A See and Treat Paradigm for Cervical Pre-cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A See and Treat Paradigm for Cervical Pre-cancer (NCT02477124), a non-randomized na study. The primary outcome is Number of Participants for Which the Standard of Care Diagnosis and the Diagnosis of the TVDC Image Agree.

2. Clinical Question

3. Trial Design

The registry describes a parallel-model, non-randomized study with no masking. It lists the transvaginal digital colposcope and standard-of-care screening as interventions. Enrollment is recorded as 514. The sponsor is Duke University, the phase is recorded as NA, and the status is terminated. The recorded start is 2015-09 and the primary completion is 2020-03. The reason for termination is not reported in the registry record.

The study summary says the purpose is to assess whether the transvaginal colposcope produces results similar to colposcopy for cervical cancer screening. It describes the device as investigational and says it is not approved by the U.S. Food and Drug Administration. The endpoint description says standard-of-care images and research-device images are available for 514 participants, with 2 photos for each step in the colposcopy procedure process. The timeframe for the primary endpoint is up to approximately 10 minutes.

Although the registry uses a parallel model and lists two interventions, the two arm descriptions both state that participants receive standard-of-care screening and then have images collected with the transvaginal colposcope. The record does not explain how those descriptions relate to separate arms or identify a distinct group receiving standard-of-care screening alone. Its own comparison note cautions that allocation is not randomized and the arms are not a like-for-like control comparison. Accordingly, the design should not be interpreted as a randomized test of the device against an independent control group.

The registry lists one outcome measure and zero analyses. It does not report a statistical analysis plan, sample-size justification for the trial, or a reason for the terminated status. Those omissions limit what can be concluded about the planned and completed analyses.

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

Analysis population for the primary outcome, as defined in the registry: We have SOC Images from these 514 participants AND the images obtained with our research device. Therefore we have 2 photos for each step in the colposcopy procedure process

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 for Which the Standard of Care Diagnosis and the Diagnosis of the TVDC Image Agree, measured over Up to approximately 10 minutes.

The primary endpoint is titled as the number of participants for whom the standard-of-care diagnosis and the diagnosis of the TVDC image agree. Its unit is participants, and its timeframe is up to approximately 10 minutes. In plain terms, it is intended to count instances in which the two diagnoses match. The registry’s population description says that standard-of-care images and research-device images are available for 514 participants and that there are 2 photos for each step in the colposcopy procedure process.

An agreement endpoint is relevant when a new imaging approach is being compared with an existing diagnostic approach. Matching diagnoses can describe concordance between the approaches. Agreement alone, however, does not establish that either diagnosis is correct; that would require an appropriate reference standard and a clearly specified analysis. The registry record does not report such details or the actual number of participants whose diagnoses agreed.

The endpoint is expressed as a count, not as a reported percentage or an effect estimate. The record does not provide the number with agreement, the number with disagreement, an analysis denominator, or uncertainty around the result. Thus, the endpoint definition indicates what was intended to be counted, but the posted results do not quantify the observed agreement.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The posted information supports a limited description of the statistical question, rather than a reconstruction of a full analysis. The primary outcome is a participant count for agreement between standard-of-care diagnosis and diagnosis from a TVDC image. The registry lists one outcome measure and zero analyses. It does not report the statistical plan, a model, a hypothesis test, an agreement coefficient, or an approach to uncertainty. The methods below explain the concepts relevant to interpreting this endpoint in general; they are not claims about procedures used in this study.

What does an agreement count represent?

A count records how many participants meet the endpoint definition. For this study, the endpoint definition concerns a match between the two diagnoses. A count by itself is not an agreement rate: interpreting the extent of agreement also requires knowing the relevant denominator and how participants without complete or assessable paired information were handled. The registry gives an enrollment count and describes images for 514 participants, but it does not post the observed agreement count, disagreement count, or endpoint analysis denominator.

For categorical diagnostic results, an agreement analysis may distinguish exact matching from partial matching, or may assess agreement across categories. Those choices can change the meaning of the result. The registry record does not report the diagnosis categories, a rule for deciding that results agree, or whether agreement was assessed for each procedure step or at another level. It also does not state whether readings were made independently or whether assessors had access to other diagnostic information. Without those details, the intended endpoint is understandable at a broad level, but its operational definition cannot be fully evaluated.

Why distinguish agreement from accuracy?

As a general statistical concept, agreement compares classifications from methods, whereas accuracy compares a method’s classifications with a suitable reference for the underlying condition. Agreement can be informative when evaluating a new imaging method, but it does not by itself show that matching diagnoses are correct. A method could match another method while both share errors. To assess accuracy, an analysis would need a reference standard and a prespecified way to compare results with it. Such a reference-standard analysis is not reported in the registry record.

What would uncertainty add?

A point estimate describes the observed result in the analyzed participants. A confidence interval expresses uncertainty around that estimate under the chosen statistical model and assumptions. For an agreement outcome, uncertainty helps readers judge how precisely the observed concordance is estimated and how well it might represent the population of interest. The posted record contains no agreement estimate and no confidence interval, so neither the level nor precision of agreement can be assessed.

A p-value, when used, addresses compatibility between data and a specified null hypothesis under a statistical model; it does not measure the size or practical importance of an effect. The registry does not report a hypothesis test or p-value. It also does not report whether the intended question was framed as superiority, non-inferiority, or equivalence. A general aim of producing “similar results” is not enough to identify a formal equivalence analysis. Such an analysis ordinarily requires a prespecified margin and a suitable estimate with uncertainty; these details are not reported in the registry record.

What do the design labels imply?

Non-randomized allocation means participants were not assigned at random according to the registry’s design label. In general, non-randomized comparisons may be affected by differences between groups, and statistical adjustment cannot be assumed unless it is described. Here, the registry also says the arms are not a like-for-like control comparison. Its arm descriptions both include standard-of-care screening followed by device image collection, so the record does not clearly define distinct comparison groups. The parallel-model label does not resolve that ambiguity.

Masking is listed as none. In general, lack of masking can matter when people involved in assessment know which method or images they are evaluating. The record does not specify assessor roles or reading procedures, so the possible effect on outcome assessment cannot be determined. The registry’s description of 2 photos for each step in the colposcopy procedure process indicates multiple images, but it does not explain how multiple images were combined into a participant-level diagnosis or count.

How should planning illustrations be read?

The stated sample-size assumptions describe a planning illustration based on the observed effect, not the trial’s own design. They specify a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect to detect expressed as a stated fraction of the observed effect. These assumptions do not establish that the trial was designed or analyzed that way. Any resulting sample-size illustration is a planning quantity, not observed power and not evidence that an effect was detected. The registry record does not report the trial’s own sample-size justification.

In general, planning calculations depend on the assumed effect, variability, analysis method, and design. A calculation based on an observed effect is especially dependent on that estimate and should not be read as a retrospective measure of whether the completed study was adequately powered. No numerical planning outputs are included in the registry record described for this outcome, and the record does not report an observed primary effect to evaluate.

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("Transvaginal Digital Colposcope (TVCD)")
n      <- c(514)
events <- c(514)
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  Transvaginal Digital Colposcope (TVCD)  514  514  100.0  99.3  100.0  99.3  100.0  99.3  100.0

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and zero analyses. That is not enough to determine how many statistical tests, outcome definitions, or subgroups were planned or performed. A multiplicity strategy, such as an adjustment for testing several outcomes, is not reported in the registry record. With no reported primary-outcome analysis, there is also no reported p-value to interpret in relation to multiplicity.

Missing-data handling is not reported in the registry record. The endpoint description says images are available for 514 participants, but it does not state how incomplete images, unreadable images, missing diagnoses, or other unevaluable cases were handled. It does not give an analysis denominator or describe whether any participants were excluded from the endpoint assessment. Therefore, the enrolled count should not automatically be treated as the number included in a primary-outcome analysis.

The analysis population is also not reported in the registry record. There is no description of an intention-to-treat, per-protocol, or other analysis set, nor of rules for assigning participants to such a set. These distinctions can matter because the set analyzed and the handling of incomplete paired information influence what an agreement estimate represents.

As a general statistical point, prespecified outcome definitions, missing-data rules, and analysis populations help readers assess the risk of selective reporting and understand which participants contribute to an estimate. The absence of these details does not prove that particular methods were or were not used. It means their use and consequences cannot be evaluated from the registry record.

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 registry lists a Transvaginal Digital Colposcope (TVCD) group with 514 participants at risk for serious adverse events and 0 affected. It also lists 514 at risk for other adverse events and 0 affected. The recorded deaths are 0, and no top serious adverse-event terms are listed.

These are registry counts, not a comparison of safety between randomized groups. The study is non-randomized, and the record’s comparison note says the arms are not a like-for-like control comparison. The safety entry identifies a device group but does not provide a separate standard-of-care-only safety group for comparison. No rates or uncertainty estimates are reported, and none should be inferred from the counts alone.

A reported count of 0 affected means no events are recorded in that registry category for the listed group; it does not establish that no adverse events occurred outside the scope or completeness of the posted record. The registry does not describe how adverse events were solicited, their observation period, or the completeness of safety follow-up. The endpoint timeframe of up to approximately 10 minutes pertains to the primary outcome and should not be treated as the safety follow-up duration. The safety figures therefore describe what is entered in the record, with limited information about ascertainment and comparison.

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

The study record presents a question about whether diagnoses associated with transvaginal digital colposcope images agree with standard-of-care diagnoses. The primary outcome is defined as a participant count of agreement. However, the posted results do not include the observed count or an analysis. The record therefore does not establish the extent of agreement, its precision, or whether the study met a statistical criterion for similarity.

Even if an agreement result were available, it would need to be interpreted according to the methods used. Agreement is not, by itself, proof of diagnostic accuracy. A reader would need to know how diagnoses were categorized, which participants were included, how incomplete information was handled, whether assessments were independent, and whether a reference standard was used. Those details are not reported in the registry record.

The design also limits what can be concluded from a comparison between the listed interventions. Allocation is non-randomized, and the registry cautions that the arms are not a like-for-like control comparison. Both arm descriptions include standard-of-care screening followed by device image collection. Thus, the record does not support attributing a difference in outcomes to the device relative to an independent control group.

The record reports enrollment of 514 and a terminated status, but does not report why the study was terminated or whether termination affected the availability or completeness of outcome data. It reports no primary endpoint analysis. The posted safety counts are zero affected in the listed categories, but the registry does not describe event ascertainment or follow-up sufficiently to interpret those counts as a comprehensive safety assessment.

In short, the registry establishes the intended endpoint and some design and enrollment details, but not the primary result needed to answer the central agreement question. It does not establish that the device is accurate, equivalent, superior, or inferior to standard-of-care colposcopy. Those conclusions would require results and methods that are not reported in the registry record.

14. Why This Trial Matters Statistically

This trial is statistically notable because its central question concerns diagnostic agreement, while its registry record does not include the agreement result. The distinction matters: an outcome definition tells readers what investigators intended to measure, but only an observed estimate and its uncertainty show what the data indicate. Without those results, the statistical question remains unanswered in the posted record.

The design labels and arm descriptions also illustrate why careful reading matters. A parallel model can suggest distinct groups, but the registry describes both arms as receiving standard-of-care screening followed by device imaging. It separately warns that the allocation is non-randomized and that the arms are not a like-for-like control comparison. Those qualifications prevent a simple causal interpretation of a device-versus-control contrast.

More generally, agreement should not be mistaken for accuracy. Matching diagnoses can be a useful measure of concordance, but interpreting diagnostic performance may require an appropriate reference standard and a clearly specified assessment process. The record does not provide those details. The same general principle applies to uncertainty: a count without an analysis denominator or interval cannot show how precisely agreement was estimated.

The status and enrollment information are not substitutes for outcome reporting. Enrollment of 514 does not reveal how many participants contributed evaluable paired assessments, and terminated status alone does not explain why results are absent. For biostatisticians and clinical researchers, the record highlights the value of reporting the endpoint definition alongside the actual result, analysis population, missing-data handling, uncertainty, and assessment procedures. Those elements are necessary for readers to judge what a diagnostic-comparison study can support.

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.

Several limitations are specific to the registry description. The intervention is named transvaginal digital colposcope (TVDC), while the arm title uses TVCD. The two arm descriptions both state that standard-of-care screening is followed by device image collection. The record does not explain whether these are genuinely distinct groups or how their descriptions should be reconciled. It also explicitly cautions that the study is non-randomized and that the arms are not a like-for-like control comparison.

The primary endpoint is defined as a count of participants whose diagnoses agree, but the actual count, disagreement count, analysis denominator, and statistical estimate are not reported. The record does not state the diagnosis categories or operational rule for agreement. Although it describes images for 514 participants and 2 photos for each step in the procedure process, it does not explain how image-level information was converted into a participant-level diagnosis or how multiple images were handled.

Masking is listed as none, while roles and reading procedures are not described. The record does not report whether image assessments were independent or how assessors handled discordant readings. It does not report a reference-standard analysis, missing-data rules, analysis population, or statistical analysis plan. These gaps limit assessment of both endpoint validity and potential bias.

The study status is terminated, but the reason is not reported in the registry record. The record provides enrollment and safety counts but no primary endpoint analysis. Its safety information does not describe ascertainment or follow-up, and it does not provide a comparable control group. These limitations mean the registry record supports a description of the study’s stated purpose and design, but not a complete assessment of its diagnostic or safety findings.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A See and Treat Paradigm for Cervical Pre-cancer — NCT02477124.
• Registry record last updated 2023-05-18; status last verified 2023-05.
• Results first posted 2023-05-18.
• 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

NCT02477124 is a Duke University study titled “A See and Treat Paradigm for Cervical Pre-cancer.” The registry records cervical pre-cancer as the condition, enrollment of 514, a non-randomized allocation, no masking, and a parallel model. The study status is terminated; the recorded start is 2015-09 and primary completion is 2020-03. The reason for termination is not reported in the registry record.

The study concerns a transvaginal digital colposcope and standard-of-care screening. Its primary endpoint is the number of participants whose standard-of-care diagnosis agrees with the diagnosis of the TVDC image, with a timeframe of up to approximately 10 minutes. The registry lists one outcome measure and zero analyses. It does not report the primary endpoint result, an analysis denominator, or uncertainty around an estimate.

The registry lists 0 serious adverse events, 0 other adverse events, and 0 deaths among the recorded safety group, with 514 at risk in each adverse-event category. These counts are not a like-for-like randomized safety comparison. The arm descriptions and non-randomized design further limit comparative interpretation. Overall, the record describes the intended diagnostic-agreement question but does not post results that answer it.