By Jennifer Cawson
Breast News: Newsletter of the NHMRC National Breast Cancer Centre
Vol. 6, No. 1, Spring 2000
The paper Ten year risk of false positive screening mammograms and clinical
breast examination 1 studied women participating in a breast cancer screening
program of a private health insurance company (HMO) in Boston, US.
Background
A random sample of 2400 women was reviewed to determine whether further tests
were required for a suspicious finding on mammography or clinical examination,
and whether cancer had been detected. A test was classified as false positive if
suspicious results required further investigation, and cancer was not diagnosed
within 12 months. The costs were examined.
Results
Ninety-six per cent of women underwent screening mammography and clinical
examination - median number of mammograms four, and median number of clinical
examinations five.
Breast cancer was diagnosed in 88 women. Fifty-eight cancers were detected by
mammography and seven by clinical examination at the time of screening. Eleven
women had cancer diagnosed after 12 months had lapsed and so were not classified
as screen detected, and 23 had true interval cancers.
False positives occurred with 6.5% of mammograms and 3.7% of clinical
examinations. Twenty-four per cent of women had at least one false positive
mammogram, and 13% had at least one false positive clinical examination. This is
higher than in other countries 2,3.
The cumulative risk of having at least one false positive after 10 screenings
was 49% for mammography and 25% for clinical examination. False positive testing
required 870 consultations, 539 mammograms, 186 ultrasounds and 188 biopsies.
The cumulative false positive biopsy rate is estimated at 18.6% from mammography
and 6.2% from clinical examination. About $US1 million was expended on screening
and $US320,000 on work-up tests.
Commentary
False positive mammography is accompanied by increases in psychological 4,5,6,7
and economic cost. However, a recent paper found no reduced intention to
participate in screening and no significant long-term psychological morbidity 8.
The screening model described does not exist in Australia, and the costs of
medical care are far higher in the US than here. No comparable studies have been
published here, however BreastScreen Australia has an extensive database and
comprehensive accreditation standards. In Victoria, an annual statistical report
9 collates rates of assessment and tests performed. Data are collected regarding
the relative costs of administration, recruitment, screening and assessment
10,11.
Clearly, the risk of morbidity of the program must be weighed against the
benefit of early cancer detection .The total number of cancers detected (88)
seems high for 2400 women in 10 years. However, the cancers are a mix of screen
detected, delayed screen-detected and interval cancers.
A difficulty relates to the definition of a screen-detected cancer (detected
within a year of screening), and the practice in the US of mammographic
follow-up for probably benign lesions. This makes comparability with other
countries difficult. BreastScreen accreditation standards allow only 5% of women
to be reviewed in this way. In 95% of cases, a definitive decision regarding
management (ie clear or biopsy) is made at the assessment visit.
The participation rate of women in the target group in Victoria is lower at 54%,
but the re-screen rate of 80% is high 9. In a 10-year period, the average number
of mammograms for women is about the same in both systems.
The recall rate in the prevalent round is comparable in Victoria (6.1%) but
lower (3.4%) in recurrent rounds. The trend for assessment rate to be lower with
increasing age is similar in the study and in Victoria.
In Victoria, for 2400 women at a 6% assessment rate, about 250 mammograms, 120
ultrasounds, 130 clinical examinations and 100 biopsies would be performed, a
much lower rate of tests than in the study.
In Victoria, the cost of screening 2400 women four times in 10 years, at current
rate of $54 per screen and double-read, would be about $AUD500,000. The cost of
assessment and biopsy would be about $AUD220,000. Hence the ratio of cost of
screening and assessment is similar to the study, but the actual costs are
lower. This high rate of work-up examinations may be due to the loose program
structure compared with the highly organised model of screening in Australia.
In conclusion, the aims to optimise cancer detection and minimise morbidity are
shared by both countries. The economic and probably the personal costs are
greater in the US.
Source :
http://pandora.nla.gov.au/nph-arch/2000/H2000-Sep-1/http://www.nbcc.org.au/pages/info/resource/nbccpubs/brnews/00spr/spr6b.htm