This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: https://info.ahrq.gov. Let us know the nature of the problem, the Web address of what you want, and your contact information.
Please go to www.ahrq.gov for current information.
Radiologists sometimes vary widely in their interpretation of
the same mammograms
U.S. radiologists looking at the same mammogram are likely to
interpret it quite differently, according to a study supported in
part by the Agency for Health Care Policy and Research (HS07845).
Inaccuracy in mammogram interpretation may mean that breast
cancer goes undetected or is detected at a later stage when it is
more difficult to treat successfully. If mammography is to
realize its potential to reduce breast cancer deaths,
accreditation programs that certify the technical quality of
radiographic equipment and images should also certify the
accuracy of mammogram interpretation, concludes Craig A. Beam,
Ph.D., of the Medical College of Wisconsin.
Dr. Beam and his colleagues compared interpretation of the same
set of 79 screening mammograms by a 1992 random sample of 108
radiologists from American College of Radiology-accredited
mammography centers across the United States. The actual breast
disease status of the 79 women whose mammograms were interpreted
had previously been established either by biopsy or 2-year
followup. There was a 53 percent range of variability between the
minimum and maximum sensitivity among radiologists in the sample.
While some radiologists referred 100 percent of women with cancer
for biopsy, others referred only 47 percent.
Fifty percent of the radiologists demonstrated at least 80
percent sensitivity (the proportion of women with cancer who were
referred for biopsy) in interpreting the mammograms meaning that
half of the radiologists missed detecting 20 percent or fewer of
possible cancers on the mammograms. One-fourth of the
radiologists had a sensitivity exceeding 88 percent, but the
other 25 percent had a sensitivity of less than 67 percent. The
researchers estimated at least a 45 percent range of variability
in radiologists' specificity, that is, in the rate of
recommendation for routine or short-term followup given to women
without cancer. The overall specificity for women with normal
mammograms was 90 percent.
More details are in "Variability in the interpretation of
screening mammograms by U.S. radiologists," by Dr. Beam, Peter M.
Layde, M.D., M.Sc., and Daniel C. Sullivan, M.D., which appears
in the January 22, 1996 Archives of Internal Medicine 156,
Choice of treatment for early-stage breast cancer often
depends on hospital characteristics
A woman's chance of having breast-conserving surgery, that is,
lumpectomy plus radiation instead of mastectomy (breast removal)
for early-stage breast cancer depends on the hospital where she
is treated. Although the outcomes of both treatments are similar,
not all hospitals have adopted the breast-conserving approach.
Urban hospitals and teaching hospitals are more apt to adopt this
approach than other hospitals, according to a study by Agency for
Health Care Policy and Research investigators and former AHCPR
administrator, J. Jarrett Clinton, M.D.
They used hospital data on women discharged with early-stage
breast cancer from 1981 to 1987 from more than 500 hospitals
included in the Healthcare Cost and Utilization Project (HCUP) to
model the influence of hospital characteristics on the use of
breast-conserving surgery, while controlling for patient and
hospital characteristics. Analysis showed that the overall rate
of radical mastectomies (removal of breast tissue, some chest
muscles, and axillary lymph nodes) decreased by nearly 21 percent
per year. The rate of modified radical mastectomies (large
muscles of the chest that move the arms are preserved) and simple
mastectomies (only breast tissue is removed) remained relatively
constant. In contrast, the rate of breast-conserving surgeries
increased by 9 percent per year.
Women treated at nonteaching hospitals were 27 percent less
likely to have breast-conserving surgery than those treated at
academic medical centers. Also, women treated at urban hospitals
were 43 percent more likely to have breast-conserving surgery
than those treated at rural hospitals. This difference is
commonly attributed to a slower diffusion of new technologies and
innovations in rural areas. The authors point out, however, that
the choice of mastectomy may be reasonable in light of the
limited access in rural areas to radiation therapy.
For more information, see "Treating early-stage breast cancer:
Hospital characteristics associated with breast-conserving
surgery," by Mary E. Johantgen, Ph.D., R.N., Rosanna M. Coffey,
Ph.D., D. Robert Harris, Ph.D., and others, in the October 1995
American Journal of Public Health 85(10), pp. 1432-1434.
Treatment intensity varies by age in elderly women with
breast cancer even after controlling for other medical
Elderly women account for 44 percent of all newly diagnosed cases
of breast cancer, and 93 percent of them are diagnosed before the
disease has metastasized. According to a study supported by the
Agency for Health Care Policy and Research (HS06589), elderly
women are treated less aggressively the older they become, and
this trend is not attributable to the increased presence of other
medical problems with advancing age. It shows that older women
are less likely to undergo surgery, more likely to receive
breast-conserving surgery as opposed to mastectomy, and less
likely to receive radiotherapy following breast-conserving
A research team led by Sheldon M. Retchin, M.D., M.S.P.H., of the
Medical College of Virginia, examined the influence of patient
age and coexisting medical conditions (comorbidity) on initial
treatment for breast cancer using Virginia Cancer Registry
records for 2,252 women ages 66 years or older. The women were
diagnosed as having nonmetastatic, invasive breast cancer between
1984 and 1989. Women with breast cancer who were 85 years of age
and older had about one-third the odds of women 66 to 74 years of
age to be treated initially by surgery, about half the odds of
undergoing radical mastectomy (removal of breast and chest tissue
and axillary lymph nodes) instead of breast-conserving surgery,
and were much less likely than the younger elderly women to
receive radiation therapy within 4 months after surgery.
Women ages 75 through 84 years did not have significantly
different odds of undergoing any surgery or receiving
breast-conserving surgery than women ages 66 through 74 years. It
has been suggested that more coexisting medical conditions in the
very elderly may contraindicate more aggressive treatment
options. In this study, however, even after adjustment for
comorbidity which tends to increase with age, age remained a
strong and significant predictor of treatment, according to the
For more details, see "The effect of age and comorbidity in the
treatment of elderly women with nonmetastatic breast cancer," by
Craig J. Newschaffer, M.S., Lynne Penberthy, M.D., M.P.H.,
Christopher E. Desch, M.D., and others, in the January 8, 1996
Archives of Internal Medicine 156, pp. 85-90.
Return to Contents
Proceed to Next Article