This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Assessment of Periodic Screening of Women With Denser Breast Using WBUS and DBT (NCT02643966), a na na study. The primary outcome is Cancer Yield Per 1,000.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02643966 |
| Phase | NA |
| Condition | Breast Cancer |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 6266 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Wendie Berg |
| Results posted | Yes |
2. Clinical Question
Population
Women with dense breasts in a screening population. The study description concerns technologist-performed whole breast ultrasound for breast cancer screening. The registry record does not provide further eligibility criteria or a fuller description of the population included in each analysis.
Intervention
Whole breast ultrasound. The arm description states that all women receive both 3D mammography and whole breast ultrasound. The interpretation order varies for each of two radiologists.
Comparator
There is no separate control arm described. Because women receive both imaging approaches, the record does not describe a like-for-like comparison between a group receiving ultrasound and a group not receiving it. The study is listed as single-group, with allocation not applicable.
Primary question
The primary outcome is cancer yield per 1,000 over 3 Years. The record says results from the first radiologist observer are presented and identifies Screens 2 & 3 as including women analyzable after the end of Screening Round 2. The numerical yield and its 95% confidence interval are not reported in the registry record.
In practical terms, the study asks about cancers detected in a dense-breast screening setting where ultrasound is used alongside 3D mammography. It is not described as a randomized test of ultrasound against mammography alone: both approaches are included in the arm description, and there is no separate control group. The listed primary outcome is a detection yield, not a direct measure of long-term health outcomes or a comparison of harms and benefits. The record also lists a secondary outcome concerning types of abnormalities detected by ultrasound and tomosynthesis, but does not give its findings. Interpretation should therefore focus on what the stated endpoint is intended to describe and on the limits imposed by the missing numerical results.
3. Trial Design
The registry record identifies NCT02643966 as a completed study with a single-group model. Allocation is listed as not applicable, and masking is listed as none. The intervention is whole breast ultrasound, described as a device. The arm description says all women receive both 3D mammography and whole breast ultrasound for screening. It also says that the order of interpretation varies for each of two radiologists. The record does not state that this order was randomized, so variation in order should not be described as random assignment.
The listed enrollment is 6266. The study started in 2015-12, reached primary completion in 2022-05, and has completed status. These registry fields describe the study timeline and enrollment; they do not establish how many participants contributed to the primary analysis. For Screens 2 & 3, the outcome description specifies women who were analyzable after the end of Screening Round 2. The number included in that analysis, and the reasons other participants may not have been analyzable, are not reported in the registry record.
Because the study is single-group and includes both imaging approaches, it does not provide a parallel control group for estimating a treatment contrast. The recorded primary endpoint is cancer yield per 1,000, with a 3 Years timeframe and results described for the first radiologist observer. The registry record does not explain how the analysis accounts for any dependence between screening observations, or how results from the radiologists are combined or compared. Those details should not be inferred from the fact that the interpretation order varies.
The sponsor is Wendie Berg. The listed condition is breast cancer. The study description says its purpose is to evaluate the role of technologist-performed whole breast ultrasound for detecting breast cancer in the screening population of women with dense breasts. That statement describes the research aim, rather than a finding. The numerical primary results 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.
First Screening Round
| Stage | Whole Breast Ultrasound |
|---|---|
| STARTED | 6266 |
| COMPLETED | 6178 |
| NOT COMPLETED | 88 |
Reasons for not completing this period:
| Reason | Whole Breast Ultrasound |
|---|---|
| ineligible | 1 |
| excluded from analysis | 2 |
| Lost to Follow-up | 56 |
| Withdrawal by Subject | 28 |
| Death | 1 |
Between First and Second Screening
| Stage | Whole Breast Ultrasound |
|---|---|
| STARTED | 6178 |
| COMPLETED | 6020 |
| NOT COMPLETED | 158 |
Reasons for not completing this period:
| Reason | Whole Breast Ultrasound |
|---|---|
| Lost to Follow-up | 65 |
| Withdrawal by Subject | 82 |
| women who skipped year 2 exam | 11 |
Second Screening Round
| Stage | Whole Breast Ultrasound |
|---|---|
| STARTED | 6020 |
| COMPLETED | 5937 |
| NOT COMPLETED | 83 |
Reasons for not completing this period:
| Reason | Whole Breast Ultrasound |
|---|---|
| Withdrawal by Subject | 20 |
| Lost to Follow-up | 57 |
| Physician Decision | 2 |
| Death | 4 |
Between Second and Third Screening
| Stage | Whole Breast Ultrasound |
|---|---|
| STARTED | 5937 |
| COMPLETED | 5680 |
| NOT COMPLETED | 257 |
Reasons for not completing this period:
| Reason | Whole Breast Ultrasound |
|---|---|
| Lost to Follow-up | 75 |
| Death | 1 |
| Withdrawal by Subject | 181 |
Analysis population for the primary outcome, as defined in the registry: Screens 2 \& 3 include women who were analyzable after the end of Screening Round 2
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: After exclusions, we analyzed the results from 6,179 women overall
| Characteristic | Unit | Whole Breast Ultrasound |
|---|---|---|
| Participants analysed | 6179 | |
| Age, Continuous | years | 54.8 (SD 8.8) |
| Sex: Female, Male: Female | Participants | 6179 |
| Sex: Female, Male: Male | Participants | 0 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 19 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 6160 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 2 |
| Race (NIH/OMB): Asian | Participants | 49 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 2 |
| Race (NIH/OMB): Black or African American | Participants | 106 |
| Race (NIH/OMB): White | Participants | 5991 |
| Race (NIH/OMB): More than one race | Participants | 29 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 |
| Region of Enrollment: United States | Participants | 6179 |
| Breast Density: scattered | Participants | 203 |
| Breast Density: heterogeneously dense | Participants | 4989 |
| Breast Density: extremely dense | Participants | 987 |
6. Primary Endpoint
The registry lists the primary outcome as Cancer Yield Per 1,000, measured over 3 Years.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | Cancer yield per 1,000 |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is cancer yield per 1,000 over 3 Years. Yield expresses cancer detection in relation to a screening denominator, which can make detection results easier to interpret than a raw count alone. The record gives the unit as cancer yield per 1,000 and identifies a 95% confidence interval as the measure of dispersion. It does not provide the numerical yield, confidence limits, or further detail about the denominator and its construction; those details are not reported in the registry record.
A yield endpoint is used to describe detection within the screening process. It does not by itself establish whether detected cancers would otherwise have become clinically apparent, whether screening changes later health outcomes, or how the imaging approaches compare causally. Those questions require additional outcome definitions and analyses. The registry record lists a 3 Years timeframe, but does not clarify how that period maps to the observations contributing to the estimate.
The outcome description says that results from the first radiologist observer are presented. It also specifies that Screens 2 & 3 include women analyzable after the end of Screening Round 2. This is important context for understanding which observations the reported endpoint is intended to cover. The endpoint estimate and its confidence interval are not reported in the registry record, so its magnitude and precision cannot be assessed from the available results information.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies a primary outcome, but does not report its numerical result. It therefore does not support a statement about the size of the cancer yield or the precision of that estimate. The listed measure is cancer yield per 1,000, with a 95% confidence interval, over 3 Years. Without the estimate and interval, readers cannot assess whether the observed yield was high or low in the study population, or whether the data support a narrow or wide range of plausible values.
In general, a confidence interval conveys uncertainty around an estimate under the statistical method used. A narrower interval indicates greater precision under that method; a wider interval indicates less precision. The interval does not correct for biases in study design, outcome ascertainment, or analysis. In this case, the interval limits themselves are not reported in the registry record, and the method used to construct them is not described. Precision therefore cannot be judged from the registry entry.
The design is also central to interpretation. The study is listed as single-group, and its arm description states that all women receive whole breast ultrasound and 3D mammography. There is no separate control arm described. The record therefore does not provide a between-group estimate of the effect of adding ultrasound, nor does it describe a randomized comparison. The reported cancer yield, if available, would describe detection in the stated screening context; it would not by itself isolate the contribution of either imaging approach.
The outcome description narrows the intended result to the first radiologist observer and identifies Screens 2 & 3 among women analyzable after the end of Screening Round 2. This limits any generalization to results from other observers, rounds, or populations unless corresponding information is reported. The number analyzed and the handling of observations that were not analyzable are not reported in the registry record. The record also does not give the result for the listed secondary outcome on types of abnormalities detected by ultrasound and tomosynthesis.
These points do not make the study aim unimportant; they define what can and cannot be concluded from its registry results. The record describes a screening study and its endpoint, but the missing outcome estimate prevents quantitative interpretation. No claim that ultrasound increased or decreased detection, or that one imaging approach performed better than another, can be drawn from the registry information available.
9. Statistical Methods Explained
The registry record describes a count-type outcome, cancer yield per 1,000, and gives a 95% confidence interval as the dispersion measure. It does not report the estimate, the interval limits, or the detailed statistical method used. The concepts below explain how such measures are generally interpreted; they are not a description of an unreported analysis plan.
What does a yield measure?
In general, a yield expresses the number of events detected relative to a defined amount of screening activity. A rate or yield needs a clear numerator and denominator. The registry gives the unit per 1,000, but does not specify the denominator construction, the event-counting rules, or whether the denominator is based on participants, examinations, or another unit. Those details are not reported in the registry record. Without them, readers cannot independently reconstruct the estimate or determine precisely what population of screening activity it represents.
The listed primary outcome has a 3 Years timeframe. A timeframe tells readers the period associated with outcome measurement, but it does not alone explain the observation schedule, the treatment of repeated screening, or whether every participant contributed the same follow-up. Those analytic details are not reported in the registry record. They matter because a rate can be interpreted differently depending on how observation time and repeated measurements are handled.
How should a confidence interval be read?
General explanation: a confidence interval is a range calculated from data using a statistical procedure. It describes the uncertainty of an estimate under that procedure and its assumptions. It is not a range containing every individual outcome, and it is not a guarantee that the true value lies within the particular interval. Precision is assessed by the interval's width, while the estimate indicates the central result. For this study, the registry specifies a 95% confidence interval but does not report the estimate or limits. Neither the result's precision nor its compatibility with alternative values can therefore be evaluated.
Confidence intervals also address sampling uncertainty, not every possible source of error. They cannot by themselves account for selection into analysis, inconsistent measurement, missing data, or limitations in study design. A precise estimate from a single-group study would still describe that study's observed screening context; it would not establish a causal contrast with a separate control group. Here, the numerical interval and its method are unavailable in the registry record.
Why can count data need special care?
General explanation: counts are non-negative and can have distributions unlike measurements that vary continuously around a mean. A count model may describe expected event counts, while a rate model relates counts to an exposure denominator. Which approach is suitable depends on how the outcome and observation process are defined. The registry's label and unit do not identify the exact model used.
Over-dispersion is a general issue in count analysis. It occurs when the observed variation is greater than a model assumes. If a model understates variability, standard errors may be too small and confidence intervals may be too narrow. Researchers can use methods that allow extra variation when justified by the data and design. The registry record does not report whether over-dispersion was assessed or how it was handled. It is therefore not possible to say that it affected this analysis.
Repeated screening observations can also be related within a participant, and results from readers can be correlated when they interpret the same examinations. General statistical methods may account for such dependence when it is present. The record says the order of interpretation varies for each of two radiologists and that primary results are from the first radiologist observer. It does not describe a correlation structure or an adjustment for dependence. No particular approach should be assumed.
What does the planning illustration mean?
The registry information describes a planning illustration based on the observed effect, rather than the study's own design. It specifies a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the target effect expressed as a stated fraction of the observed effect. These are assumptions for an illustration, not evidence that the study was designed or analyzed as a two-arm trial. No corresponding participant-per-arm values or target-effect fraction are reported in the registry record.
A sample-size calculation estimates the amount of information a hypothetical design might require under stated assumptions. It is not an observed study result and does not tell readers whether the completed study had a particular probability of detecting an effect. The illustration should not be called observed power. The study itself is listed as single-group, and the registry record does not report a two-arm design or a sample-size justification. For time-to-event calculations, the registry note describes totals of events using the Schoenfeld method; this does not establish that the primary cancer-yield outcome was analyzed as time-to-event data.
10. Multiplicity, Missing Data and Analysis Populations
The registry record lists four outcome measures and six analyses. It does not report a multiplicity strategy, such as a prespecified hierarchy, adjustment to control the chance of false-positive findings across analyses, or designation of which analyses are confirmatory. Without that information, a reader cannot determine how the collection of analyses was intended to support formal statistical claims. The counts alone do not establish that every analysis tested a separate hypothesis or that every analysis received equal emphasis.
The primary outcome is identified as cancer yield per 1,000, and a secondary outcome concerns types of abnormalities detected by ultrasound and tomosynthesis. The numerical findings for these outcomes are not reported in the registry record. It also does not describe how the six analyses relate to the four outcome measures, or how multiple readers, screening rounds, or analytic populations were handled. Those relationships should not be guessed.
The primary outcome description says that Screens 2 & 3 include women who were analyzable after the end of Screening Round 2. This identifies an analysis population in broad terms, but the number analyzed, the criteria for being analyzable, and the flow from enrollment to analysis are not reported in the registry record. Missing-data methods, including any imputation or sensitivity analyses, are also not reported. The record does not say whether missing observations were excluded, treated in another way, or considered in an assessment of bias.
These gaps limit assessment of both multiplicity and selection into analysis. They do not prove that the investigators used no such methods; they mean the registry record does not describe them. A complete interpretation would require the analysis methods and population definitions, together with the numerical results for the stated outcomes.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Whole Breast Ultrasound | 0 | 6266 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Whole Breast Ultrasound | 6 of 6266 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Whole Breast Ultrasound | 0 of 6266 |
The registry record lists a Whole Breast Ultrasound group with 6266 at risk for serious adverse events and 0 affected. It also lists 6266 at risk for other adverse events and 0 affected. The record lists 6 deaths. These are the posted counts; the registry entry does not provide further detail about the timing, causes, or relationship of the deaths to screening or study procedures.
At-risk counts describe the group represented in the safety record, but do not establish how the safety population was defined or whether it matches the population analyzed for the primary outcome. The registry record does not describe the collection period, event definitions beyond the serious and other categories, or the method used to determine whether events were related to the intervention. The absence of affected participants in the listed categories should therefore be read as a reported count, not as proof that no event of any kind occurred or that the study ruled out risk.
Safety counts also cannot be compared with another imaging strategy from this record because no separate control group is described. The listed deaths should not be interpreted as caused by or unrelated to the study; that relationship is not reported in the registry record. These safety figures are descriptive and do not replace the missing detail needed to understand event ascertainment or attribution.
12. What the Result Does, and Does Not, Mean
The registry record establishes the study's purpose and broad design: evaluating technologist-performed whole breast ultrasound for breast cancer detection in a screening population of women with dense breasts. It describes use of ultrasound alongside 3D mammography, a single-group model, and a primary endpoint of cancer yield per 1,000 over 3 Years. It also identifies results from the first radiologist observer and a defined analyzable population for Screens 2 & 3.
What the record does not establish is the numerical value of the primary outcome. The yield and its 95% confidence interval are not reported in the registry record. As a result, the size of the observed detection yield and the precision of its estimate cannot be assessed. The registry record also does not report the numerical findings for the listed secondary outcome on types of abnormalities detected by ultrasound and tomosynthesis.
The design does not provide a separate control group. Since the arm description includes both whole breast ultrasound and 3D mammography for all women, the record does not isolate the effect of ultrasound relative to screening without ultrasound. An observed yield, if reported, would describe detection in this study's stated context; it would not by itself show that one approach caused more or fewer cancers to be detected or establish effects on later health outcomes.
Nor does the available information establish how results varied by radiologist, how repeated screening observations were handled, or how missing observations affected analysis. The record states that the order of interpretation varies for each of two radiologists, but does not say that order was randomized. It also does not provide the number included in the primary analysis. The appropriate conclusion is limited: the registry describes a completed screening study and its intended endpoint, while leaving the endpoint estimate and important analysis details unreported.
13. Why This Trial Matters Statistically
This study raises a statistical question that is important in screening research: how should cancer detection be described when participants receive more than one imaging approach and the principal outcome is a yield? A yield can characterize detection in a defined screening setting, but it is not equivalent to a randomized estimate of the benefit of adding one test. That distinction matters because the study is listed as single-group and its arm description includes both whole breast ultrasound and 3D mammography.
The endpoint also focuses attention on the unit of analysis. Cancer yield per 1,000 needs a clearly specified numerator and denominator, and interpretation depends on which screening observations are included. The registry record provides the unit, timeframe, observer context, and an analyzable-population description for Screens 2 & 3. It does not provide the calculation details, analysis count, or primary estimate. These omissions make it difficult to evaluate how broadly the endpoint represents screening activity and how precisely it was estimated.
Reader-related measurement is another relevant feature. The record says interpretation order varies for each of two radiologists and that the presented primary results are from the first observer. This makes the observer context important, while leaving the way reader variation was analyzed unreported. It is not possible to determine from the registry entry whether the findings were consistent across readers or how any dependence was addressed.
The study is therefore statistically instructive even though the numerical result is unavailable: it highlights the difference between describing detection and estimating a comparative effect. A clear interpretation requires both the endpoint estimate and the design context. Without the estimate, confidence interval, and analysis details, readers should not infer the magnitude or comparative meaning of cancer detection from the study's aim alone.
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 central limitation of the registry information is that it describes the primary endpoint but does not report the cancer-yield estimate or its 95% confidence interval. This prevents assessment of magnitude and precision. The definition of the denominator, the event-counting rules, and the method for calculating the interval are also not reported in the registry record. Without these details, the result cannot be reconstructed or fully interpreted.
The single-group design limits comparative interpretation. The arm description says women receive both whole breast ultrasound and 3D mammography, and no separate control group is described. Thus, the registry record does not show a contrast between ultrasound plus mammography and mammography alone. It also does not report whether or how the order of interpretation was assigned, beyond stating that the order varies for each of two radiologists.
The outcome description specifies the first radiologist observer and a population for Screens 2 & 3 that includes women analyzable after the end of Screening Round 2. The registry record does not state how many women contributed to that analysis, how analyzability was defined, or how missing observations were handled. It does not report results from other observers or provide enough information to assess consistency across readers or rounds.
Additional limitations in interpretation arise from the lack of reported analysis methods. Multiplicity handling, the statistical model, any treatment of over-dispersion or within-participant dependence, and the relationship between the listed outcomes and analyses are not reported in the registry record. The safety entry lists counts but does not describe event timing or attribution. These are limitations of the information available in the registry record; they should not be taken as evidence that a particular method was or was not used.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Assessment of Periodic Screening of Women With Denser Breast Using WBUS and DBT — NCT02643966.
- Registry record last updated 2023-06-29; status last verified 2023-06.
- Results first posted 2023-06-29.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT02643966 is a completed, single-group study of whole breast ultrasound in screening women with dense breasts. The arm description states that all women receive whole breast ultrasound and 3D mammography, with interpretation order varying for each of two radiologists. Enrollment is listed as 6266; the study started in 2015-12 and reached primary completion in 2022-05.
The primary endpoint is cancer yield per 1,000 over 3 Years, with a 95% confidence interval. The outcome description identifies the first radiologist observer and specifies an analyzable population for Screens 2 & 3 after the end of Screening Round 2. The numerical endpoint result, interval, and analysis details are not reported in the registry record.
The registry lists a secondary outcome concerning types of abnormalities detected by ultrasound and tomosynthesis. Its safety entry reports 0 serious adverse events, 0 other adverse events, and 6 deaths. Details about the deaths and their relationship to study procedures are not reported.