Summary Background There is uncertainty about overdiagnosis in screening mammography. Methods We aimed to estimate the effect of screening on breast cancer incidence and overdiagnosis in the NHS Breast Screening Program in England. The study included 57,493 cases and 105,653 controls, with cases defined as women diagnosed between 47 and 89 years of age with primary breast cancer, invasive or ductal carcinoma in situ , in 2010 or 2011. Wherever possible, They selected two controls per case, matched on date of delivery and screening area. Conditional logistic regression was used to estimate the effect of screening on breast cancer risk, adjusting for possible self-selection bias. The results were combined with national incidence data to estimate absolute rates of overdiagnosis. Overdiagnosis was calculated as the cumulative excess of cancers diagnosed in the 50 to 77 age group in a woman who was screened every three years from age 50 to 70 compared with a woman who was not screened . to screening exams. Results The estimated number of overdiagnosed cases in women who attended all the program tests was 679.3 per 100,000 without adjustment for self-selection bias and 261.2 per 100,000 with adjustment. These corresponded to an estimated 9.5% of screening-detected cancers overdiagnosed without adjustment and 3.7% with adjustment for self-selection. Conclusions The NHS Breast Screening Program in England confers, at worst, modest levels of overdiagnosis. |
Comments
In the UK, breast cancer accounts for 31% of all new cancers diagnosed in women each year, and cases have increased by 4% in the last decade. Despite the increase in cases, deaths from breast cancer are decreasing. This is partly due to the early diagnosis of breast cancer through the NHS Breast Screening Programme.
The NHS Breast Screening Program invites anyone registered with a GP as a woman aged 50 to 70 for an assessment every three years. Breast cancer screening uses mammograms to check the breasts for signs of cancer. Despite the benefits of screening, there remains some debate about the potential harms of screening, particularly overdiagnosis.
Overdiagnosis is not the same as when a test finds something abnormal that turns out not to be cancer (a false positive) .
Overdiagnosis is the diagnosis of a real cancer that grows so slowly that it would never actually have caused any symptoms.
Overdiagnosis may be a consideration when women are considering whether or not to participate in breast cancer screening. Until now, estimates of overdiagnosis have varied widely, from less than 5% of cancers detected by screening to more than 30%. When a cancer is diagnosed, we cannot know if it is overdiagnosed or not, so all cancers need treatment. It is important for women to have reliable estimates of overdiagnosis so they can make an informed decision about whether to be screened or not.
To quantify overdiagnosis in the NHS Breast Screening Programme, Queen Mary researchers conducted a study of 57,493 breast cancer cases diagnosed in 2010 or 2011, combined with 105,653 controls.
They estimated the effect of screening on breast cancer risk and combined the results with national incidence data to estimate absolute rates of overdiagnosis. Overdiagnosis was calculated as the cumulative excess of cancers diagnosed in women aged 50 to 77 years who attended screening every three years from age 50 to 70 compared with women who did not attend screening.
The estimated number of cases of overdiagnosis in women who attended all examinations in the program was just under 3 per 1000 , corresponding to an estimated 3.7% of cancers detected by overdiagnosed examinations.
This is considerably lower than has been suggested in the past. The authors conclude that the NHS Breast Screening Program is associated, at worst, with modest overdiagnosis of breast cancer.
Stephen Duffy, professor of cancer screening at Queen Mary University of London and joint principal investigator, said:
“These results provide some reassurance that participation in the NHS Breast Screening Program confers only a low risk of overdiagnosed breast cancer. "Together with the results of our previous study on the effect of screening on breast cancer mortality, this indicates that the benefit of screening in preventing breast cancer deaths outweighs the small risk of overdiagnosis."
Peter Sasieni, Academic Director of the Clinical Trials Unit and Professor of Cancer Prevention at King’s College London, and joint principal investigator, said:
“Overdiagnosis is a quite complex issue. If we had stopped following women at age 70, nearly 1% of women tested would apparently have been overdiagnosed, but by waiting another 7 years, more women than never tested will be diagnosed with breast cancer and three-quarters of excess cancers turn out to have been diagnosed early rather than overdiagnosed. What this means is that most previous studies have overestimated the harms of participating in breast cancer screening until age 70.”
This study was funded by the NIHR Policy Research Programme.
