0

The full content of Annals is available to subscribers

Subscribe/Learn More  >
Original Research |

Overdiagnosis of Invasive Breast Cancer Due to Mammography Screening: Results From the Norwegian Screening Program

Mette Kalager, MD; Hans-Olov Adami, MD, PhD; Michael Bretthauer, MD, PhD; and Rulla M. Tamimi, ScD
[+] Article and Author Information

Grant Support: By the Norwegian Research Council and Frontier Science.

Potential Conflicts of Interest: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M11-1560.

Reproducible Research Statement:Study protocol and data set: Not available. Statistical code: Available from Dr. Kalager (e-mail, mailto:mkalager@hsph.harvard.edu).

Requests for Single Reprints: Mette Kalager, MD, Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115; e-mail, mailto:mkalager@hsph.harvard.edu.

Current Author Addresses: Drs. Kalager and Adami: Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115.

Dr. Adami: Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, POB 281, SE-171 77 Stockholm, Sweden.

Dr. Bretthauer: Department of Transplantation Medicine, Oslo University Hospital Rikshospitalet, 0027 Oslo, Norway.

Dr. Tamimi: Channing Laboratory, Brigham and Women's Hospital and Harvard Medical School, 181 Longwood Avenue, Boston, MA 02115.

Author Contributions: Conception and design: M. Kalager, M. Bretthauer.

Analysis and interpretation of the data: M. Kalager, H.O. Adami, M. Bretthauer, R.M. Tamimi.

Drafting of the article: M. Kalager, H.O. Adami, M. Bretthauer.

Critical revision of the article for important intellectual content: M. Kalager, H.O. Adami, M. Bretthauer, R.M. Tamimi.

Final approval of the article: M. Kalager, H.O. Adami, M. Bretthauer, R.M. Tamimi.

Provision of study materials or patients: M. Kalager.

Statistical expertise: M. Kalager.

Obtaining of funding: M. Kalager.


From Harvard School of Public Health, Brigham and Women's Hospital, and Harvard Medical School, Boston, Massachusetts; Karolinska Institutet, Stockholm, Sweden; and Oslo University Hospital Rikshospitalet, Oslo, Norway.


Ann Intern Med. 2012;156(7):491-499. doi:10.7326/0003-4819-156-7-201204030-00005
Text Size: A A A

Background: Precise quantification of overdiagnosis of breast cancer (defined as the percentage of cases of cancer that would not have become clinically apparent in a woman's lifetime without screening) due to mammography screening has been hampered by lack of valid comparison groups that identify incidence trends attributable to screening versus those due to temporal trends in incidence.

Objective: To estimate the percentage of overdiagnosis of breast cancer attributable to mammography screening.

Design: Comparison of invasive breast cancer incidence with and without screening.

Setting: A nationwide mammography screening program in Norway (inviting women aged 50 to 69 years), gradually implemented from 1996 to 2005.

Participants: The Norwegian female population.

Measurements: Concomitant incidence of invasive breast cancer from 1996 to 2005 in counties where the screening program was implemented compared with that in counties where the program was not yet implemented. To adjust for changes in temporal trends in breast cancer incidence, incidence rates during the preceding decade were also examined. The percentage of overdiagnosis was calculated by accounting for the expected decrease in incidence following cessation of screening after age 69 years (approach 1) and by comparing incidence in the current screening group with incidence among women 2 and 5 years older in the historical screening groups, accounting for average lead time (approach 2).

Results: A total of 39 888 patients with invasive breast cancer were included, 7793 of whom were diagnosed after the screening program started. The estimated rate of overdiagnosis attributable to the program was 18% to 25% (P < 0.001) for approach 1 and 15% to 20% (P < 0.001) for approach 2. Thus, 15% to 25% of cases of cancer are overdiagnosed, translating to 6 to 10 women overdiagnosed for every 2500 women invited.

Limitation: The study was registry-based.

Conclusion: Mammography screening entails a substantial amount of overdiagnosis.

Primary Funding Source: Norwegian Research Council and Frontier Science.

Figures

Grahic Jump Location

Figure 1. Implementation of the different groups in the breast cancer screening program in different regions from 1986 to 2005.

The Appendix describes each of the 6 regions. Four groups are defined dependent on the county and the calendar year. The current screening group consists of women invited to screening. The current nonscreening group consists of women in areas where screening was not offered in the period that screening was offered. The historical groups consist of women residing in the different counties in the 10-y period before screening was offered. Data were obtained from Kalager et al (13).

Grahic Jump Location
Grahic Jump Location

Figure 2. Yearly breast cancer incidence rates for women aged 50 to 79 y in the screening and nonscreening groups from 1986 to 2005.

The researchers calculated the incidence rates of the current vs. historical nonscreening groups (the time effect). The incidence rate ratios of women aged 50–79 y in the program group include both changes in risk factors (the time effect), the increase in incidence due to lead time (lead-time effect), and the overdiagnosed cases. To account for lead time, we included women aged 70–79 y (older than the upper age limit of the screening program). We calculated overdiagnosis as the incidence rate ratios among women aged 50–79 y in the current vs. historical screening groups divided by the incidence rate ratios in the current vs. historical nonscreening groups. The peak in incidence observed in 1996–1997 is due to detection of prevalent cancer in the first screening round in region I (40% of the country). In theory, the increase in incidence due to lead time should be compensated for by a subsequent decrease in incidence among older age groups and may only be observed for age groups no longer offered screening (22).

Grahic Jump Location
Grahic Jump Location

Figure 3. Age-specific invasive breast cancer incidence rates for the 4 groups.

This figure shows the incidence rates for the current and historical screening groups (top), the incidence rates for the current and historical nonscreening groups (middle), and incidence rate ratios comparing the ratio of the incidence rate in the current and historical groups for both the screening and nonscreening groups (bottom). Only women aged 50–69 y in the current screening group were invited to undergo mammography screening.

Grahic Jump Location
Grahic Jump Location

Appendix Figure 1. Defined daily doses for menopausal hormone therapy among all women in Norway from 1990 to 2010.

Systemic treatment consisted of Anatomical Therapeutic Chemical group G03C-estrogens and G03F-combined estrogen and progesterone, and local treatment consisted of Anatomical Therapeutic Chemical group G03C. Adapted from the Norwegian Institute of Public Health.

Grahic Jump Location
Grahic Jump Location

Appendix Figure 2. The defined daily doses for menopausal hormone therapy for women aged 60 to 69 y in the Norway counties from 2004 to 2010.

Hormone therapy consisted of the Anatomical Therapeutic Chemical group G03C-estrogens and G03F-combined estrogen and progesterone. Data were obtained from the Norwegian Prescription Database.

Grahic Jump Location
Grahic Jump Location

Appendix Figure 3. The defined daily doses for menopausal hormone therapy for women aged 50 to 59 y in the Norway counties from 2004 to 1010.

Hormone therapy consisted of the Anatomical Therapeutic Chemical group G03C-estrogens and G03F-combined estrogen and progesterone. Data were obtained from the Norwegian Prescription Database.

Grahic Jump Location
Grahic Jump Location

Appendix Figure 4. Yearly breast cancer incidence rates for women aged 50 to 69 y in the nonscreening groups and defined daily doses for menopausal hormone therapy for women of all ages in Norway from 1990 to 2005.

Incidence rate is calculated from the data in the text, and data on hormone therapy are obtained from Appendix Figure 1. The incidence rate is smoothed by using a 3-y moving average.

Grahic Jump Location
Grahic Jump Location

Appendix Figure 5. Rates of death from various causes among Norwegian women from 1951 to 2005.

Rates are age-standardized to the Norwegian population of 1 January 1981. Death from breast cancer is also shown in a larger format (bottom). Data obtained from the Norwegian Institute of Public Health.

Grahic Jump Location

Tables

References

Letters

NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Comments

Submit a Comment
The percent of overdiagnosed breast cancers is over-adjusted and underestimated
Posted on April 11, 2012
Per-HenrikZahl, Doctor
Norwegian Institute of Public Health
Conflict of Interest: None Declared

Kalager and colleagues [1] report 15-25% overdiagnosis of invasive cancers in the Norwegian public screening program. In 2004 we reported that screening lead to 51% more breast cancer which could not be explained by earlier diagnosis and called this phenomenon overdiagnosis [2]. In 2012 we updated our study and reported the same high level of overdiagnosis [3]. We now had longer follow-up and included 11 631 breast cancers diagnosed after screening start while this paper only has 3862 cases. The difference is, however, not explained by data per se but by the methods for adjusting for underlying incidence increase and lead time.

There is no incidence increase of breast cancer in Norway for women under age 50 or above 69 years [2] and we have argued that the incidence increase in the screened age group before screening started is explained by private screening and not by hormone use [3]. When screening first started in 1996-97, DCIS rates increased from 10 to 50 per 100 000. When many other counties started with screening in 2003, DCIS rates increased from 25 to 50 per 100 000 [4]. Thus, private screening was extensive and the intensity was increasing before public screening started. Furthermore, after the program started, use of hormones has dropped 80% (70% decline in sales figures combined with 33% population increase) without any decline in the breast cancer incidence [3]. Today we can exclude hormones as a major explanation for the incidence increase before screening started because there has not come any decline after women stopped using hormones.

Overdiagnosed tumors should not be adjusted for lead time because it is not possible to exclude the overdiagnosed cancers when calculating lead time and by including cancers diagnosed after age 69 we dilute the estimate. To illustrate this problem, let us suppose there are 200 cancers in age group 50-69 and 100 in the age group 70-79 years in an un-screened cohort. Suppose the corresponding numbers for a screened cohort is 300 and 75 - i.e. screening detects 100 extra cancers but only 25 are earlier diagnosis. By using their method we estimate 25% overdiagnosis (375/300). Our estimate is 38% (275/200) and it is correctly adjusted for earlier diagnosis (cancers that would appear after age 69).

In addition, all DCIS are overdiagnosed - by definition if they regress - and if they progress, we should adjust the cancer incidence for every DCIS that has progressed.

References

1. Kalager M, Adami H-O, Bretthauer M, Tamimi RM. Overdiagnosis of invasive breast cancer due to mammography screening: Results from the Norwegian Screening program. Ann Intern Med 2012; 156: 491-499.

2. Zahl P-H, Strand BH, Maehlen J. Breast cancer incidence in Norway and Sweden during introduction of nation-wide screening: prospective cohort study. BMJ 2004; 328: 921-924.

3. Zahl P-H, Maehlen J. Overdiagnosis of breast cancer after 14 years with screening. Tidsskr Nor Laegeforen 2012; 132: 414-417

4. Zahl P-H, Maehlen, Welch HG. Re: The natural history of invasive breast cancers detected by screening mammography [Letter]. Arch Intern Med 2009; 169: 1000-1.

Conflict of Interest:

None declared

The Effects of Mammography Screening and Overdiagnosis on All-Cause Mortality
Posted on April 19, 2012
IsmailJatoi, University of Texas Health Science Center
MD, PhD
Conflict of Interest: None Declared

Mammography screening reduces cause-specific (breast cancer-specific) mortality, but it seems to have no benefit in reducing all-cause mortality (1), (2). Kalgar et al's recent study may shed some light on this observation (3). These authors demonstrated that mammography screening results in a significant overdiagnosis of breast cancer. A woman overdiagnosed with breast cancer will be subjected to cancer-specific treatments (surgery, systemic therapy, radiotherapy), all of which are associated with a very small increase in the risk of death (4). Thus, the overdiagnosis of breast cancer will result in overtreatment, and the small excess in all-cause mortality associated with overtreatment may potentially offset any benefit of mammography screening on breast cancer- specific mortality.

References

1. Jatoi I, Anderson WF. Cancer screening. Curr Probl Surg. Sep 2005;42(9):620-682.

2. Black WC, Haggstrom DA, Welch HG. All-cause mortality in randomized trials of cancer screening. J Natl Cancer Inst. Feb 6 2002;94(3):167-173.

3. Kalager M, Adami HO, Bretthauer M, Tamimi RM. Overdiagnosis of invasive breast cancer due to mammography screening: results from the norwegian screening program. Ann Intern Med. Apr 3 2012;156(7):491-499.

4. Bird BR, Swain SM. Cardiac toxicity in breast cancer survivors: review of potential cardiac problems. Clin Cancer Res. Jan 1 2008;14(1):14 -24.

Conflict of Interest:

None declared

The rate of overdiagnosis is overestimated
Posted on April 20, 2012
TorHaldorsen, Researcher, Steinar Tretli and Giske Ursin
Cancer Registry of Norway
Conflict of Interest: None Declared

To the editor: We think Kalager et al (1) gives an exaggerated picture of overdiagnosis in the Norwegian Breast Cancer Screening Program. The main issue is whether the excess incidence during the period of prevalent and subsequent screening (mainly 50-69 years in Norway) is compensated by the deficit in incidence after screening (2). The estimation of overdiagnosis in Approach 1 assumes that the "Current screening group" contains observations during (50-69 years) and after screening (70-79 years). However, this is not the case. Even in Region 1 where screening started first, only 37 percent of person-years 70-79 years are among women after ended invitation to screening. The rest of person- years are mostly from unscreened cohorts (born 1926 and before). In addition older birth cohorts 1927-28 were invited to a second screening when they were 70-71 years old. For the regions starting screening later, only 0-23 percent of observations 70-79 years old are among women after screening. There can obviously not be a post-screening deficit in incidence after age 70 in women who have not, ever, been invited to screening. The authors therefore underestimate the incidence deficit after screening.

Another bias in Approach 1 is the inclusion of data on prevalent screening for all ages (50-69 years) in the "Current screening group". The excess in incidence is higher for prevalent than for subsequent screening (2). Since women will experience the prevalent screening just once, too much weight is put on data from prevalent screening. The bias will lead to overestimation of excess incidence during screening.

Approach 2 is solely based on observations from the subsequent screening period. The results will rely heavily on the assumptions and precision of data. A minimal requirement for this risky modeling enterprise should be use of individual data on screening history. Their exclusion of data from the "first screening round" in each region does not exclude all prevalent cancers. To do so they would also need to exclude cancers detected in later screening rounds in 50-51 year olds and in older women participating for the first time on a later invitation. The inclusion of data on prevalent cancers will lead to an overestimation of the excess incidence in the subsequent period. We conclude that the rate of overdiagnosis is overestimated in Kalager et al (1).

References

1. Kalager M, Adami HO, Bretthauer M, Tamimi RM. Overdiagnosis of invasive breast cancer due to mammography screening: results from the norwegian screening program. Ann Intern Med 2012 Apr 3;156(7):491-499. 2. Boer R, Warmerdam P, de Koning H, van Oortmarssen G. Extra incidence caused by mammographic screening. Lancet 1994 Apr 16;343(8903):979.

Tor Haldorsen MSc Steinar Tretli PhD Giske Ursin MD, PhD

Conflict of Interest:

None declared

Error in numbers
Posted on April 20, 2012
Ragnhild S.Falk, Statistician, Per Skaane, Solveig Hofvind
Cancer Registry of Norway
Conflict of Interest: None Declared

To the editor: Kalager et al (1) use data from the Norwegian Breast Cancer Screening Program, Cancer Registry of Norway and Statistics Norway for quantification of overdiagnosis due to mammography screening. We are concerned about some of the numbers that are reported. According to the fifth paragraph of Results approximately 500 000 women have been invited to the program at the end of 2005 while the correct number is 592 155. In the same paragraph the absolute number of overdiagnosis appears to be wrongly estimated. The authors estimate percentage of overdiagnosis to be 15% to 25%. Then the observed 7793 cancers in the current screening group represent 115% and 125%. The authors have erroneously set these to 100%. Thus, the absolute estimates become1016 (7793 - (7793/1.15)) and 1559 (7793-(7793/1.25)), and not 1169 and 1948 as were published.

In Table 2 outcomes per 2500 women are presented. There are 20 cases of breast cancer without screening and between 26 and 30 cases with screening. This yields a percentage of overdiagnosis from 30% ((26 - 20)/20) to 50% ((30-20)/20), which is twice the estimate presented earlier in the same paragraph. The person-years in this paper are based on ecological data, computed from population statistics from January 1st each year. Analyses are based on results from four defined population groups: screening groups and nonscreening groups in a historical period (1986-1995) and in a current period (1996-2005). We added the numbers from the current period for ages 50-69 years in Table 1. The number, 4 204 064 (1 866 741 + 2 337 323), is different from what can be found in Statistics Norway; 4 518 003 (2). This discrepancy of almost 314 000 person-years is also found in a previous paper (3) by some of the same authors. These person-years should have been included in the denominators to rates of central importance in these studies. We ask the authors for a clarification of the pointed numbers.

Ragnhild S?rum Falk MSc Per Skaane MD, PhD Solveig Hofvind PhD

References

1. Kalager M, Adami HO, Bretthauer M, Tamimi RM. Overdiagnosis of invasive breast cancer due to mammography screening: results from the norwegian screening program. Ann Intern Med 2012; 156:491-9.

2. Statistics Norway. StatBank Norway. Table 05839. Accessed at http://statbank.ssb.no/statistikkbanken/Default_FR.asp?PXSid=0&nvl=true&PLanguage=0&tilside=selectvarval/define.asp&Tabellid=05839 on 20 April 2012.

3. Kalager M, Zelen M, Langmark F, Adami HO. Effects of screening mammography on breast-cancer mortality in Norway. N Engl J Med 2010; 363:1203-10.

Conflict of Interest:

None declared

Overdiagnosis in Norway
Posted on April 24, 2012
Karsten J.Joergensen, MD
The Nordic Cochrane Centre
Conflict of Interest: None Declared

We have estimted 37% overdiagnosis of invasive breast cancer in Norway, and 52% when including ductal carcinoma in situ (DCIS) [1]. This means that one in three breast cancers detected in the screened age group is overdiagnosed. We have estimated 33% overdiagnosis in Denmark [2], which has the longest period worldwide that allows a contemporaneous comparison between screened and non-screened areas. There are three main reasons for the difference between our estimates for Norway and the estimate of 18-25% overdiagnosis of invasive breast cancer in the present study [3]. We prefer the analysis that relies on observed numbers rather than on an assumed lead time. First, we included DCIS; second, we did not include the non-screened age group 70-79 years; and third, we define overdiagnosis in a slightly different way. Kalager et al. assess that including DCIS requires special considerations that they could not honour. But their results indicate that overdiagnosis of DCIS is likely more common than we assumed, as we counted half the identified DCIS lesions as overdiagnosed [2]. If DCIS lesions identified at screening were not overdiagnosis, their removal should lower the frequency of invasive cancers later on. This did not happen in Norway [3], or elsewhere [4,5]. Early stage invasive breast cancers skyrocket in incidence with screening, and rates of late stage invasive breast cancers have not been reduced. Given that DCIS constitutes between 10-25% of breast cancers in a screened population, we believe it is essential to include a proportion of these in an estimate of overdiagnosis if we are to make a meaningful comparison of the benefits and harms of screening. Kalager et al. included the largely non-screened age-group 70-79 years in order to take into account a compensatory reduction in breast cancer incidence in previously screened women (lead-time bias). We found a decline in this age group in the screened areas in Norway in the relevant time period, but also a similar reduction in the non-screened areas [2]. We conservatively chose to compensate for the decline in our estimate of overdiagnosis, subtracting it from the observed incidence increase in the screened age group 50-69 years. Simply including the age group 70-79 years substantially dilutes the true level of overdiagnosis. When estimating overdiagnosis, we asked the question "by how many percent did breast screening increase incidence." Kalager et al. seems to have asked the question "how large a percentage of the breast cancers were overdiagnosed." Consider a bottle of wine that is increased in price from 10 to 15 dollars. We would say that the price have increased by 50%, whereas Kalager et al. would say that one third of the new price is excessive. Similarly, if a drug increases the number of headaches by 50%, we say exactly this. Recognizing this difference is of importance to the way we communicate estimates of overdiagnosis to healthcare professionals and invited women. The difference in the level of overdiagnosis between Norway and Denmark is likely caused by Denmark having substantially lower participation and a deliberately conservative attitude towards micro-calcifications, leading to particularly low levels of DCIS [2].

References Cited

1. Joergensen KJ, Goetzsche PC. Overdiagnosis in publicly organised mammography screening programmes: systematic review of incidence trends. BMJ 2009;339:b2587.

2. Joergensen KJ, Zahl PH, Goetzsche PC. Overdiagnosis in organised mammography screening in Denmark. A comparative study. BMC Women',s Health 2009;9:36.

3. Kalager M, Adami HO, Bretthauer M, Tamimi RM. Overdiagnosis of invasive breast cancer due to mammography screening: results from the Norgwegian screening program. Ann Int Med 2012;156:491-9.

4. Autier P, Boniol M, Middleton R, Dor? J-F, H?ry C, Zheng T, et al. Advanced breast cancer incidence following population based mammographic screening. Ann Oncol 2011;22:1726-35.

5. Nederend J, Duijm LEM, Voogd AC, Groenewoud JH, Jansen FH, Louwman MWJ. Trends in incidence and detection of advanced breast cancer at biennial screening mammography in The Netherlands: a population based study. Breast Cancer Res 2012;14:R10. doi:10.1186/bcr3091.

Conflict of Interest:

None declared

Author's Response
Posted on May 31, 2012
MetteKalager, MD, Hans-Olov Adami, MD, PhD, Michael Bretthauer, MD, PhD, Rulla M. Tamimi, ScD
University of Oslo, Institute of Health and Society, Department of Health Management and Health Econ
Conflict of Interest: None Declared

To the editor,

Falk et al. state that the correct number of women invited to the screening program is 592,155. However, based on data provided to us by the Cancer Registry of Norway, approximately 496,157 women were invited to the screening program in our study period. Further, Falk et al. may have misunderstood our estimates of overdiagnosis. We estimated the amount of overdiagnosis based on four groups of women, comparing current versus historical groups (1). The actual number of invited women is not included in any of our estimates. As shown in our paper, we calculated the number of overdiagnosed women as the number of women with breast cancer in the screening group (7793) x the amount of overdiagnosis (15% and 25%, respectively). Even if the numbers of invited women we received from the Cancer Registry were increased to reflect the latest release of data (as given in the letter by Falk et al.), our estimated number of women with overdiagnosis would remain similar, ranging from 1169 to 1948.

There are several ways to calculate overdiagnosis. Falk et al. have used the number of expected cases without screening (a method also suggested by J?rgensen) as the denominator in their estimates of overdiagnosis. We used the number of women with cancer in the screening cohort as the denominator. Using our estimate of overdiagnosis (number of women in the screening group in the denominator) and including the number of cancers that were detected among women aged 50 to 69 years in the screening group who would have been diagnosed at later ages (due to lead time, approximately 10 per 2500 women invited), 6/(20+6+10) or 10/(20+10+10) is in perfect accordance with our estimate of 15 to 25 % overdiagnosis. However, if we used the nonscreening cancers in the denominator as suggested by Falk et al. and J?rgensen, the amount of overdiagnosis would be 30% and 50% respectively (1169/3948 and 1948/3948).

Our method for person-years calculation is described in detail in the statistical appendix accompanying both the present study and a study published in NEJM (1,2). In brief, it takes two years for all individuals age 50 to 69 years to become enrolled in the program in each county (region). Further, as also pointed out in the letter by Haldorsen et al., it takes two years for women aged 50 to 51 years to be enrolled in the program. Falk et al. summed the total number of inhabitants in the different counties, but this does not take into account the gradual enrollment of both women aged 50 to 69 years in the first round (2 years) in all counties, and the gradual enrollment of women aged 50 to 51 years in each county every second year. By taking into account the gradual enrollment of women in the program, the correct person-years in the screening group for the age group 50 to 69 is 2,337,323.

Haldorsen et al. raise concern that there are biases that may lead to an overestimate in overdiagnosis. As discussed in the paper, the follow-up time after the prevalent rounds is short in some counties. To compensate for this, we used two different approaches, and performed a sensitivity analysis including only data from region 1 where the screening program has been running for ten years. If the claims of bias by Haldorsen et al. were true and had a meaningful impact, we would expect the level of overdiagnosis to be lower in the analyses enriched for women with multiple screening rounds; however we did not find this to be true. In an analysis restricted to women aged 50-74 years in region 1 only, where approximately 64% of women older than 69 had previously been screened, a 22% of overdiagnosis was observed (in the paper we found 18% overdiagnosis when we included women aged 50 to 79 years) (1). Further, after 14 years of screening a 7% nonstatistically significant reduction in incidence of invasive cancer was seen among women aged 70 to 74 years previously screened compared to women not previously screened (3). Thus, it does not seem as the expected drop in incidence among older women (called post- screening deficit by Haldorsen et al.), compensates for the large increase seen in women in the screening age. In approach 2, prevalent cancers were excluded, also among women aged 50-51 years in all rounds and counties. Thus, our sensitivity analyses addressing Haldorsen et al?s concerns do not support the claim that our estimate of overdiagnosis is overestimated.

Zahl argues that the increase in incidence among women aged 50 to 69 years in the nonscreening group is explained by opportunistic or ?private? screening. Appendix figure 4 shows the yearly incidence for women aged 50 to 69 corresponding with the yearly use of hormone treatment during menopause. As mentioned in our discussion, these data show that opportunistic screening cannot explain our observations. We agree with Zahl that including DCIS would have lead to higher estimates of overdiagnosis. Detection and removing DCIS is expected to be followed by a decrease in incidence of invasive breast cancer. In our analysis, this possible decrease is included in the estimate. Some DCIS will not progress to invasive cancer, all of these are per definition overdiagnosed.

References

1. Kalager M, Adami HO, Bretthauer M, Tamimi RM. Overdiagnosisi of invasive breast cancer due to mammography screening: Results from the Norwegian screening program. Ann Intern Med 2012;156:491-9.

2. Kalager M, Zelen M, Langmark F, Adami HO. Effect of screening mammography on breast-cancer mortality in Norway. N Engl J Med 2010;363:1203-10[PMID:20860502]

3. Zahl PH, M?hlen J. Overdiagnositkk av brystkreft etter 14 r med mammografiscreening. Tidsskr Nor legeforen 2012;132:414-7. [Norwegian]

Conflict of Interest:

None declared

Submit a Comment

Summary for Patients

Overdiagnosis of Invasive Breast Cancer With Mammography Screening

Summaries for Patients are a service provided by Annals to help patients better understand the complicated and often mystifying language of modern medicine.

Summaries for Patients are presented for informational purposes only. These summaries are not a substitute for advice from your own medical provider. If you have questions about this material, or need medical advice about your own health or situation, please contact your physician. The summaries may be reproduced for not-for-profit educational purposes only. Any other uses must be approved by the American College of Physicians.

Read More...

Clinical Slide Sets

Terms of Use

The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.

Toolkit

Buy Now

to gain full access to the content and tools.

Want to Subscribe?

Learn more about subscription options

Advertisement
Related Articles
Related Point of Care
Topic Collections
Forgot your password?
Enter your username and email address. We'll send you a reminder to the email address on record.
(Required)
(Required)