addition to an intervention-associated difference. The record does not report an adjustment strategy that would clarify these sources of variation.

Consequently, neither a positive nor a negative effectiveness conclusion is warranted from the posted registry results. The absence of a posted analysis is not evidence of no effect; it is an absence of results needed to estimate and evaluate an effect. The same caution applies to the listed vaccination and mitigation outcomes. Statistical interpretation would require outcome values, an identified analysis population, the comparison method, and measures of uncertainty. Those results and methods are not reported in the registry record.

14. Why This Trial Matters Statistically

This study raises an important statistical question about how to evaluate interventions delivered through community health worker interactions and digital content when the comparator comes from an earlier study phase. Such designs may be used when a concurrent randomized control is not part of the study approach, but interpretation then depends on careful alignment of populations, outcome definitions, and measurement periods. The registry record describes the historical-control structure but does not report how these issues were handled analytically.

The self-reported testing endpoint also highlights the distinction between an outcome’s definition and evidence about its value. Knowing that an outcome counts respondents who reported antigen or PCR testing in the past 60 days clarifies what is intended to be measured. Without the observed count, respondent denominator, and an analysis, it does not show the level of testing or whether groups differed. This distinction is central to reading clinical registries: a completed study and a detailed outcome label do not by themselves provide an interpretable treatment-effect estimate.

From a statistical perspective, the record also illustrates why uncertainty and analysis populations matter. A comparison needs an explicit estimate and an account of who contributed data; precision needs an uncertainty measure; missing responses need transparent handling. Because the primary outcome is restricted to respondents and the analysis results are absent, those elements cannot be evaluated. The listed adverse-event entries do not answer the efficacy question, and planning assumptions do not replace observed comparative results.

The trial’s stated aim addresses testing and vaccination in vulnerable populations, but the registry results available for interpretation do not quantify outcomes for those aims. The methodological significance is therefore in the design questions and reporting needs the study presents, rather than in a demonstrated statistical effect.

15. 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 most direct limitation for interpreting the registry entry is that no primary endpoint result or statistical analysis is reported. The record identifies a baseline self-report count, but not its observed value, respondent denominator, uncertainty, or comparison with historical controls. As a result, the testing outcome cannot be evaluated for magnitude or precision.

The endpoint is collected only from participants who answered the question. The number of respondents and the handling of missing responses are not reported in the registry record. This makes it impossible to assess how much of the enrolled group contributed to the endpoint or whether nonresponse could have affected the observed data.

Recruitment is described as convenience sampling, and allocation is non-randomized. The historical control participants came from Phase II, whereas the intervention group was recruited for the later phase. The record does not report how group differences, time-related changes, or differences in surveys and study context were addressed. These features limit causal interpretation of any eventual group contrast.

The primary outcome’s timeframe is baseline, which does not by itself provide a post-intervention measure. The registry does not report an analysis connecting baseline and later outcomes or an adjusted comparison with the historical control. The statistical analysis plan, sample-size justification, and details of outcome validation are also not reported in the registry record.

The record’s secondary-outcome list repeats vaccination and does not provide secondary results. It is not reported in the registry record how outcome multiplicity was addressed. Safety entries report zero events, but methods and monitoring periods are not described. These limitations leave both the efficacy interpretation and the scope of safety reporting uncertain.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Examining How a Text Message-based Educational Intervention Influences COVID-19 Testing — NCT07074171.
• Registry record last updated 2026-01-06; status last verified 2025-12.
• Results first posted 2026-01-06.
• 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.

19. Record Summary

NCT07074171 is a completed, non-randomized, unmasked Phase III study of a text message-based COVID-19 education intervention supported by community health workers. The registry describes convenience recruitment from intervention priority census block groups and a historical control group from the Phase II embedded study. Enrollment is 384. The study start date is 2023-04-03, and the primary completion date is 2024-12-12.

The primary endpoint is the number of participants who self-reported using a COVID-19 antigen or PCR test in the past 60 days at baseline. Data were collected only from participants who answered the question. The registry lists 0 analyses and does not report a primary outcome result or comparative effect. Safety entries list 0 serious adverse events, 0 other adverse events, and 0 deaths in both groups. The record’s design and outcome definitions are available, but statistical results needed to evaluate testing behavior are not reported in the registry record.