NPC Archive Item: Overdiagnosis in breast cancer screening

NOTE – This is an archive post from the NPC and has not been updated since first publication. Therefore, some hyperlinks may no longer be working.
MeReC Rapid Review NPC Logo

30th July 2009

A systematic review of published trends in the incidence of breast cancer, before and after introduction of mammography screening, identifies that one in every three breast cancers identified in national screening programmes are overdiagnosed, i.e. they are inconsequential and result in unnecessary treatment. It is important that the absolute risks and benefits of breast cancer screening, including the level of overdiagnosis, are communicated to women in a balanced way so that they can make an informed decision to have breast cancer screening or not.

Level of evidence:
Level 2 (limited-quality patient-oriented evidence) according to the SORT criteria.

Health professionals should ensure that women who are considering breast cancer screening are provided with balanced and accurate information on the risks and benefits of screening, so that they can make an informed choice. A patient decision aid, which is based on the UK NHS Breast Screening Programme’s published data, may help in communicating the risks and benefits of screening and is available on the breast cancer section of NPC.

What is the background to this?
Breast cancer screening was introduced in the UK in 1988 for women aged 50 to 64 years, with national coverage by 1990. It was extended to women aged 65 to 70 in 2002, and is being extended further to women aged 47 to 73. Information is available on the national breast cancer screening programme in the UK on the NHS Breast Screening Programme website. Included is a downloadable patient leaflet “Breast Screening: The Facts”.

Overdiagnosis refers to the detection of abnormalities that will never cause symptoms or death during a patient’s lifetime, for example, when the cancer grows so slowly that the patient dies of other causes before it produces symptoms, or when the cancer remains dormant (or regresses). The issue of overdiagnois is well recognised in the case of prostate cancer (see the Prostate Cancer Risk Management website) for which a national screening programme is not recommended. Overdiagnosis of breast cancer in screening programmes may result in many women receiving unnecessary cancer-specific treatments. These can have a serious impact on quality of life and are, themselves, associated with a very small increased risk of death. The likelihood of this harm has to be balanced against the benefits that will be obtained in some women by early detection and effective treatment of breast cancers, for example, by reducing the risk of death from breast cancer and/or avoiding the need for mastectomy.

A previous Cochrane review of randomised trials of mammography screening from several countries indicated a clear trend for overdiagnosis of breast cancer, with a pooled  overall estimate for overdiagnosis of 30%.Taking into account the absolute reduction in breast cancer mortality and the degree of overdiagnosis, the results of that review indicated that for every 2,000 women invited for screening throughout 10 years, one would have her life prolonged. However, in addition, 10 healthy women would be diagnosed as having breast cancer, who would not have been diagnosed if there had not been screening, and will be treated unnecessarily. The authors were uncertain as to whether screening does more good than harm.

The authors of the present study tested the hypothesis that overdiagnosis could have been overestimated in this previous review, as the trials considered did not have lifelong follow-up. Therefore, they carried out a systematic review of published trends in the incidence of breast cancer before and after the introduction of screening programmes around the world, in order to identify the degree of overdiagnosis.

What does this study claim?
This study analysed the trends in the incidence of breast cancer before and after introduction of mammography screening in five countries. It identified that an increase in the incidence of breast cancer was closely associated with the introduction of the screening and little of this increase was compensated for by a drop in the incidence of breast cancer in previously screened women. From meta-analysis, the total overdiagnosis of breast cancer in publicly available mammography screening programmes was estimated as 52% (95% confidence interval [CI] 46% to 58%), i.e. about one in three are overdiagnosed.

So what?
This article provides valuable information that confirms the presence of overdiagnosis in breast cancer screening and provides a more reliable estimate than previously available of the extent to which it occurs within national screening programmes. As it is not known which women who are diagnosed as having breast cancer will benefit from treatment, unnecessary treatment for many women is inevitable. It is important that balanced information on the risks and benefits of breast cancer screening are communicated to women, so that they can make an informed decision on whether to be screened or not.

The current “Breast Sceening: The Facts” patient leaflet is designed to ensure that women are told what screening can and cannot achieve. The leaflet contains information on the benefits of breast cancer, in terms of the number of lives that can be saved, and how early detection may prevent the need for mastectomy. It describes the possibility of pain from the mammography procedure and the small risk associated with X-rays. It also describes the test as imperfect, and points out the possibility of false positive results and the worry this might cause, as well as false negative results, where a cancer is present but not detected. However it does not describe the harms in absolute terms, and does not mention overdiagnosis and the possibility and consequences of unnecessary treatment.  Providing this additional information would provide a more balanced view of the risks and benefits and be helpful to women, when considering  whether or not to undertake the screening or not. The patient decision aid (which is based on the UK NHS Breast Screening Programme’s published data) on the breast cancer section of NPC might be helpful in this respect. As the editorial accompanying the study says: “Mammography is one of medicine’s ‘close calls’ — a delicate balance between benefits and harms — where different people in the same situation might reasonably make different choices. Mammography undoubtedly helps some women but hurts others. No right answer exists, instead it is a personal choice.”

More information on the management of breast cancer can be found on the relevant section of NPC.

Please comment on this blog in the NPC discussion rooms, or using our feedback form.

Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you.

Study details

Jørgensen KJ, Gøtzsche PC. Overdiagnosis in publicly organised mammography screening programmes: systematic review of incidence trends. BMJ 2009;339:b2587 doi:10.1136/bmj.b2587

Design: Systematic review of published trends (using linear regression analysis) in incidence of breast cancer before and after the introduction of screening in: the UK; Manitoba, Canada; New South Wales, Australia, Sweden; and parts of Norway. Meta-analysis was used to estimate the extent of overdiagnosis.

Patients: women, according to age groups: 35 to 49 years (too young to be screened); 50 to 69 years (screening age); and 70 to 84 years (exceeded age for screening).

Intervention and comparison: Incidence data covering at least seven years before screening and seven years after screening had been fully implemented, and including both screened and unscreened groups. Data from unscreened age groups were used to identify changes in background incidence and compensatory drops in incidence among older, previously screened women, and adjustments made. From analyses of the change in the rate of breast cancer incidence prior to the screening period and the rates subsequent to screening, overdiagnosis was estimated as the rate ratio between the results for the last observation year determined by linear regression and the expected incidence.

Outcomes and results: The introduction of screening was closely related to an increase in the incidence of breast cancer in each of the five regions/countries studied. Data from three countries (not the UK) showed a drop in incidence as the women exceeded the age limit for screening, but the reduction was small and the estimate of overdiagnosis was compensated for in this review. Meta-analysis estimated the overall overdiagnosis of breast cancer (including carcinoma in situ) as 52% (95%CI 46% to 58%); the estimate for England and Wales was 57% (95%CI 53% to 61%).

Sponsorship: None