Cindy S. Lee

The National Mammography Database: Preliminary Data

OBJECTIVE The purposes of our study were to analyze screening mammography data submitted to the National Mammography Database (NMD) since its inception to confirm data collection feasibility, to draw parallels to data from the Breast Cancer Surveillance Consortium (BCSC), and to examine trends over time. We also retrospectively evaluated practice-level variation in terms of practice type, practice setting, census region, and annual volume. MATERIALS AND METHODS Data from 90 mammography facilities in the NMD registry were analyzed. The registry receives mammography data collected as part of standard clinical practice, including self-reported demographic information, clinical findings, screening mammography interpretation, and biopsy results. Outcome metrics calculated were cancer detection rate, recall rate, and positive predictive values for biopsy recommended (PPV2) and biopsy performed (PPV3). RESULTS The NMD successfully collected and analyzed data for 3,181,437 screening mammograms performed between January 2008 and December 2012. Mean values for outcomes were cancer detection rate of 3.43 per 1000 (95% CI, 3.2-3.7), recall rate of 10% (95% CI, 9.3-10.7%), PPV2 of 18.5% (95% CI, 16.7-20.2%), and PPV3 of 29.2% (95% CI, 26.2-32.3%). No statistically significant difference was seen in performance measurements on the basis of practice type, practice setting, census region, or annual volume. NMD performance measurements parallel those reported by the BCSC. CONCLUSION The NMD has become the fastest growing mammography registry in the United States, providing nationwide performance metrics and permitting comparison with published benchmarks. Our study shows the feasibility of using the NMD to audit mammography facilities and to provide current, ongoing benchmark data.

Beginner’s Guide to Practice Quality Improvement Using the Model for Improvement

Radiologists in the United States are required to complete the Practice Quality Improvement (PQI) program as part of their Maintenance of Certification by the ABR. The Institute for Healthcare Improvements (IHI) Model for Improvement (MFI) offers an alternative to the 3-phase approach currently advocated by the ABR. The MFI implicitly assumes that many interventions will need to be tested and refined for any meaningful project, and provides a project management approach that enables rapid assessment and improvement of performance. By collecting data continuously, rather than simply before and after interventions, more interventions can be tested simultaneously and projects can progress more rapidly. In this article, we describe the ABRs 3-phase approach, and introduce the MFI and how it can be employed to affect positive changes. Using a radiology case study, we demonstrate how one can utilize the MFI to enable rapid quality improvement.

Association of Patient Age With Outcomes of Current-Era, Large-Scale Screening Mammography: Analysis of Data From the National Mammography Database

This study analyzes mammography screening data from the National Mammography Database to determine the age cutoff point for screening.

Screening for Breast Cancer in Women Age 75 Years and Older

OBJECTIVE Older women undergoing regular mammography experience significant reductions in breast cancer mortality, except in women with severe comorbidities or limited life expectancy. Optimizing screening strategies requires informed discussions of benefits and risks given each womans health status. CONCLUSION This article will review the benefits and risks of screening mammography in women older than 75 years within the context of life expectancy and comorbidities and summarize the current recommendations from professional organizations for screening mammography in older women.

Assessing the Recall Rate for Screening Mammography: Comparing the Medicare Hospital Compare Dataset With the National Mammography Database

OBJECTIVE High-quality screening mammography has been shown to substantially reduce mortality from breast cancer. Recall rate is a principal performance metric for screening mammography because it directly relates to the rate of false-positive examinations. This study aims to compare the recall rate derived using two sources-the claims-based Hospital Compare (HC) dataset from the Centers for Medicare & Medicaid Services versus the National Mammography Database (NMD) from the American College of Radiology-to understand the implications in pay-for-performance and quality improvement activities. MATERIALS AND METHODS This study retrospectively compared the recall rate reported by NMD facilities with that reported in the HC dataset. Site matching was performed by facility name and zip code, followed by manual verification. Scatterplots, correlations, a paired t test, and Bland-Altman analysis were performed to assess association between the two measures. RESULTS During the period from October 1 to December 1, 2016, 92 facilities were unambiguously matched using 2014-2015 records in both datasets. The recall rates were positively correlated (r = 0.428, p < 0.001), but the mean HC recall rate (8.5% ± 2.86% [SD]) was significantly (p < 0.001) lower than the mean NMD recall rate (10.6% ± 3.90%). CONCLUSION The NMD and HC are two commonly used datasets for measuring screening mammography recall rate. Although recall rates are correlated at the individual facility level, there are important differences that have implications for quality improvement and pay-for-performance.

BI-RADS Category 5 Assessments at Diagnostic Breast Imaging:Outcomes Analysis Based on Lesion Descriptors

RATIONALE AND OBJECTIVES The Breast Imaging-Reporting and Data System (BI-RADS) atlas defines category 5 assessments as appropriate only for lesions that are almost certainly cancer, with a positive predictive value (PPV) of ≥95%. This study aims to demonstrate the feasibility of classifying lesions at diagnostic breast imaging with sufficiently high PPV to merit category 5 assessments, and to identify those lesion descriptors that yield such a high PPV. MATERIALS AND METHODS For this Health Insurance Portability and Accountability Act compliant and IRB exempt study, we reviewed diagnostic breast imaging examinations (mammography and/or ultrasound) assessed as highly suggestive of malignancy (BI-RADS category 5). Pathology diagnosis was considered the gold standard. PPV3 (biopsy performed) was calculated, and the BI-RADS descriptors for each lesion were analyzed. RESULTS Among 22,564 consecutive diagnostic breast imaging examinations between January 2010 and September 2015, we identified 239 exams (1.1%) assessed as BI-RADS category 5 (mean age 62.5 years). Malignancy (invasive breast carcinoma and/or ductal carcinoma in situ) was diagnosed in 233 examinations (PPV3 97.5% and 95% confidence interval: 96.2%-98.8%). The most common lesion types were mass (170) and calcifications (116). Of the 220 examinations involving both mammography and ultrasound, no category 5 lesions had <3 suspicious BI-RADS descriptors, only three lesions had three suspicious descriptors, but the remaining 217 lesions (98.6%) had ≥4 suspicious descriptors. CONCLUSION In clinical practice, it is feasible to make BI-RADS category 5 assessments with the intended ≥95% PPV. To justify a category 5 assessment, at least four suspicious BI-RADS descriptors should be identified at the combination of diagnostic mammography and ultrasound examinations.

Harmonizing Breast Cancer Screening Recommendations: Metrics and Accountability

OBJECTIVE The purposes of this article are to summarize breast cancer screening recommendations and discuss their differences and similarities and to explain the differences between two national databases to aid in interpretation of their benchmarks. CONCLUSION The American College of Radiology, American Cancer Society, and U.S. Preventive Services Task Force all agree that annual mammography beginning at age 40 saves the most lives, and all acknowledge a womans right to choose when to begin and stop screening. The National Mammography Database (NMD) differs from the Breast Cancer Surveillance Consortium database in that it acquires data using the same approach used by almost all mammography facilities in the United States. Therefore, NMD benchmarks, which include standard metrics, provide more meaningful comparisons to help mammography facilities and radiologists improve performance.

Does mild cognitive impairment increase the risk for developing postoperative cognitive impairment

cognitive impairment (MCI). The current study relates cardiac function (cardiac output and ejection fraction) to early cognitive and neuroimaging markers of cerebrovascular and Alzheimer’s disease (AD) pathology among individuals with MCI. Methods: 3T cardiac and brain MRI and neuropsychological data were collected on 10 MCI participants free from clinical stroke (62-84 years, 7567; 46% women). Cardiac output and ejection fraction, determined by cardiac MRI, were related to neuropsychological and neuroimaging markers of vascular (executive functioning, fractional anisotropy) and AD pathology (learning and memory, hippocampal volume). Results: Partial correlations (pr), adjusting for systolic blood pressure and body surface area, suggest that cardiac output is significantly associated with Digit Symbol (pr1⁄40.72, p1⁄40.04), California Verbal Learning Test Long Delay Recall (pr1⁄40.95, p1⁄40.001), Biber Figure Learning Test Trials 1-5 Recall (pr1⁄40.13, p1⁄40.04), and hippocampal volume (pr1⁄40.76, p1⁄40.04). Partial correlations, adjusting for systolic blood pressure, suggest that ejection fraction is significantly associated with Color-Word Interference (pr1⁄4-0.66, p1⁄40.05). In all cases, as cardiac function decreased, the integrity of the brain aging phenotype also decreased. Conclusions: Our preliminary data based on a very small sample of MCI patients suggest that cardiac function is related to brain aging phenotypes indicative of early Alzheimer’s disease (memory, hippocampal volume) and cerebrovascular changes (executive functioning). Additional cross-sectional data collection and longitudinal follow-up up this cohort will provide essential information in determining if cardiac function is associated with accelerating maladaptive brain aging among older adults at risk for cognitive decline. The successful completion of this research will increase understanding about the complex interrelations between subclinical impairments in cardiac function and cognitive progression in MCI.