Characteristics Of Mammography Facilities Associated Accuracy Defined

Main Category: Radiology / Nuclear Medicine
Also Included In: Women's Health / Gynecology;  Cancer / Oncology
Article Date: 24 Jun 2008 - 0:00 PDT

email to a friend   printer friendly   view / write opinions   rate article

Four characteristics of mammography facilities have been found to be associated with higher overall levels of accuracy in screening, reports a study published in the Journal of the National Cancer Institute (online June 10). The study showed that higher interpretative accuracy is seen with facilities offering only screening mammography, which have breast imaging specialists to interpret mammograms, which undertake single readings (as opposed to double) and which undertake independent audits at least twice yearly.

"These findings are important because a referring physician or the patient herself is much more likely to have the opportunity to choose the facility where the mammogram is performed and interpreted than they are to choose the radiologist who will interpret the mammogram," write the authors, Stephen Taplin and colleagues from the National Cancer Institute (Bethesda, Maryland, USA).

With more than 27 million women undergoing screening mammography world-wide each year, there is growing concern to improve technique performance. The two main factors known to influence interpretation of mammograms are the characteristics of the women being screened (factors like breast density, age and the time since their last mammogram) and the experience of the interpreting radiologist (things like years on the job and reading volumes). However, studies have suggested that these characteristics account for only 10 % of variations in performance.

In the current observational study, Taplin and colleagues set out to determine the extent to which screening performance varies according to the characteristics of different screening facilities. The cross-sectional survey based study involved 484,463 screening mammograms, taking place between January 1996 and December 2001, that were performed on 237 669 women at 44 facilities.

Breast cancer outcomes for participants in the study were ascertained by linking the data with regional surveillance registries, local tumour registries and breast pathology databases. Results show that on average, the facilities identified cancer when it was present (sensitivity) in 79.6 % of breast cancer cases that occurred within one year and correctly categorised a mammogram as cancer-free (specificity) 90.2 % of the time.

The authors used the area under the curve (AUC) method to look for differences between facilities in terms of overall accuracy. This statistical method is routinely used to flag up differences between different sets of data. The AUC was higher among facilities that offered screening mammograms alone versus those that offered screening and diagnostic mammograms (0.943 vs 0.911, P = .006), had a breast imaging specialist interpreting mammograms compared to those who did not (0.932 vs 0.905, P = .004), that did not perform double reading versus independent double reading versus consensus double reading (0.925 vs 0.915 vs 0.887, P = .034), or conducted audit reviews two or more times per year versus annually versus at an unknown frequency (0.929 vs 0.904 vs 0.900, P = .018).

The authors were surprised that neither facility volume, nor the method of audit review, were found to be associated with greater interpretive accuracy. Furthermore, double reading methods did not show any advantages.

One of the limitations of the study design, they say, was that it did not take account the different methods of double reading that are used in different centres. Other limitations included possible unaccounted differences among women and radiologists. The authors took account of breast density, age and time since last mammogram in women, and years of experience and reading volumes in radiologists, but say there may be other factors that were not taken into account.

"Understanding how facility characteristics influence interpretive accuracy could allow women and physicians to choose a mammography facility based on characteristics that are more likely to be associated with higher quality. Radiologists could also change the facilities' structures or processes to include practices that improve interpretive accuracy," conclude the authors, adding that the associations identified require further prospective evaluation.

Reference

Mammography Facility Characteristics Associated With Interpretive Accuracy of Screening Mammography
Taplin S, Abraham L, Barlow WE et al.
Journal of the National Cancer Institute 2008, 100: 876-887

Cancer Research Summaries are overviews of important cancer research findings that have been reported in leading cancer publications. The Cancer Research Summaries are provided by the Cancer Media Service (CMS) in collaboration with Nature Clinical Practice Oncology.

All previous Cancer Research Summaries can be viewed here.

"This summary is provided by the European School of Oncology's Cancer Media Service"

Cancer Media Service