Estimates of overdiagnosis of invasive breast cancer associated with screening mammography.
ABSTRACT To estimate the extent of overdiagnosis of invasive breast cancer associated with screening in New South Wales, Australia, a population with a well-established mammography screening program which has achieved full geographic coverage.
We calculated overdiagnosis as the observed annual incidence of invasive breast cancer in NSW in 1999-2001 (a screened population) minus the expected annual incidence in this population at the same time, as a percentage of the expected incidence. We estimated expected incidence without screening in 1999-2001 from the incidence of invasive breast cancer in: (1) women in unscreened age groups (interpolation method); and (2) women in all age groups prior to the implementation of screening (extrapolation method). We then adjusted these estimates for trends in major risk factors for breast cancer that may have coincided with the introduction of mammography screening: increasing obesity, use of hormone replacement therapy (HRT) and nulliparity. Finally, we adjusted for lead time to produce estimates of expected incidence in 1999-2001. These were compared with the observed incidence in 1999-2001 to calculate overdiagnosis of breast cancer associated with screening.
Overdiagnosis of invasive breast cancer among 50-69 year NSW women was estimated to be 42 and 30% using the interpolation and extrapolation methods, respectively.
Overdiagnosis of invasive breast cancer attributable to mammography screening appears to be substantial. Our estimates are similar to recent estimates from other screening programmes. Overdiagnosis merits greater attention in research and in clinical and public health policy making.
SourceAvailable from: PubMed Central[Show abstract] [Hide abstract]
ABSTRACT: To determine the optimal method for quantifying and monitoring overdiagnosis in cancer screening over time. Systematic review of primary research studies of any design that quantified overdiagnosis from screening for nine types of cancer. We used explicit criteria to critically appraise individual studies and assess strength of the body of evidence for each study design (double blinded review), and assessed the potential for each study design to accurately quantify and monitor overdiagnosis over time. PubMed and Embase up to 28 February 2014; hand searching of systematic reviews. English language studies of any design that quantified overdiagnosis for any of nine common cancers (prostate, breast, lung, colorectal, melanoma, bladder, renal, thyroid, and uterine); excluded case series, case reports, and reviews that only reported results of other studies. 52 studies met the inclusion criteria. We grouped studies into four methodological categories: (1) follow-up of a well designed randomized controlled trial (n=3), which has low risk of bias but may not be generalizable and is not suitable for monitoring; (2) pathological or imaging studies (n=8), drawing conclusions about overdiagnosis by examining biological characteristics of cancers, a simple design limited by the uncertain assumption that the measured characteristics are highly correlated with disease progression; (3) modeling studies (n=21), which can be done in a shorter time frame but require complex mathematical equations simulating the natural course of screen detected cancer, the fundamental unknown question; and (4) ecological and cohort studies (n=20), which are suitable for monitoring over time but are limited by a lack of agreed standards, by variable data quality, by inadequate follow-up time, and by the potential for population level confounders. Some ecological and cohort studies, however, have addressed these potential weaknesses in reasonable ways. Well conducted ecological and cohort studies in multiple settings are the most appropriate approach for quantifying and monitoring overdiagnosis in cancer screening programs. To support this work, we need internationally agreed standards for ecological and cohort studies and a multinational team of unbiased researchers to perform ongoing analysis. © Carter et al 2015.BMJ Clinical Research 01/2015; 350(jan07 5):g7773. DOI:10.1136/bmj.g7773 · 14.09 Impact Factor
[Show abstract] [Hide abstract]
ABSTRACT: Debate about the extent of breast cancer over-diagnosis due to mammography screening has continued for over a decade, without consensus. Estimates range from 0-54%, but many studies have been criticized for having flawed methodology. In this study we used a novel study design to estimate over-diagnosis due to organised mammography screening in South Australia (SA). To estimate breast cancer incidence at and following screening we used a population-based, age-matched case-control design involving 4931 breast cancer cases and 22914 controls to obtain OR for yearly time intervals since women’s last screening mammogram. The level of over-diagnosis was estimated by comparing the cumulative breast cancer incidence with and without screening. The former was derived by applying OR for each time window to incidence rates in the absence of screening, and the latter, by projecting pre-screening incidence rates. Sensitivity analyses were undertaken to assess potential biases. Over-diagnosis was estimated to be 8% (95%CI 2-14%) and 14% (95%CI 8-19%) among SA women aged 45-85 years from 2006-2010, for invasive breast cancer and all breast cancer respectively. These estimates were robust when applying various sensitivity analyses, except for adjustment for potential confounding assuming higher risk among screened than non-screened women, which reduced levels of over-diagnosis to 1% (95%CI -5 -7%) and 8% (95%CI 2-14%) respectively when incidence rates for screening participants were adjusted by 10%. Our results indicate that the level of over-diagnosis due to mammography screening is modest and considerably lower than many previous estimates, including other for Australia. © 2014 Wiley Periodicals, Inc.International Journal of Cancer 08/2014; 136(6). DOI:10.1002/ijc.29124 · 6.20 Impact Factor
[Show abstract] [Hide abstract]
ABSTRACT: This study aimed to: (1) Estimate cumulative risk of recall from breast screening where no cancer is detected (a harm) in Australia; (2) Compare women screened annually versus biennially, commencing age 40 versus 50; and (3) Compare with international findings. At the no-cost metropolitan program studied, women attended biennial screening, but were offered annual screening if regarded at elevated risk for breast cancer. The cumulative risk of at least one recall was estimated using discrete-time survival analysis. Cancer detection statistics were computed. In total, 801,636 mammograms were undertaken in 231,824 women. Over 10 years, cumulative risk of recall was 13.3 % (95 % CI 12.7-13.8) for those screened biennially, and 19.9 % (CI 16.6-23.2) for those screened annually from age 50-51. Cumulative risk of complex false positive involving a biopsy was 3.1 % (CI 2.9-3.4) and 5.0 % (CI 3.4-6.6), respectively. From age 40-41, the risk of recall was 15.1 % (CI 14.3-16.0) and 22.5 % (CI 17.9-27.1) for biennial and annual screening, respectively. Corresponding rates of complex false positive were 3.3 % (CI 2.9-3.8) and 6.3 % (CI 3.4-9.1). Over 10 mammograms, invasive cancer was detected in 3.4 % (CI 3.3-3.5) and ductal carcinoma in situ in 0.7 % (CI 0.6-0.7) of women, with a non-significant trend toward a larger proportion of Tis and T1N0 cancers in women screened annually (74.5 %) versus biennially (70.1 %), χ (2) = 2.77, p = 0.10. Cancer detection was comparable to international findings. Recall risk was equal to European estimates for women screening from 50 and lower for screening from 40. Recall risk was half of United States' rates across start age and rescreening interval categories. Future benefit/harm balance sheets may be useful for communicating these findings to women.Breast Cancer Research and Treatment 12/2014; 149(1). DOI:10.1007/s10549-014-3226-x · 4.47 Impact Factor