Study Evaluating Changes In Mammographic Breast Density: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study Evaluating Changes In Mammographic Breast Density (NCT00774267), a registered phase not stated study. The primary outcome is Percent Change From Baseline in Mammographic Breast Density at Month 24.

2. Clinical Question

The study objective was to evaluate quantitative changes in breast density over the stated follow-up period. The primary endpoint is defined as percent change from baseline in mammographic breast density at Month 24, with a mean reported as the endpoint parameter and standard deviation as its dispersion measure. This frames the question as one about change in an imaging measurement within each group. It does not, by itself, specify a formal contrast between groups or establish that the groups can be compared as if treatment assignment were randomized.

The endpoint description says digitized mammogram pairs were assessed using software by a single trained radiologist, using only the left craniocaudal view. The registry defines the analyzed population as participants with technically acceptable mammograms available at both primary study baseline and Month 24. Numerical outcome estimates, group analysis counts, and inferential results are not reported in the registry record, so the size and precision of any group changes cannot be assessed from this record.

3. Trial Design

This is an ancillary study titled “Study Evaluating Changes In Mammographic Breast Density.” Its official title identifies it as an ancillary study of Protocol 3115A1-303-WW and describes assessment of mammographic breast-density changes associated with bazedoxifene acetate/conjugated estrogens, raloxifene, and placebo in postmenopausal women. Pfizer is listed as the sponsor. The registry lists an enrollment of 507, a start of 2009-01, primary completion in 2010-04, and status as completed.

The interventions listed are BZA 20 mg/CE 0.45 mg, BZA 20 mg/CE 0.625 mg, raloxifene 60 mg, and placebo. The brief summary describes daily dosing and states that the primary endpoint is change in mammographic breast density between baseline and Month 24 for each group. The arm descriptions do not add further information about how participants were assigned or how the groups were constituted.

Allocation is not stated. The registry specifically cautions that this is not a randomized comparison and that the arms are not a like-for-like control comparison. Accordingly, the listed groups should not be treated as a conventional randomized set of treatment and control arms based on this record alone. Details of the underlying protocol’s design, eligibility criteria, masking, allocation procedures, and statistical plan are not reported in the registry record.

The enrollment figure describes the study record, not the number with an evaluable mammogram pair for the primary endpoint. The endpoint population is limited to all available participants with technically acceptable mammograms at baseline and Month 24, and the analyzed count for each group is not reported in the registry record. The record identifies the study as completed but does not report numerical primary endpoint results or 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: All available participants were analyzed. Here, 'N' (number of participants analyzed) signifies those participants who had technically acceptable mammograms available at primary study baseline and Month 24.

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 Percent Change From Baseline in Mammographic Breast Density at Month 24, measured over primary study baseline, Month 24.

The primary endpoint is percent change from baseline in mammographic breast density at Month 24. It is a continuous measurement, and the endpoint parameter is the mean; standard deviation is identified as the measure of dispersion. The stated timeframe is primary study baseline to Month 24. The endpoint description defines the analyzed population as all available participants with technically acceptable mammograms at both time points.

Density was assessed from digitized mammogram pairs. A single trained radiologist used software to determine breast density, and the assessment used only the left craniocaudal view. These details identify the source and method of the measurement, but the registry does not report technical validation information or further image-reading procedures.

General: A continuous imaging measure can describe the amount of change in a target characteristic over time. A mean gives a summary of the group’s measured values, while a standard deviation describes how spread out those values are around the mean. These summaries help characterize the measurement, but they do not alone establish a causal treatment effect or its clinical importance. That interpretation also depends on the study design, the analysis population, uncertainty estimates, and the context for the measure. Numerical endpoint estimates and uncertainty measures are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous variable: percent change from baseline in mammographic breast density at Month 24. The registry identifies the mean as the parameter and standard deviation as the dispersion measure. These summaries would describe the average change and the variability of observed changes within each group. The registry does not report their numerical values.

What does a mean describe?

General: The arithmetic mean combines all measured values into one central summary. It can be useful when the scientific question concerns average change, as it does in the stated endpoint. However, an average can conceal variation among participants. A mean change does not imply that every participant changed by a similar amount or even in the same direction. The distribution of individual measurements and the number analyzed are important context for interpreting the mean.

What does a standard deviation describe?

General: A standard deviation describes how dispersed individual values are around the mean. A larger spread indicates more heterogeneity in the measurements, while a smaller spread indicates values are more concentrated around the mean. It is not the same as the uncertainty in the mean. Precision of the mean depends on both the spread and the number of observations, along with the assumptions used for inference. Because the registry does not report endpoint standard deviations or analyzed counts, precision cannot be assessed from this record.

How should the paired measurement be understood?

The endpoint population is defined by availability of technically acceptable mammograms at baseline and Month 24. General: When measurements are collected at two times for the same participant, change can be calculated within that participant and then summarized across participants. Pairing can account for stable differences between people, but the details of the actual calculation and model matter. The registry names percent change from baseline but does not report the computational formula, handling of unusual values, or any adjustment for other variables. Those details are not reported in the registry record.

What do uncertainty estimates add?

General: A confidence interval communicates the uncertainty around an estimate under a statistical model. It can convey a range of values compatible with the observed data and assumptions. A p-value, when used, evaluates how unusual data at least as extreme as those observed would be under a specified null model. Neither measure gives the probability that a treatment works, and neither substitutes for examining the design, the endpoint, and the size of the estimated change. The registry record does not report confidence intervals, p-values, or other inferential results for the primary endpoint.

What is the role of the planning illustrations?

The stated sample-size illustrations are based on the observed effect and are explicitly described as planning illustrations rather than the study’s own design. Their stated assumptions include a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the target effect is expressed as a stated fraction of the observed effect. For non-time-to-event outcomes, the planning quantity is participants per arm for 80% power or another specified power level where listed. General: Such calculations describe a hypothetical future study under chosen assumptions. They are sensitive to the assumed effect, variability, allocation, and analysis method. They are not observed power, do not describe the precision achieved in this study, and should not be confused with the number analyzed. The registry does not state that these assumptions governed the study’s actual design.

The registry does not report the model used for the endpoint, any covariate adjustment, procedures for checking assumptions, or a statistical analysis plan. Those omissions limit interpretation beyond the endpoint definition and measurement description.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Bazedoxifene 20 mg/Conjugated Estrogen 0.45 mg", "Bazedoxifene 20 mg/Conjugated Estrogen 0.625 mg", "Raloxifene 60 mg", "Placebo")
n   <- c(129, 105, 125, 126)
m   <- c(-0.39, -0.05, -0.23, -0.42)
sp  <- c(1.75, 1.68, 1.76, 1.72)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Bazedoxifene 20 mg/Conjugated Estrogen 0.45 mg  129  -0.390  1.750  -0.695  -0.085
CARM  Bazedoxifene 20 mg/Conjugated Estrogen 0.625 mg  105  -0.050  1.680  -0.375  0.275
CARM  Raloxifene 60 mg  125  -0.230  1.760  -0.542  0.082
CARM  Placebo  126  -0.420  1.720  -0.723  -0.117

11. Multiplicity, Missing Data and Analysis Populations

The primary endpoint is described as a change in breast density for each group. Four intervention groups are listed, but the registry record does not specify which between-group contrasts, if any, were intended as primary comparisons. It also does not report a multiplicity strategy, such as adjustment for multiple comparisons or a hierarchy among tests. When several groups or outcomes are compared, the chance of at least one false-positive finding can increase unless the analysis plan accounts for the number and structure of tests. General: That concern applies to interpretation of multiple inferences; it does not establish that any particular method was required or used in this study.

The endpoint population is defined as all available participants with technically acceptable mammograms at baseline and Month 24. The number analyzed in each group and the reasons that other enrolled participants may not have contributed endpoint data are not reported in the registry record. Methods for handling missing mammograms, technically unacceptable images, or other unavailable data are not reported in the registry record. The record does not describe imputation, sensitivity analyses, or an assessment of whether missingness could affect the results.

The endpoint description gives an analysis-population rule, but does not provide the actual population counts or a fuller analysis-set definition. Whether the analysis included all assigned participants, only those with evaluable images, or another specified set beyond the stated availability criterion is not reported in the registry record. A statistical analysis plan, interim-analysis details, and any prespecified subgroup analyses are not reported in the registry record. The reported enrollment therefore cannot be substituted for the number contributing to the endpoint summaries.

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

The registry record establishes that the study aimed to measure percent change in mammographic breast density from baseline to Month 24 among postmenopausal women in four listed groups. It also describes the image-reading method and defines which participants could contribute to the primary endpoint. Those details explain the intended measurement, not the observed result.

Numerical changes, group-specific analysis counts, and measures of statistical uncertainty are not reported in the registry record. Therefore, the record cannot establish how much breast density changed, how variable the changes were, or how precisely any group mean was estimated. It also cannot establish whether group results differed or whether any difference was statistically distinguishable from random variation.

The listed interventions and placebo group do not, by themselves, demonstrate a valid comparative treatment effect. Allocation is not stated, and the registry explicitly cautions that the arms are not a like-for-like control comparison. No conclusions about comparative efficacy or causation can be drawn from the record’s endpoint description alone.

General: A measured change in an imaging characteristic is not automatically equivalent to a change in a clinical outcome. Its interpretation depends on what the measure captures, how reliably it is measured, and how it relates to outcomes important to participants. The registry record does not report evidence linking the measured changes in this study to clinical outcomes, nor does it provide a threshold for interpreting the importance of a particular change.

The completed status means the record lists the study as completed; it does not supply the missing outcome findings. The information available supports an account of the research question and measurement approach, while leaving the results and their implications undetermined.

13. Why This Trial Matters Statistically

This study is statistically informative as an example of how an endpoint definition can be more detailed than the results reported in a registry record. It specifies a continuous outcome, the time points of interest, the summary parameter, the dispersion measure, the imaging view, and the image-reading approach. Those elements help readers understand what the researchers intended to quantify, even though numerical findings are not reported.

The population rule also matters. The primary endpoint uses available participants with technically acceptable mammograms at both baseline and Month 24, rather than simply identifying all enrolled participants as analyzed. Without the group-specific counts and information about unavailable images, the relationship between enrollment and the endpoint analysis remains unclear. This distinction is important whenever an outcome requires usable measurements at multiple time points.

The registry’s caution about allocation is equally important. A list of several interventions and a placebo group can invite a direct comparison, but a comparative interpretation requires information about how groups were formed and how contrasts were analyzed. The record says allocation is not stated and the arms are not a like-for-like control comparison, so the listed groups should not be read as a randomized treatment comparison.

The record also highlights the difference between study results and planning calculations. Planning illustrations based on an observed effect can describe hypothetical sample-size requirements under specified assumptions; they are not evidence that the study achieved a particular precision or had a particular observed power. Separating those concepts helps prevent planning quantities from being mistaken for empirical findings.

Overall, the record allows a careful reading of the intended measurement and its limitations, while showing why transparent outcome estimates, analysis counts, uncertainty measures, and design details are necessary for statistical interpretation.

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.

A central limitation of the available record is that it does not report numerical primary endpoint results. Group means, standard deviations, analysis counts, and inferential statistics are not reported in the registry record. As a result, neither the magnitude nor the precision of the breast-density changes can be evaluated.

The analysis population depends on technically acceptable mammograms at baseline and Month 24, but the record does not report how many participants met that criterion in each group. It also does not explain the availability or acceptability of images for participants who did not contribute to the endpoint. The extent and possible consequences of missing endpoint data are therefore unclear.

Measurement information is limited to digitized mammogram pairs assessed by a single trained radiologist using software and only the left craniocaudal view. Additional information about image acquisition, software procedures, reproducibility, quality control, and handling of technical failures is not reported in the registry record. The record also does not state the formula used for percent change or describe the endpoint’s statistical model.

Design information needed for comparative interpretation is limited. Allocation is not stated, and the registry cautions that the arms are not a like-for-like control comparison. Details about eligibility, group formation, masking, prespecified comparisons, covariate adjustment, and the statistical analysis plan are not reported in the registry record. Multiplicity procedures and sensitivity analyses are also not reported.

These limitations concern what can be learned from the registry record; they do not establish what may be available in other study materials. On this record alone, the study’s measurement objective is clear, but its numerical findings and comparative interpretation remain unavailable.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Study Evaluating Changes In Mammographic Breast Density — NCT00774267.
• Registry record last updated 2014-04-08; status last verified 2014-03.
• Results first posted 2013-12-20.
• 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

The registry describes a completed ancillary study of changes in mammographic breast density among postmenopausal women. Pfizer is the listed sponsor, and the record lists enrollment of 507. The study evaluates daily BZA 20 mg/CE 0.45 mg, daily BZA 20 mg/CE 0.625 mg, raloxifene 60 mg, and placebo. The primary endpoint is percent change from baseline in mammographic breast density at Month 24, summarized by a mean and standard deviation.

Digitized mammogram pairs were assessed by a single trained radiologist using software, with only the left craniocaudal view used. The endpoint population is defined as participants with technically acceptable mammograms at baseline and Month 24. Allocation is not stated, and the registry cautions that the arms are not a like-for-like control comparison. Numerical endpoint results, group analysis counts, uncertainty estimates, and details of the statistical plan are not reported in the registry record.