Study Questioning UK Breast Screening Program Is Flawed

Zosia Chustecka

June 14, 2013

A study claiming that the national breast screening program in England has not reduced mortality from breast cancer, widely reported this week in the British media, is seriously flawed, a mammography expert told Medscape Medical News.

The study, published in the June issue of the Journal of the Royal Society of Medicine, analyzed mortality statistics for England and for a region near Oxford to look at the rate of breast cancer deaths before and after the national breast screening program was launched in 1988.

The researchers, Toqir Mukhtar and colleagues from the Department of Public Health at Oxford University, United Kingdom, conclude that "mortality statistics do not show an effect of mammographic screening on population-based breast cancer mortality in England."

This conclusion was widely reported in the British media, and several top newspapers took this as evidence that the national breast screening program is not working. There has been a fierce debate about this program in recent years. A government review of all the data concluded last year that the program does save lives, but also leads to overdiagnosis.

Recent media reports have focused on the contradiction between the finding by the Oxford team of no reduction in breast cancer deaths and that of the government-backed review. For instance, a front-page article in The Telegraph proclaimed that "25 years of breast cancer screening has failed to significantly reduce deaths from the disease, according to a landmark study."

Far from being a landmark study, this research has "flawed methodology" and was rejected by a cancer journal when previously submitted for publication, Daniel Kopans, MD, professor of radiology at Harvard Medical School and senior radiologist at Massachusetts General Hospital, in Boston, told Medscape Medical News.

"This is one of those papers where one has to wonder what happened to peer review," Dr. Kopans noted.

"The major problem with analyses like these is the lack of direct patient data. They merely looked at registry information; they did not actually know whose cancers were found by screening and what percent of women who died were actually screened," he explained.

Dr. Kopans also questions the conclusion. When looking at mortality data presented in Figure 1 in the publication, "to my eye, it appears that the death rate from breast cancer was fluctuating above 40 per 100,000 women from 1979 to 1993. Since 1993, the death rate has been dropping rapidly. Given that the national breast screening program began in 1988, why do they not think that the decline in deaths is related to screening?"

Analysis of Mortality Data

For their study, the researchers looked at the Oxford region, using data compiled in the Oxford Record Linkage Study from 1979 to 2010. These data came from death registration data supplied by the Office for National Statistics, the researchers note.

The team found 20,987 death certificates with female breast cancer coded as the underlying cause or mentioned anywhere on the certificate.

Mukhtar and colleagues also analyzed data for the whole of England. For this, they used data on breast cancer incidence from 1975 to 2008 (982,135 cases) and mortality from 1971 to 2009 (444,186 deaths) from Cancer Research UK. In this dataset, all of the deaths were coded with female breast cancer as underlying cause, the researchers note.

The researchers used joinpoint regression to age-standardized trends in incidence and mortality rates, which allowed them to identify when a trend in a series of annual rates changes.

Their analysis showed that in the Oxford region, breast cancer mortality rates peaked in 1985 and started to decline prior to the introduction of the national breast screening program in 1988.

The researchers found that there was an annual reduction in breast cancer mortality from 1979 and 2009 of 2.1%. This was the same for women 50 to 64 years of age, who had been screened (the national program begins screening women at 50 years), as it was for women 40 to 49 years of age, who had not been screened.

When they analyzed the data for England, they found that the downward trend was greatest in women younger than 40 years. The mortality rate fell 2% per year from 1988 to 2001, and 5% per year from 2001 to 2009.

"There was no evidence that declines in mortality rates were consistently greater in women in age groups and cohorts that had been screened at all, or screened several times, than in other unscreened women, in the same time periods," the researchers write.

"We permuted the data in a number of different ways, over an observation period of 39 years, but the data show that, at least as yet, there is no evidence of an effect of mammographic screening on population-level breast cancer," they conclude.

Dr. Kopans, who was asked by Medscape Medical News to comment on the findings, noted that "there are many uncontrolled factors that cannot be determined when this type of analysis is used. The only way to really analyze the effect of screening is with direct patient data for individual women, so that it is known how the cancer was detected and what happened to the patients."

This type of data analysis with no direct patient information can be grossly misleading. For example, a large number of women who died after the introduction of the screening program were diagnosed before they had access to screening. Obviously, they could not benefit from it, Dr. Kopans explained.

J R Soc Med. 2013;106: 234-242. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.