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 Image Evaluation of Philips Philips MammoDiagnost DR Full Field Digital Mammography System (FFDM) (NCT00999596), a registered phase not stated study. The primary outcome is FFDM (Full Field Digital Mammography) Mammogram Scores.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00999596 |
| Phase | Phase not stated |
| Condition | Breast Cancer |
| Allocation | Not Stated |
| Design model | Not Stated |
| Masking | Not Stated |
| Enrollment | 107 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Philips Healthcare |
| Results posted | Yes |
2. Clinical Question
Population
The registry describes women undergoing screening or diagnostic mammography. Overall enrollment was 107, but the image-rating description concerns 6 mammography image sets from women participating in the study. The registry states that this was not a statistical sample under the FDA FFDM Guideline and that the required participant count changed during the study. The relationship between the enrolled population and the women represented by the image sets is not reported in the registry record.
Intervention
The image sets were mammograms from the Philips Digital System, identified as the Philips MammoDiagnost DR full field digital mammography system. The registry says the images were evaluated using hardcopy film. Two MQSA-certified mammography readers rated each image set pass or fail under a typical MQSA evaluation.
Comparator
The stated purpose refers to comparing FFDM with screen-film mammography for breast cancer detection accuracy. However, the registry describes an FFDM arm and does not provide a screen-film arm or a like-for-like control comparison. Allocation is not stated, and the record explicitly indicates that this was not a randomized comparison. A comparative accuracy estimate cannot be recovered from the described outcome information.
Primary question
The registry’s stated aim is to compare the accuracy of Philips FFDM and screen-film mammography among women undergoing screening or diagnostic mammography. The reported primary measure instead concerns pass/fail scores for FFDM image sets. The record does not report the resulting scores, a screen-film result, or an analysis comparing the two methods. Thus, the registry description establishes what was intended and what was rated, but it does not establish the comparative answer.
3. Trial Design
This is a completed clinical image evaluation study of the Philips MammoDiagnost DR full field digital mammography system. Philips Healthcare is listed as the sponsor. The registry lists a start date of 2010-04, a primary completion date of 2011-01, and completed status. Enrollment is recorded as 107.
The brief summary describes an investigation using hardcopy film and states an aim of comparing the accuracy of FFDM with screen-film mammography in detecting breast cancer. It also notes that, under new FDA guidance for FFDM, comparison of accuracy with screen-film was no longer required. The registry’s listed arm is FFDM, described as mammograms from the Philips Digital System. A screen-film arm, allocation method, and randomized assignment are not described. The registry explicitly notes that this was not a randomized comparison and that the arms are not a like-for-like control comparison.
The image evaluation is described as involving 6 mammography image sets, with 4 images per set. Two MQSA-certified mammography readers rated the image sets pass or fail. The description reports a total of 12 scores. The stated timeframe for the primary outcome is Day 1. The record also says the image sets were not a statistical sample under the FDA FFDM Guideline, which specified a minimum of 6 film sets from participants; it notes that the number of participants required by the FDA changed mid-study.
Details needed to characterize the conduct more fully—including how images were selected, how readers were assigned or masked, and how the image ratings were combined—are not reported in the registry record. The listed enrollment should not be assumed to be the number of participants contributing image sets or to represent the number included in an outcome analysis.
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: The trial was to evaluate performance of a FFDM system. During image acquisition FDA modified the trial criteria from an efficacy trial to a SE trial. Images from our diag site prior to Nov could not be used in this study as mag views were now required. Mag views were acquired in Jan-Feb 2011. Selected images were used from our screen site.
Overall Study
| Stage | FFDM (Full Field Digital Mammography) |
|---|---|
| STARTED | 107 |
| New FDA Guideline Released | 6 |
| COMPLETED | 6 |
| NOT COMPLETED | 101 |
Reasons for not completing this period:
| Reason | FFDM (Full Field Digital Mammography) |
|---|---|
| Statistical population not required | 101 |
Analysis population for the primary outcome, as defined in the registry: This was not a statistical sample per FDA FFDM Guideline. FDA Guideline specified that a minimum of 6 film sets from the participants be analyzed. The number of participants required by FDA changed mid-study.
5. Baseline Characteristics
The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.
| Characteristic | Unit | FFDM (Full Field Digital Mammography) |
|---|---|---|
| Participants analysed | 107 | |
| Age, Categorical: <=18 years | Participants | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 107 |
| Age, Categorical: >=65 years | Participants | 0 |
| Sex: Female, Male: Female | Participants | 107 |
| Sex: Female, Male: Male | Participants | 0 |
| Region of Enrollment: United States | participants | 104 |
| Region of Enrollment: Germany | participants | 3 |
6. Primary Endpoint
The registry lists the primary outcome as FFDM (Full Field Digital Mammography) Mammogram Scores, measured over Day 1.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | FFDM Mammogram set |
| Endpoint classification | not reduced to a single comparison |
The primary measure is titled “FFDM (Full Field Digital Mammography) Mammogram Scores.” Its listed unit is an FFDM mammogram set, and its measurement type is recorded as a number. The description explains that 6 mammography image sets, each containing 4 images, were rated pass or fail by 2 MQSA-certified mammography readers after a typical MQSA evaluation. It reports a total of 12 scores and gives Day 1 as the timeframe.
Pass/fail scoring is a categorical way to record whether an image set met an evaluation standard. In general, such ratings can summarize reader judgments against specified criteria. They do not, by themselves, show how accurately a system detects disease unless the outcome definition and suitable reference information are also available. The registry’s stated broader aim concerns detection accuracy, but the described measure is an evaluation of FFDM image sets. The record does not explain how the individual ratings were combined into the named number measure, what constituted the final score, or how a screen-film result was obtained.
The registry does not report a numeric result for this endpoint. Without the reported scores and a clearly described comparison, the measure cannot establish the magnitude or precision of any difference in accuracy.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry establishes that an image evaluation was conducted and describes the basic rating procedure: 6 FFDM image sets, 4 images per set, and 2 readers assigning pass/fail ratings. It reports 12 scores in total. These details describe the evaluation process, but they are not outcome results. The primary outcome’s numerical value is not reported in the registry record.
The stated purpose concerns comparative accuracy between FFDM and screen-film mammography. Yet the registered arm description is for FFDM, and the record does not provide a screen-film result or a comparative analysis. It also identifies the study as not randomized and says the arms are not a like-for-like control comparison. Accordingly, the stated aim should not be mistaken for a reported comparative finding. No estimate of an accuracy difference, confidence interval, or test of that difference is reported in the registry record.
The record lists 107 enrolled participants, but the image rating description refers to 6 image sets. It says the image sets were not a statistical sample under the FDA FFDM Guideline and notes that the FDA-required participant count changed during the study. The enrolled count therefore cannot be treated as the number of image sets rated or as the analysis denominator for the primary outcome. The selection and analysis populations are not reported in the registry record.
In general, a confidence interval communicates the range of effect sizes compatible with an estimate under the analysis assumptions. It is useful for judging precision, not only whether a result crosses a threshold. Here, there is no reported effect estimate or interval to interpret. The registry also lists no analyses. That means the posted record does not provide a statistical result from which to assess uncertainty or distinguish a small effect from a large one.
The adverse-event listing reports 0 serious events among 6 at risk and 0 other events among 6 at risk in the FFDM group. These counts refer to the listed safety group; they do not supply the missing image-rating outcome or comparative accuracy analysis. Deaths are not reported in the registry record. Taken together, the record supports a description of the planned image-rating activity and listed safety counts, but not a conclusion about comparative detection accuracy or the precision of such a comparison.
9. Statistical Methods Explained
The outcome description combines a set-level measure with individual pass/fail reader judgments. It says the measure is a number, identifies the unit as an FFDM mammogram set, and describes 12 ratings arising from 6 image sets assessed by 2 readers. It does not report the final count or explain how the individual judgments were converted into a single endpoint value. The analysis method for the primary outcome is not reported in the registry record.
What does a pass/fail rating represent?
In general, a binary rating records which of two categories applies to an observation. It can be straightforward to summarize, but interpretation depends on a clear definition of the categories and on how ratings are linked to the research question. For an accuracy question, a rating alone is not necessarily an accuracy estimate. Accuracy requires a defined target and information that allows reader judgments to be evaluated against that target. The registry describes pass/fail evaluation of FFDM image sets but does not report a diagnostic accuracy statistic or the information used to derive one.
The measure’s title and operational description also leave an important interpretive gap. The title refers to mammogram scores, the unit refers to an image set, and the description reports pass/fail ratings. The registry does not say whether the intended outcome was a count of passing sets, a count of failing sets, a reader-level score, or another summary. These possibilities have different meanings, so a reader should not infer one from the wording alone.
Why does the reader structure matter?
General statistical principle: observations that share the same image set may be related rather than independent. Here, the same 6 image sets were rated by 2 readers. The two ratings associated with an image set may therefore contain overlapping information. Treating all ratings as though each came from an unrelated image set could misstate uncertainty. Analyses of reader evaluations commonly need to account for the structure of the data, but the registry does not report whether or how that structure was addressed.
The record also does not report reader-level results, agreement statistics, or a method for resolving different ratings. Such details would help distinguish variation between readers from variation across image sets. Their absence does not show that no such procedures were used; it means they are not reported in the registry record.
What can be inferred from the absence of analyses?
The registry lists 1 outcome measure and 0 analyses. An outcome measure states what was intended to be assessed; an analysis specifies how observations are used to produce an estimate or comparison. With no analysis listed and no numeric endpoint result, there is no reported statistical estimate to interpret. The record does not report a p-value, confidence interval, adjusted estimate, or model for the primary outcome.
General statistical principle: a confidence interval expresses uncertainty around an estimated quantity under the analysis assumptions. A p-value, when used, summarizes how compatible data are with a specified null model; it does not measure the size or clinical importance of an effect. Neither quantity can be reconstructed from the description of ratings alone. The registry record does not report the data needed for an independent recalculation.
Planning illustrations based on an observed effect are conceptually different from results of a trial’s own design. The registry does not report a sample-size justification for this study, nor does it report a statistical plan specifying a target effect or power. Any planning illustration should be read as a hypothetical calculation under its stated assumptions, not as observed power or evidence that the study was designed to detect a particular effect.
General statistical principle: power describes the long-run probability that a specified analysis will detect a specified effect under assumptions. It does not tell whether the observed study result is reliable, and a post hoc calculation using an observed effect is not a substitute for a prespecified design rationale. In this record, the specific analysis, effect estimate, and corresponding uncertainty are not reported in the registry record.
The registry’s note that the image sets were not a statistical sample under the FDA FFDM Guideline is relevant to generalization. A minimum number of film sets specified for an evaluation does not by itself establish that the image sets represent all patients or all clinical settings of interest. The record says the participant requirement changed mid-study, but does not report how that change affected selection, enrollment decisions, or analysis.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 1 outcome measure and 0 analyses. It does not report a statistical analysis plan, the set of hypotheses tested, or any adjustment for multiple comparisons. Because there are no analyses listed, the registry record does not allow an assessment of whether multiplicity was addressed in the study’s own analysis. The absence of listed analyses should not be interpreted as proof that no analyses were performed.
The primary outcome description identifies pass/fail ratings from 2 readers on 6 image sets, but does not describe whether reader-specific summaries, image-set summaries, or another endpoint definition was used. The registry does not report how disagreement was handled, whether any subgroups were considered, or whether any secondary analyses were prespecified. These details matter because multiple outcomes or analytic choices can create more than one opportunity to identify a seemingly notable result. Without the analysis record, the extent of that issue cannot be assessed.
Missing-data handling and analysis populations are not reported in the registry record. The enrollment count is 107, whereas the rating description refers to 6 image sets; the record does not explain the relationship between those counts, identify exclusions, or state how incomplete ratings would have been handled. It also does not report whether all eligible images were evaluated or whether any ratings were unavailable. No complete-case, imputation, or other missing-data method is described.
The listed sample-size planning assumptions are illustrative rather than a reported justification for enrollment. The registry does not report a protocol-based sample-size calculation, an interim-analysis plan, or rules for stopping or changing the study. The note that the FDA-required participant number changed mid-study does not specify the statistical consequences of that change. These omissions limit assessment of the intended error control, the analyzed population, and the robustness of any result.
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 |
|---|---|---|---|
| FFDM (Full Field Digital Mammography) | 0 | 6 | 0.0% |
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 |
|---|---|
| FFDM (Full Field Digital Mammography) | 0 of 6 |
The registry lists a FFDM safety group with 6 participants at risk for serious adverse events and 0 affected. It also lists 6 at risk for other adverse events and 0 affected. These are reported counts for the safety listing, not the number enrolled overall. The record does not explain how the safety group relates to the full enrollment or to the participants whose image sets were rated.
These counts should be read descriptively. A zero count in the listed group means no events of that category are recorded for that group in the registry entry; it does not establish that no events occurred outside the group, that every participant was observed for the same duration, or that event ascertainment was complete. The registry gives no additional detail about event definitions, collection procedures, follow-up, or attribution.
Deaths are not reported in the registry record. The safety counts do not provide information about mammography accuracy and should not be used to infer a comparative benefit or harm against screen-film mammography, because a like-for-like comparator group is not described. The listed safety figures are therefore limited to the reported event counts and denominators.
12. What the Result Does, and Does Not, Mean
The record documents a completed evaluation of FFDM image sets and describes a pass/fail rating procedure conducted by 2 MQSA-certified readers. It also records 107 participants enrolled and explains that the image sets were not a statistical sample under the FDA FFDM Guideline. These are useful details about the study context and the intended image evaluation.
The registry does not report the primary endpoint value, a screen-film result, or a statistical analysis comparing FFDM with screen-film mammography. Consequently, the stated comparative aim cannot be treated as a reported comparative result. There is no reported effect estimate or confidence interval with which to judge the direction, magnitude, or precision of a difference in detection accuracy. The record does not support a conclusion about whether either approach performed better.
The distinction between the stated study purpose and the available endpoint description is important. The purpose refers to accuracy in detecting breast cancer, whereas the described outcome is pass/fail evaluation of FFDM image sets. The registry does not provide the endpoint definition or reference information needed to translate those ratings into an accuracy measure. Nor does it explain how the 12 scores were combined.
In general, an image-set evaluation can answer a question about whether images meet a specified evaluation standard if the standard and scoring method are clearly defined. A diagnostic accuracy comparison asks a different question and requires suitable data for both methods and a defined basis for determining disease status. The registry entry does not report the information needed to make that comparison.
The listed safety counts are limited to the group with 6 at risk and show 0 serious and 0 other adverse events recorded. Those descriptive counts do not change the limits of the efficacy interpretation. Overall, the record describes an evaluation process but does not report results sufficient to quantify accuracy, compare systems, or assess statistical precision.
13. Why This Trial Matters Statistically
This study illustrates why a registered research question and a reported endpoint need to be read together. The stated purpose concerns comparative accuracy of FFDM and screen-film mammography, while the registered outcome describes pass/fail ratings of FFDM image sets. Without a reported comparator result and an analysis connecting the ratings to accuracy, the stated aim cannot be answered from the registry entry alone.
The record also makes the distinction between enrollment and the unit of measurement important. It lists enrollment of 107, but describes ratings of 6 image sets and 12 scores. The registry says the image sets were not a statistical sample under the FDA FFDM Guideline. These counts refer to different aspects of the study and should not be substituted for one another. Clear reporting of who contributed images, which images were analyzed, and how ratings were summarized would help readers understand the scope of the evidence.
Reader-based evaluations introduce another statistical issue: ratings may be related when different readers assess the same images. General statistical practice is to account for the structure of the observations when estimating uncertainty. The registry describes the reader arrangement but does not report the analysis method, reader agreement, or uncertainty estimates. This prevents assessment of how the design was reflected in the analysis.
The entry is also a reminder that registry completeness affects interpretation. It reports a primary measure but no analyses or numerical result. It does not report the sample-size justification, missing-data approach, or analysis population. Without those details, readers cannot evaluate the precision, robustness, or possible selection effects of a result.
These issues matter even when the study is an image evaluation rather than a conventional treatment comparison. The endpoint definition, sampling approach, reader structure, comparator, and analysis all determine what a result can support. Here, the registry provides some design details but not the results and methods needed for a comparative statistical conclusion.
14. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The main limitation for interpretation is that the registry does not report the primary outcome value or any analysis. Although it describes pass/fail scoring, it does not report how many image sets passed or failed, the reader-level results, or how ratings were combined. No uncertainty estimate or comparative analysis is reported.
The study’s stated aim is comparative, but the registry lists only an FFDM arm and explicitly says the arms are not a like-for-like control comparison. A screen-film arm and its results are not described. The record also notes that allocation was not stated and that the study was not randomized. These features preclude interpreting the listed information as a randomized head-to-head comparison.
The sampling information is limited. The registry records enrollment of 107 but describes 6 image sets. It states that the image sets were not a statistical sample under the FDA FFDM Guideline and that the number of participants required by FDA changed mid-study. The selection process, relation of the image sets to enrollment, and impact of the change are not reported in the registry record. Generalizability from the evaluated image sets therefore cannot be assessed from this entry.
The registry does not report reader masking, image selection procedures, reference-standard details, agreement or adjudication procedures, missing-data handling, analysis populations, or the statistical method. It does not report an analysis plan, sample-size justification, or interim analyses. These omissions should be understood as gaps in the registry record, not evidence that such procedures were necessarily absent from the study.
Safety reporting is also limited to counts for 6 participants at risk in the listed group. Event definitions and follow-up details are not reported, and deaths are not reported in the registry record. The listed counts cannot characterize risk beyond that recorded group.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Image Evaluation of Philips Philips MammoDiagnost DR Full Field Digital Mammography System (FFDM) — NCT00999596.
- Registry record last updated 2012-12-04; status last verified 2012-11.
- Results first posted 2012-05-02.
- 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
NCT00999596 is a completed clinical image evaluation study sponsored by Philips Healthcare. The registry lists a start date of 2010-04, primary completion date of 2011-01, and enrollment of 107. The stated purpose was to compare the accuracy of Philips FFDM and screen-film mammography, while the registered outcome describes pass/fail ratings of FFDM image sets.
The outcome description covers 6 image sets, with 4 images per set, assessed by 2 MQSA-certified readers, producing 12 scores. The registry says these image sets were not a statistical sample under the FDA FFDM Guideline and notes that the FDA-required participant number changed during the study. It lists 1 outcome measure and 0 analyses, but does not report the outcome value or a comparative accuracy result.
The safety listing records 0 serious and 0 other adverse events among 6 at risk in the FFDM group. Deaths are not reported in the registry record. The available entry describes the evaluation and some safety counts, but does not permit assessment of comparative accuracy or statistical precision.