Recent debates on the value of mammography screening have highlighted its major downfall – the high chance of having false positive results, which lead to unnecessary biopsies. Ultrasonography after a suspicious finding on a mammogram can help further discriminate potentially benign from cancerous lesions, however these too are far from perfect. Thus, many biopsies are performed needlessly.
This week news came out of the annual Radiological Society of North America (RSNA) meeting that, according to two small studies, a newer type of utlrasonography, elastography, could help better identify benign lesions from cancer. The incomplete results of these studies were presented at the meeting, but they have yet to be published. Readers of such stories should note the results with some caution: scientific presentations have not been subject to a rigorous peer-review process. Published articles undergo review by experts who can comment on the quality of the design, analysis and interpretation of the results.
This story does little to expand upon the press releases provided on the RSNA website. (Available at: http://www.rsna.org/Media/rsna/RSNA09_newsrelease_target.cfm?id=441). It appears to lift content directly from the press release, both in the numbers provided, the text itself and the quotes from the story lead authors. The story gives no evidence that any independent reporting was done beyond lifting and slightly rewriting what is written in the press release. As such, the story adequately describes the design of the studies, the novelty of elastography and quantifies the benefits. However, it fails to discuss the availabililty of elastography, and does not describe the potential costs or harms of the test.
The lack of independent expert perspectives in the story was a major shortcoming.
Especially in this time frame when breast cancer has received so much confusing news coverage, it is imperative that a story do a complete job of explaining what these results mean to women who are considering mammography and who are concerned about the risk of having to undergo biopsies that are of no benefit. This story failed to deliver that information.
The story makes no mention of the cost of elastography. The use of a second additive technology will clearly add to the costs of diagnosis unless the sensitivity and specificity of the test is sufficiently high to reduce unnecessary biopsies. The potential cost implications of this added technology are not noted in the story. Although the true economic impact of the technology is not known at present, the potential cost implications of the method should have been noted.
The story repeats the quantification of benefits as they are presented in the press release but these differ slightly from what is published in the abstracts. The story mentions that elastography correctly identified 98% of cancerous lesions and 82% of benign lesions, however the abstract states that elastography identified 76% of benign lesions. It isn’t clear where this discrepancy comes from and for that reason we must judge this unsatisfactory. The story states that these numbers are better than conventional utrasonography but provides no quantification for comparison. The story should have placed them both into context with other technologies including MRI, computer assisted detection and full field mammography as well as ultrasonography.
The story mentioned no harms of elastography, which are similar to conventional ultrasonography. Excessive false positive results could lead to needless biopsies in the case of both skin and breast lesions. Alternatively, false negative results of the new test could lead to undiagnosed breast or skin cancer. The story focuses on the reduced likelihood of biopsy in both cases failing to comment on the potential downside.
While a sophisticated reader may note that the results were to be presented at a meeting, have not been published or subjected to peer review and are preliminary (since the study is ongoing), the story fails the average reader. The story could have emphasized that the peer-review process often picks up problems in the design, conduct and analysis of the studies, whereas presentations at scientific meetings are not subject to such scrutiny. Results presented at meetings should come with this note of caution to readers. The comments of the lead author of the breast lesion study suggesting elastography can eliminate the need for biopsies seem premature at best based on the preliminary results of this small study. The results of the study to date are presented incompletely and do not allow the reader to determine the relative sensitivity and specificity of the elastography as compared to ultrasound.
The story does not engage in disease mongering. It accurately states that most breast lesions are benign.
The story quotes only experts who were involved in the study. Even worse, the two experts who are quoted are also quoted in the press release. The story should have also pointed out that the senior author is listed as an investigator for the company that produces the device used for the study. This is a major flaw of the story. Persepctives from independent sources could have tremendously improved the quality of the story.
The story compares elastography to traditional ultrasonography and biopsy but fails to mention the availability of MRI, full field mammography, computer assisted ultrasound and other technologies available or under development.
The story makes no mention of the availability of elastography. The test is still under evaluation and is not widely available outside of research settings.
The story adequately describes the novelty of elastography.
The story appears to lift content directly from the Radiology Society of North America’s press release, available at: http://www.rsna.org/Media/rsna/RSNA09_newsrelease_target.cfm?id=441. Not only are the same numbers used in the story as the press release, but the same quotes and very similar text are used in both. The story gives no evidence of any independent reporting other than reading the press release to generate the story.