Just How Effective are Medical Screening Tests?
Mon, 17 Nov 2003
A investigative report in the Guardian (UK) takes a critical view of medical screening for early detection of latent diseases. Contrary to the promoters of medical screening tests, there is scant evidence that early detection actually saves lives. The Guardian raises doubts about whether the risks (from false positives and subsequent, unnecessary invasive interventions), the cost, and the anxiety and discomfort are worth it. Essentially, medical screening tests are experimental fishing expeditions.
To save a single life from cervical cancer, 1,000 women have to be screened once every five years for 35 years. Similarly, the US baby-screening program “detects just one genetic abnormality in 7,800 scans and inevitably, since testing is imprecise, involves many false alarms.”
Other questions arise: What value are screening tests if there is no cure or effective treatment for a given disease? And most importantly, screening tests and early intervention may have disastrous results if radical procedures (surgery) are done when a benign form of a disease requiring no intervention, is wrongly identified as lethal. An example of large scale harm from mass screening occurred in Japan where 80% of infants were being screened for Neuroblastoma tumors, the second most common form of early-childhood cancer, after leukemia. When neuroblasoma was detected the infants underwent surgery, radiation and chemotherapy. But it was learned that most of the babies had a benign form of neuroblastoma requiring no intervention. What’s more, the number of deaths from the lethal form of neuroblastoma remained the same.
See: WHY MAMMOGRAMS ARE NOT THE ANSWER. Search and Destroy
by Shannon Brownlee April 12, 2002
The Guardian reports (below): “So far, the technology is not sufficiently advanced to distinguish between cancers that require treatment and those that do not. But the impulse of most doctors, given any finding of cancer, is to take action – which at the least will involve the removal of tissue for further testing, and may also mean more drastic surgery, a course of drugs, radiation or chemotherapy.”
The main rationale for screening for disease appears to be Big Business– like much of medicine today.
Just how effective are medical screening tests?
More than you need to know?
These days, we are screened for a whole raft of diseases, from cervical and prostate cancers to heart defects. We’re told such tests save lives. But just how effective are they?
Jerome Burne investigates
Saturday November 15, 2003
One of the rituals at boys’ schools in the 1950s was the yearly medical inspection. We’d all line up in our underpants and, one by one, have our chests listened to, our throats looked at, and then came the bit that made the occasion so memorable – the doctor would cup our testicles in his hand and order us to cough. What he was testing for we were never told, and no one ever confessed to having some terrible disorder identified as a result, so we were left to indulge in lurid speculation.
How primitive that all seems today. Hi-tech early diagnosis of disease is one of the fastest-growing areas of medicine – if you haven’t already seen brilliantly sharp colour pictures of your heart and lungs or read a print-out of your DNA, with the variations linked with disease highlighted, you very likely soon will. In the US, four million newborn babies are screened annually for more than 30 different genetic disorders, most extremely rare (a decade ago, a mere two or three could be identified this way). In the UK, such tests on babies are so far carried out only in very particular circumstances, but as adults we can send off $200 (£120) and a swab from the inside of our cheek to a laboratory in the US and get back a list of the genes that may be “your ticking time bombs for illness”.
Within the past year, several private diagnostic centres have opened in London. One can show you images of your heart that would not disgrace the pages of a Dorling Kindersley home anatomy textbook and can spot problems long before other methods. Another combines conventional and complementary methods to test your immune system and the function of individual organs. Meanwhile, we are all urged to have regular screening for various cancers, notably prostate, cervical and breast. All of which makes sense, surely – after all, cars have a regular service to spot potential problems before they leave you stranded on the motorway, so shouldn’t we check to see which of our bits are wearing out?
Not necessarily. There is a growing lobby that says such screening is not nearly as good an idea as it sounds. So far, the technology is not sufficiently advanced to distinguish between cancers that require treatment and those that do not. But the impulse of most doctors, given any finding of cancer, is to take action – which at the least will involve the removal of tissue for further testing, and may also mean more drastic surgery, a course of drugs, radiation or chemotherapy. Nor is it clear that screening achieves its main purpose – saving lives – on the scale that might be expected.
In April, for instance, a report in the British Medical Journal (BMJ) on the cervical screening programme in Bristol concluded that 1,000 women would have to be screened once every five years for 35 years to prevent just one death from cervical cancer (the least common of the three cancers widely screened for). This amount of testing involves an awful lot of women suffering anxiety from abnormal results and subsequent treatment. Similarly, the US baby-screening programme detects just one genetic abnormality in 7,800 scans and inevitably, since testing is imprecise, involves many false alarms. In July, another BMJ feature concluded that women must be told more plainly about the harm that can result from breast screening. In other forms of screening, the cost can be more than just anxiety or a biopsy. The latest figures, published in the Lancet, indicate that thousands of men who have invasive treatments for localised, symptomless prostate cancer will be left impotent and incontinent because of removal of a tumour that, if left alone, would have caused no harm.
The father of mass screening was UK physician Horace Dobell who, in 1861, outlined the benefit of regular check-ups of “ostensibly healthy persons” for tuberculosis. Like today’s supporters of scanning, Dobell assumed that “therapeutic efforts are more effective at earlier stages”. The idea caught on, with encouragement from the health insurance industry. But by 1918 one of the biggest drawbacks, one that is still with us, had emerged: the boost it gave to unnecessary medical treatments.The Journal of the American Medical Association that year reported on a study carried out among people considering life insurance, and concluded that almost everyone had some form of impairment and “more than 50% need medical or surgical attention”. However, the popularity of screening continued unabated and by the 1950s no senior US executive felt he had arrived until he’d been sent for a three-day, multi-systemic diagnostic survey at the Greenbrier Clinic, West Virginia.
The industry took something of a battering in the 1960s when sceptics began asking whether this extensive testing was making any difference. Two large trials, one in south-east London, the other in the US, followed up a total of 17,000 people over a decade to see if there was any eventual difference between those who had been screened and those who had not. They found “no convincing evidence for the effectiveness of periodic health examinations in decreasing morbidity or overall mortality”.
But the findings did little to halt the screening juggernaut. “For the past 20 years, regular screening has been a medical article of faith,” says Dr Angela Raffle, consultant in public health medicine in Bristol, consultant to the National Screening Programmes and author of the recent BMJ report on the outcomes of cervical screening. “Anyone who got up at a conference and asked for greater honesty about the harms as well as the benefits was shouted down as a heretic.”
The fundamental problem with cancer screening is what Raffle calls the dysplasia swamp, dysplasia being the medical term for abnormal cells. “If a pathologist looks at enough bits of you, he’ll find precancerous tissue somewhere. It is so common that it must be part of the normal process of tissue damage and repair. What we don’t know is which ones are going to spread and which can be safely left. That’s why we have to over-treat in order to help the small number who’d otherwise end up with something serious.”
Raffle is also involved in Bristol’s breast-screening programme which covers a million people. “Every year we detect 130 cases of breast cancer through screening,” she says. “Most of these women are very grateful and feel they owe their lives to the programme. But, extrapolating from the recent Office for National Statistics figures, the annual number of deaths prevented in our screened population is six. So, even in most of the women whose cancer [we] detect, screening has made no difference. Either they have a type of tumour that would have been curable even if left until symptoms developed, or they have a progressive form of the disease that will kill them anyway, despite being picked up on screening. There is only a small window of opportunity when a cancer is at the stage where screening can make that all-important difference.”
This sort of analysis suggests that the decision to offer mass screening is as much social and political as scientific and medical. “The fervour with which it has been promoted,” says Raffle, “means that in the past there was little proper scientific scrutiny of screening, but there is a serious discussion to be had. If you have £10m to spend, should you run a screening programme – or should you spend the money on other things, like better diagnosis and treatment, and on some of the things that the NHS doesn’t do well, like home care or prevention or palliative care?”
The debate over prostate screening is even more polarised. The public view is captured by doctor-cum-comic Phil Hammond, who has a line in one show that goes, “With a white coat on, I can stick my finger up your arse within a minute of meeting you and you’ll be grateful.” And it’s true, men have long been told that the “digital rectal examination” for possible prostate cancer is in our best interests. The statistics are certainly alarming – prostate cancer affects one in 13 men in the UK and is set to overtake lung cancer as the leading cause of death from cancer in males by 2005. So shouldn’t we be doing something about it?
Once again, the problem centres on the dysplasia swamp. Many people with a positive PSA (prostate specific antigen) test will have a slow-growing form of cancer that will cause few problems if left alone. However, a few will have the fast-growing, dangerous sort. “The tragedy,” says David Dearnaley, of the Institute of Cancer Research, “is that, at present, we have no way of telling which is which.” As a result, far too many people are being unnecessarily treated. The review of prostate screening in the Lancet put it like this: if a million men are tested, 110,000 will have raised levels of PSA, 10,000 of whom will have surgery; and of those, only 16% with localised tumours will benefit, because the disease in the remaining 84% would not have developed to a stage where it actually affected them. What makes the situation worse than cervical screening is that the results of surgery for prostate cancer can be far more drastic – of the 10,000 operated on, 10 will die, 300 will lose control of their bladder and 4,000 will become impotent.
So men diagnosed with prostate cancer have a stark choice (which may or may not be put to them): refuse surgery and have a 16% chance of the disease becoming acute and possibly fatal; or submit to surgery and face a 40% chance of being made impotent.
The medical profession is deeply split on the issue. In the US, the American Urological Association says that all men over 50 should have a check, yet the US Preventive Services Task Force advises against it. And the Lancet review concludes, “There is no clear association between intensity of screening and reduced prostate cancer.” Given this degree of doubt, you might think there would be some resistance to introducing a new form of even more detailed mass screening, especially one that involves a dose of radiation. Apparently not: one of the more fashionable gifts in the US, soon to arrive in the UK, is the whole-body CT (computerised tomography) scan, which searches for the minutest sign of cellular irregularity that might be cancer. This is rummaging in the dysplasia swamp with a vengeance.
When Professor Stephen Swensen of the US National Institutes of Health ran whole-body CT scans on 1,570 patients last year, he found 700 indicators of disease, including breast, kidney and stomach cancers. Most turned out to be false-positives but they all had to be checked out, a process that was worrying, time-consuming and unpleasant. As Swensen puts it, “The investigations adversely affected the quality of life and resulted in unnecessary diagnostic and interventionist procedures.” Not such a cool gift, then.
So what is the prudent health consumer to do? Despite the worrying statistics, there is still something reassuring about having a scan that says you’re OK. And then there is the lottery factor: you might be one of those who benefits from screening. What is needed is the habit of frank discussion with your GP. “The problem is that very little is known about the effects of sharing research-based information about healthcare effectiveness with patients and involving them in decisions about their care,” says Joan Austoker, of the department of primary care at Oxford University. In other words, will it put people off doing it?
To get mass-screening programmes under way, governments and the medical profession have invariably presented them as effective, simple and cheap. “They are often none of these things,” says Austoker, “but no one knows whether it would stop people from attending if they knew the odds.” The only way forward is to move from a paternalistic to a more democratic approach. “Above all, we need to respect patients’ autonomy, and that includes their right to decide not to undergo a screening intervention.”
What makes the scanning debate so tricky is that the criticism is counter-intuitive. It seems so obvious that catching a disease early is a good thing that it can be hard to appreciate why that isn’t necessarily true. To make matters worse, much of the debate turns on statistics – a topic that induces blank incomprehension in most people. Saying that scanning reduces the number of cancers by 50% makes it sound very worthwhile, but it’s not so impressive if that reduction is from two in 1,000 to one in 1,000. The first is called relative risk and the second absolute risk.
Presenting the information visually can make the risks and benefits much clearer. A chart recently published by Dr Tom Marshall of the department of public health and epidemiology at the University of Birmingham shows at a glance what would happen to women aged 50 who were regularly screened for breast cancer, compared with women of the same age who weren’t. This reflects, in simplified form, his contention that, in a non-screened group of 1,000, 762 would reach the age of 75, while in a screened group 768 would live that long. In other words, six women in the screened group will benefit, 218 will be referred for further tests and 68 will have a breast biopsy. “There’s a lack of clear and accurate information available to women,” he says. “We go out of our way to persuade women to have breast screening – the least we can do is to give them the facts.”
Of course, one way to improve the rather lousy odds of mass screening is to choose your patient base rather more carefully. If you test people who are already known to be at high risk, this greatly increases the chance that the signs of disorder you do pick up are going to be significant. The odds improve still further if your marker is a definitive indication that you have the disease.
It is this combination that is on offer at the European Scanning Clinic, which opened recently in Harley Street, London. Expensive stems of tropical flowers stand in tall vases; friendly assistants are on hand to help. This is private medicine at £500 a scan.
The clinic’s main advantage is an EBT (electron beam tomography) scanner that is much faster and delivers far less radiation than regular CT scans. On computer monitors, you can see the bone of the ribcage as hard-edged white with a lacy decoration of brilliant red blood vessels. The purpose of such textbook detail is to calculate your calcium count. Although countless healthy heart campaigns have firmly implanted the notion that a build-up of fatty plaque is what causes heart attacks, the danger sign is when the plaque gets turned into calcium. This machine can spot even microscopic specks of calcium in the arteries around the heart. “What’s nice about EBT is that it is absolute,” says the clinic’s publicity manager, Barry Burles. “Rather than just indicating that there may be a problem, it is definite. If you have calcium, you have heart disease.” At the moment, the only people referred to the clinic are those who have a number of the familiar risk factors for heart disease – overweight, smoking, family history, etc – but who want to be certain.
With the EBT (unlike the standard angiogram, a test to detect blockages of the arteries which involves inserting a catheter into an artery in your groin and threading it up to your heart), you can see the heart at work and blockages show up clearly. “It does give the impetus to change lifestyle,” says the chief radiographer, Tony McArthur. “Everyone, especially men, think they are 10ft tall and bulletproof. Actually seeing what’s wrong can be a wake-up call.”
Once heart disease has been confirmed, then you are referred back to your GP for treatment, which at the moment means statins – cholesterol-lowering drugs. “If there were no statins, there would not be much point in EBT,” says Burles. A trial is currently under way to see how much difference statins treatment makes to the calcium score year on year.
All of which raises yet another of the controversial issues surrounding mass screening: is it largely a device to increase sales of a particular drug? While it is obviously pointless to scan or test for conditions that have no treatment, if a positive finding does not lead to surgery, it usually means years of taking some drug or other to reduce your risks.
The big hope of those involved in promoting large-scale screening is that rapid advances in genetics and a new form of molecular analysis known as metabolomics will come to their rescue. These will allow doctors to detect people who are at higher risk of disorders, either by spotting damaging mutations in their DNA or by identifying dangerous combinations of molecules in their blood.
Metabolomics isn’t being offered to the public yet, because it requires very expensive scanning equipment, but it can already pick up early signs of a range of disorders by detecting their “metabolic signature” in blood or urine. While most conventional tests look for change in just a single molecule, metabolomics uses a sophisticated computer programme, originally developed to spot signs of credit card fraud among billions of transactions, to detect changes in their combinations.
This approach may soon lead to a huge improvement in the accuracy of prostate screening. Researchers at the University of Minnesota have reported on a way to distinguish the slow type of prostate cancer from aggressive ones. One common feature of these hi-tech screening techniques is that they are impersonal. But Kim Jobst at the newly opened Diagnostic Clinic in New Cavendish Street, London, looks into your eyes. “You can learn a lot from the eyes,” he says. “They tell you about cholesterol and your blood-sugar level, and if you’ve got any inflammatory processes at work.” He also relies on reports from a Chinese medical practitioner who examines your tongue. This reflects the clinic’s aim, to straddle the divide between the complementary world of “low energy in the gall bladder” and the conventional world of electrocardiograms.
“We’ll give you the same sort of standard physical check-up that you’ll get from a Bupa testing centre,” says Jobst, who is a regular doctor and a practising homeopath, “but we also take a more complementary and systemic approach. We believe that what’s going on in your guts may well be having an effect on how you are fighting off cancer, or that correcting vitamin and mineral deficiencies can lower high blood pressure and your risk of heart attacks.”
The place is a hypochondriac’s heaven. After a package of conventional tests (ECG for heart rhythm, PSA, triglycerides, cholesterol), I went through a combination of complementary ones (blood and hair analysis for mineral and vitamin levels). Then I was hooked up via electrodes to a couple of machines in turn: one does a “bioresonance analysis” and is said to be especially useful for spotting food sensitivities, lurking bacteria, viruses and parasites; the other is an AMI machine, which provides information about spinal alignment and internal organs using acupuncture meridians.
The end result for me was a report that ran to 50 or so pages. I’d passed the conventional tests with flying colours, but the complementary scan revealed a more complex picture, involving stress, a deficiency in the immune system and low levels of zinc, sodium and potassium. Bioresonance and AMI both pointed to some inflammation in my guts and AMI picked up on an old schoolboy lung infection, as well as a long-standing problem with my lower back.
Now, all of this would undoubtedly fall foul of Angela Raffle’s passion for evidence. Do we know that people who showed up with signs of gut problems on the bioresonance machine actually went on to develop problems? Have there been any controlled trials run over 10 or more years to see if those tested and treated did any better than those left alone?
The answer to these questions is no. Was there any evidence that, if I took the suggested supplements religiously, I would actually be, as Jobst put it, “a patient at 85 who’d drop in occasionally”? Again, the answer has to be no. There were also similarities with the vogue for whole body scanning – test for enough biomarkers and you are bound to come up with signs of abnormality. The Diagnostic Clinic offers more than 150 tests.
And yet, and yet … The decision to fund mass-screening programmes is a complex one in which are intertwined science, economics and the desire by both the government and the medical profession to be seen to be caring. The factors affecting a personal decision to have a screen are equally complex. You will not merely be reassured or warned – usually, the type of scan you have implies certain forms of treatment if something is spotted.
So, despite the lack of hard evidence, I feel personally drawn to Jobst’s approach. His account of how low levels of certain minerals and vitamins, plus poor functioning of your gut, can combine to create illness makes sense. Correcting such deficiencies seems a sensible first step in dealing with warning signs such as raised blood pressure or high cholesterol. If that doesn’t work, drugs are always a second option. And whatever the benefits may be of my supplement shopping list, taking them seems very unlikely to cause unpleasant or damaging side-effects.
Mass screening can undoubtedly save lives, but people are not being clearly informed of the pros and cons. Most people wildly overestimate both their risk of getting cancer and the protection that screening offers them. We are all health consumers now, and we should be given a much clearer statement of what is involved. What’s more, as the number of private hi-tech and genetic scans proliferates, it is worth arming yourself with the same sorts of hard questions that researchers have been asking about the public programmes. It was all so much simpler when all you had to do was cough
Guardian Unlimited © Guardian Newspapers Limited 2003
FAIR USE NOTICE: This may contain copyrighted (© ) material the use of which has not always been specifically authorized by the copyright owner. Such material is made available to advance understanding of ecological, political, human rights, economic, democracy, scientific, moral, ethical, and social justice issues, etc. It is believed that this constitutes a ‘fair use’ of any such copyrighted material as provided for in section 107 of the US Copyright Law. In accordance with Title 17 U.S.C. Section 107, this material is distributed without profit to those who have expressed a prior general interest in receiving similar information for research and educational purposes. For more information go to: http://www.law.cornell.edu/uscode/17/107.shtml If you wish to use copyrighted material for purposes of your own that go beyond ‘fair use’, you must obtain permission from the copyright owner.