The instinct, it turns out, is sound. Yesterday's paper reported research in the United States on screening for breast cancer that found one in three cancers detected and treated would not have been fatal.
That could be conservative. An Australian website of university comment (theconversation.edu.au) has been running a series by medical academics on something they call "overdiagnosis" from cancer screening.
This week four professors, Robin Bell, Bebe Loff and Robert Burton of Monash and Michael Baum of University College, London, put the names to a piece that asked, "Is routine breast cancer screening doing more harm than good?"
They cited a report in the Lancet last month of a United Kingdom screening review that estimates the ratio of over-diagnosis to deaths prevented at 3:1.
I thought a test could always tell whether a cancer was life-threatening. Apparently not. The article continued:
"As it is not currently possible to distinguish breast cancer identified through screening which will never cause harm, from cancer that will, all breast cancer identified by screening is treated - with surgery and various combinations of radiotherapy, chemotherapy, endocrine and biologic therapies.
"A woman treated needlessly will have no way of knowing that her cancer was harmless and that her wellbeing has been sacrificed, without her knowledge and consent, for the sake of others who may have benefited from screening."
It went on to suggest that the ratio of 3:1 was probably an under-estimate since it was based on trials in the 1970s and 1980s.
On the basis of advances in treatment since then, other researchers estimated a ratio of 10:1 and the authors of this article put it closer to 15:1.
I have been astonished in recent years at the number of women I know who have suffered breast cancer. Others remark on this too, and wonder if we will discover something seriously amiss in modern food or cellphones or something.
There seem to be far more women diagnosed and treated these days than we ever knew to have died of breast cancer before the introduction of screening and advances in treatment.
In every case, of course, the diagnosis has been terrifying for them and their families, and we have followed their progress through tests and gruelling treatment with constant anxiety.
Obviously the difficulty with overdiagnosis is that no matter whether the ratio is 3:1, 10:1 or 15:1, there is no way of knowing you are not the one.
The article does not oppose screening, it merely suggests women should be made aware of the overdiagnosis risk before treatment. I am not sure it would make much difference. If I was diagnosed with a cancer and given those ratios, I'd have the treatment every time.
If it was prostate cancer I'd accept the risk of infertility and incontinence from successful treatment. Better them than dead.
But I'd rather not expose myself too often to a possible overdiagnosis. I'd prefer to watch for symptoms than face more frequent checks.
They are delicately called "wellman examinations", which seems appropriate. If three out of four diagnosed cases would not advance, screening is a classic treatment of the "worried well" that is blowing public health budgets.
An earlier article on The Conversation website, by Professor Burton and colleagues Christopher Stevenson of Deakin and Mark Frydenberg of Monash, mentioned that an examination of men of various ages killed in motor accidents in the United States indicated that microscopic asymptomatic prostate cancer could be found in about one third of American men aged 30-39, rising to 70 per cent in ages 70-79.
Those numbers are far higher than the numbers who die of it. Burton and colleagues credited a 1993 study of prostate cancer with challenging what they called the "fatal cancer" myth. It found that 40 per cent of men could be treated for prostate cancer though only 8 per cent would have a cancer large enough to be detected and only 3 per cent would die of it.
Spare us the agony of tests that tell us less than we fear and more than we need to know.
