Margaret McCartney: Cancer strategy should be led by evidence

Harpal Kumar, chief executive of the charity Cancer Research UK (CRUK), is to chair the new Cancer Strategy Taskforce, which will promote “swifter diagnosis.”1 Who could argue with that? Well, I have some concerns.

GPs are in an invidious position. Few symptoms, conceivably, could never represent cancer. Referring every patient with any symptom for further tests would do enormous harm to the over-investigated population—and to the people whose worrying symptoms are investigated less swiftly because of the waiting list that GPs have created.

New draft guidelines from the National Institute for Health and Care Excellence (NICE) recommend lowering the bar so that patients are referred for further investigations when their calculated risk of cancer reaches 3%.2 NICE also proposes less sensitive tests. So, for example, it recommends a serum cancer antigen 125 test for women with persistent abdominal distension, but ultrasound examination for ovarian cancer is more accurate.3

NICE also suggests cheaper tests to assess the risk of bowel cancer in symptomatic patients, such as faecal occult blood in stool—yet we have accumulated evidence that a negative test is not reliable enough to exclude a serious underlying cause.4 Are less good tests going to cancel out any benefit of a lowered investigative threshold?

Cancer is not the only condition worth diagnosing, and tests with high rates of false negatives will lead to uncertainty. Cost effectiveness does not justify anxiety in patients or doctors. General practice is a smorgasbord of unexplained symptoms, illness, cultural expectations, rituals, slender positive predictive values, red flags that usually don’t predict cancer well, and frequent sub-threshold aches and pains that are probably nothing sinister and rarely something to be concerned about.

CRUK is “not waiting” on the taskforce, however, but is pressing ahead with various work, including “projects” on instigating self referrals for diagnostic tests, as well as its current public “awareness” campaign, Be Clear on Cancer.5 (This advises people who have had indigestion most days for three weeks to see their GP. Pilot studies showed 10% more referrals than in control areas but no statistically significant increase in diagnosing upper gastric cancers.)

CRUK has recommended that women should have breast screening,6 and it wants to increase the uptake of bowel cancer screening.7 Ethically, however, doctors should promote informed choice because screening can do harm. CRUK is also responsible for a recent fundraising campaign urging people to “battle cancer,” to the dismay of many patients.8

We need a task force whose examination of early diagnosis is led by the fair use of evidence, grounded in an understanding of the highly complex world of general practice, and which is capable of recognising ineffectiveness and the harm resulting from hype.