A recent article in The Conversation’s Health Rationing series endorsed the government’s decision to extend the BreastScreen program to women aged 70 to 74 (from 50 to 69), based on the results of a 2009 cost-effectiveness analysis of the program.
But the landscape of breast cancer screening changed irrevocably in late 2012, with the publication of a report by an independent UK panel that reviewed breast cancer screening.
The panel confirmed that screening reduces the risk of dying from breast cancer by about 20%, but also concluded that screening causes over-diagnosis. Over-diagnosed cancers are those that would never have been found without screening. They are not destined to cause symptoms or become life-threatening but, nonetheless, they lead to more women getting treatment for breast cancer.
The panel estimated that 681 cancers would be diagnosed for every 10,000 UK women screened from 50 years of age for 20 years. Of these 681 cancers, 129 would represent over-diagnosis and 43 deaths would be prevented.
In other words, for women invited to be screened, the chance of avoiding dying from breast cancer is about 0.4% and the chance of being over-diagnosed and over-treated is about 1.3%. So, for every breast cancer death prevented, three women will be over-diagnosed and over-treated.
Breast cancer screening is a finely balanced trade-off of benefit versus harm. And there are important question marks over the argument to expand the program based on cost-effectiveness analysis alone.
This is particularly problematic because the analysis itself is based solely on survival statistics, when there are also major impacts on quality of life from both potential harms and benefits of screening.
More important than economic considerations, such as whether the program’s “cost per life-year saved” is improved by screening older (or younger) women, is the question - will we do more good than harm by expanding screening?
The 2009 cost-effectiveness analysis that was used to assess the implications of expanding the screening program was done as part of a wide-ranging evaluation of the BreastScreen Australia program.
That evaluation also included an assessment of the mortality benefit (percentage of breast cancer deaths averted) by screening in Australia. It found that, while screening significantly reduced the risk of dying from breast cancer among women aged 50 to 69 years, there was no significant reduction in breast cancer mortality among women aged 70 to 74 years.
So why have we expanded a program that will likely harm women over the age of 70 through over-diagnosis, without clear evidence that it will deliver a benefit for this age group?
While the 2009 cost-effectiveness estimates included treatment costs, they did not explicitly consider the effects of over-diagnosis and subsequent over-treatment on either survival or on quality of life.
The UK panel’s report suggested that new cost-effectiveness estimates should be made, taking account of over-diagnosis. This needed to be done before the decision to expand the program was made.
The 2009 BreastScreen Australia Evaluation recommended a focus on increasing screening participation rates through social marketing and other strategies. But is it ethical to promote screening to women in their 70s when we cannot demonstrate that it will reduce the risk of dying of breast cancer in this age group?
Might we not inadvertently pressure older women to undergo breast cancer screening and treatment that they may not need nor benefit from? An example from recent research conducted by our group is revealing in this regard.
When presented with information about both the benefit of screening and the risk of harm through over-diagnosis, one study participant was angry that she and her mother (aged in her 70s) had not received this information before her mother underwent breast cancer treatment. Her mother died shortly after the operation and our participant believed the death was a direct consequence of treatment.
Breast cancer screening currently costs an average of A$136 million a year. The expansion of the program is being funded at a cost of an additional A$55 million. The opportunity cost (the forgone benefit of alternative use/s of the money spent) of the expansion is considerable.
All things considered, there may be better ways to address the burden of breast cancer that would be more beneficial and less harmful. Such strategies might include more resources directed towards better information about the benefits and harms of screening, as called for by the independent UK panel, to ensure women are able to make informed choices to screen or not.
Deliberative methods, such as citizen juries may also be effective way to proceed. These methods allow evidence on complex health decisions to be presented to communities so they can make informed recommendations about where they feel government funds would be best invested.
Better information and better consultation initiatives are priorities for screening programs now that the harm of over-diagnosis has been confirmed. Or, money could be directed to primary prevention of breast cancer through lifestyle modification, breast cancer research or other health-care priorities.