Re: False + mammograms
From: art fougner, md (evsono@pipeline.com)
Fri Sep 26 07:12:41 2003
For the radiologist - it's the false negative ( approx 15% of cancers
may be undectable on mammogram) that causes grief. failure to dx breast
ca is the radiologist's equivalent of the "failure to perform a timely
c-section."
art
At Thu, 25 Sep 2003, Joanne Bulley, MD wrote:
>
>Reuters Health Information 2003. © 2003 Reuters Ltd.
>
>Mammograms More Likely to Be False Positive in North America
>
>NEW YORK (Reuters Health) Sept 16 - Radiologists in North America are
>more prone to labeling screening mammograms as abnormal than their
>counterparts in other countries, investigators in the U.S. report.
>
>According to a report in the Journal of the National Cancer Institute
>for September 17, false-positive results are of concern because they
>increase patient anxiety, costs per case, and the number of diagnostic
>procedures performed. The tradeoff is likely to be a reduced incidence
>of missed or delayed diagnoses, which could ultimately save costs.
>
>Dr. Joann G. Elmore at the University of Washington, Seattle, and
>colleagues identified 32 reports about screening mammography results in
>community-based practices in or after 1985. Eight reports were from
>North America. One was from Hong Kong, three were from Australia, and
>the remainder were from Europe.
>
>The proportion of mammogram readings that resulted in further diagnostic
>evaluation ranged from 5.5% to 15.0% in the North American programs and
>1.2% to 12.6% in those of other countries. Weighted mean averages were
>8.4% and 5.6%, respectively (p = 0.018). The positive predictive value
>of an abnormal reading ranged from 4.4% to 12.2% in North America, and
>from 3.4% to 48.7% in other countries.
>
>The differences remained significant after adjusting for the percentage
>of women under age 50, reported calendar year, number of readers, and
>number of radiographic views per breast. Program location was the most
>significant predictor of abnormal findings.
>
>The percentage of mammograms deemed abnormal was inversely associated
>with positive predictive value, both for screening and for biopsy
>results. The proportion of abnormal mammograms was associated with
>proportion of ductal carcinoma in situ (DCIS) cases among women
>diagnosed with cancer and the number of DCIS cases diagnosed per 1000
>screens.
>
>The authors offer several possible explanations for the observed
>heterogeneity. These include characteristics of the population
>screened, such as age and proportion of those undergoing first
>screening; features of the health care system, such as malpractice
>concerns and financial incentives; and physician characteristics.
>
>"A better understanding of the sources of variability in mammography may
>lead to more effective screening programs that have a lower percentage
>of mammograms judged as abnormal without substantially lowering the
>cancer detection rate," Dr. Elmore's group concludes.
>
>J Natl Cancer Inst 2003;95:1384-1393.
>
>--
>Joanne Bulley, MD
>Keene, NH, USA
>
>-----
>Work to create peace everywhere you go and with everything you do.
>-----
>Let us accept truth, even when it surprises us and alters our views.
>-George Sand
>
--
art fougner, md
ich bin ein New Yorker
|
|
