IELTS Academic Reading Practice Test 401

In 1928, a Greek-born physician working in New York made a claim that much of his audience found hard to believe: that cancer of the womb could be detected by examining a few cells scraped from the vagina and spread on a glass slide. The claim rested on years of patient microscopy, yet it attracted little attention at the time, and more than a decade passed before the method he described became part of ordinary medical practice. The physician was George Papanicolaou, and the procedure that eventually carried his name has probably been performed more often than any other cancer test in history.

Papanicolaou was born in 1883 in Kymi, a coastal town on the Greek island of Euboea. He studied medicine at the University of Athens and, after qualifying, travelled to Germany, where he took up zoology, a subject that gave him a lasting interest in the structure and behaviour of cells. In 1913 he emigrated to the United States with his wife, Andromache, whom everyone called Mary. After a difficult period looking for academic work, he obtained a post at Cornell University Medical College in New York, and he remained there for the rest of his working life.

His early research concerned the reproductive cycle of the guinea pig. In order to establish when the animals ovulated, he prepared smears of cells from the lining of the reproductive tract, stained them and examined them under a microscope, and he demonstrated that their appearance changed in a regular sequence. Mary Papanicolaou, who worked as his assistant, supplied samples of her own cells so that he could check whether the same pattern occurred in women. It did, and by the late 1920s he was one of the few people in the world able to interpret a vaginal smear with confidence.

That skill led him to an unexpected observation. In samples taken from women with cancer of the cervix he saw cells that resembled nothing in healthy tissue: their nuclei were large, dark and irregular, and their shapes varied wildly from one cell to the next. He became convinced that a trained eye could recognise malignancy in such cells long before a tumour produced any symptoms. He presented his findings at a conference in Battle Creek, Michigan, in 1928 and published a short paper in the proceedings, suggesting cautiously that the method might be used to screen women who had no symptoms at all.

The reception was cool. The meeting was not a gathering of cancer specialists, and several pathologists who heard the idea doubted that cancer could be diagnosed without a piece of tissue cut from the tumour itself. Papanicolaou, discouraged, returned to his studies of the reproductive cycle and let the matter drop. Historians of medicine have sometimes described the episode as a clear case of a discovery arriving before the profession was ready to use it, although it is also true that his evidence at that stage was thin.

The work resumed in 1939, when he began a collaboration with Herbert Traut, a gynaecologist at Cornell. Together they examined smears from a large number of women and checked the results against tissue removed at operation, a comparison that had been missing from the 1928 report. Their book, published in 1943 as Diagnosis of Uterine Cancer by the Vaginal Smear, set out the evidence at length and included coloured illustrations of the cells. It was well received, and it did much to persuade gynaecologists that the technique was reliable.

Adoption was nonetheless gradual. The test required a doctor to take a sample, a laboratory to stain and examine it, and somebody to pay for both. In the United States, voluntary organisations that campaigned against cancer did much to promote it, and clinics offering free smears opened in several cities after the Second World War. Britain was slower: a national programme, which invited women at regular intervals, began there only in 1988. In Australia and the Nordic countries, registers that recorded every result made it possible to check that women who had been tested were actually followed up.

The effect on mortality has been considerable. In countries with long-running programmes, deaths from cancer of the cervix have fallen substantially since the 1960s, and the decline is generally attributed to screening, although better treatment has also played a part. The test changed the way doctors thought about cancer in general. Papanicolaou argued repeatedly that the smear should be offered to healthy women as a routine check rather than only to those who already had symptoms, and this idea — that a disease may be caught in a pre-cancerous state and treated before it does harm — underlies screening programmes of every kind.

Papanicolaou died in 1962, before the causes of cervical cancer were understood. In the 1980s the German virologist Harald zur Hausen showed that most cases are caused by a few strains of the human papillomavirus, work for which he later shared a Nobel Prize. Vaccines against those strains were licensed from 2006, and in many countries the smear is now being replaced or supplemented by tests that look directly for the virus. Yet the underlying principle — that a handful of cells on a slide can reveal a cancer while it is still curable — remains the one he established.

Screening is the testing of apparently healthy people so that disease can be found at an early stage, when treatment is more likely to succeed. It is quite different from diagnosis, which begins with a patient who has a complaint. A screening programme invites an entire population, or a defined group within it, to be tested, and the initiative lies with the health service rather than with the individual. Because the people tested have no symptoms, the great majority of them will be healthy, and that single fact shapes almost everything that follows.

The accuracy of a screening test is usually summarised by two numbers. Sensitivity is the proportion of people who actually have the condition and are correctly identified by the test; a test with 90% sensitivity misses one case in ten. Specificity is the proportion of people without the condition who are correctly given a negative result, so a test with 95% specificity wrongly flags five in every hundred healthy people. The two measures pull against each other. A laboratory can usually make a test more sensitive only by making it less specific, and vice versa, so deciding where to set the threshold is a judgement about which kind of error is more acceptable.

Neither number answers the question that a patient asks on receiving a positive result, which is whether the disease is really there. That is the positive predictive value, and it depends heavily on how common the condition is among the people tested. A hypothetical example makes the point. Suppose a test with 90% sensitivity and 95% specificity is applied to 10,000 people, of whom 1% have the condition. Of the 100 who have it, 90 are correctly identified. Of the 9,900 who do not, 5% — some 495 people — are wrongly told that they do. So 585 people receive a positive result and only 90 of them are genuinely ill, fewer than one in six. The same test used on a group in which the disease is ten times as common would give a far more encouraging picture.

This is one reason why screening is seldom worthwhile for rare conditions, and why concentrating it on higher-risk groups makes it more efficient. It also explains the cost that programmes impose on individuals. A woman whose mammogram looks abnormal, and who then waits a fortnight for further tests, has been harmed by the process even if she proves to be well. Studies of breast screening have found that a substantial minority of women who attend are recalled at least once for investigation, although most of them turn out not to have cancer.

Assessment is complicated further by biases that flatter screening. Lead-time bias arises because screening moves the date of diagnosis earlier without necessarily postponing death; someone diagnosed five years sooner who dies at the same age appears to have survived five years longer. Length bias is subtler. Screening is more likely to detect slow-growing disease, which is present in the body for longer before it causes symptoms than aggressive disease, so the cases it finds are on average less dangerous than those that arrive in a clinic. Overdiagnosis is the most contentious idea of all. Some abnormalities, particularly in the thyroid and the prostate, would never have caused illness if they had been left alone, yet they are found, labelled and treated, and the treatment can leave a patient incontinent or without an organ.

The South Korean experience shows how far the consequences can reach. From the late 1990s, ultrasound examination of the thyroid became widely available there, often as an add-on in clinics offering screening for other cancers. Within a decade the number of thyroid cancers diagnosed had risen roughly fifteenfold, while deaths from the disease barely changed. Prostate screening raises similar questions. Two large randomised trials produced apparently opposite results, one reporting a modest fall in deaths and the other none, and much of the disagreement has been traced to differences in how the trials were designed and how many men were treated for tumours that would never have troubled them.

Only randomised trials, in which people are allocated by chance to be screened or not, can escape these biases, and even they must measure the right thing. A trial that counts deaths from the disease is much harder to mislead than one that counts the proportion of patients alive five years after diagnosis. Well-conducted trials have produced encouraging findings. A large American study of heavy smokers, for example, reported that annual low-dose computed tomography reduced deaths from lung cancer by about a fifth, and screening of this kind is now offered to older smokers in several countries.

Where does the field stand now? Researchers are trying to raise specificity without losing cases, using risk models that combine age, family history and genetic markers to decide who should be invited at all. Blood tests that look for fragments of tumour DNA, often marketed as multi-cancer early detection tests, are being evaluated in large trials. Their promoters make ambitious claims; the honest answer is that nobody yet knows whether finding such traces early prolongs life, or merely lengthens the period during which a person is regarded as a patient. Every programme produces a predictable ratio of benefit to harm, and the least a health service can do is to state those figures plainly.

Patients in many countries are now told that they have a right to choose their doctor. In England, anyone may apply to register with any general practice whose list is open, and practices are expected to allocate each patient a named doctor. In Germany, patients may go directly to a specialist without asking anyone's permission; in the Netherlands they must register with a practice and obtain a referral before they can see a hospital consultant. In some systems the choice extends to a named clinician, in others it stops at the door of the building. The language of consumer choice has become so familiar in health policy that it is easy to forget how recently it arrived, and how unevenly it is distributed.

The appeal of choice is obvious. It treats adults as adults, rather than as supplicants dependent on the goodwill of an institution. It offers a route of escape from a doctor who is dismissive, incompetent or simply impossible to talk to, and it is especially valuable to patients whose needs are particular: a woman who would rather see a woman, a patient who needs a doctor who speaks her language, a family with a child who has a rare condition. On this view, the freedom to choose is not a luxury bolted onto a public service; it is part of what makes the service decent.

Yet much of what passes for choice is illusory. Choice depends on supply. A patient in a remote valley with one practice within twenty miles has nothing to choose between, and a patient in a city may find that every practice in reach has closed its list. Information is also unevenly distributed. It is difficult even for clinicians to judge the quality of a doctor's work, and patients, who see only the surface of it, tend to make their decisions on the basis of convenience, opening hours and hearsay. National patient surveys in Britain have repeatedly found that being able to get an appointment quickly, and seeing a doctor who knows them, matter more to people than the abstract right to pick a name from a list.

That last finding deserves attention, because continuity of care is not a soft preference. Seeing the same doctor over time allows a picture to be built up that a ten-minute consultation with a stranger cannot produce, and a growing body of research associates it with measurable benefits. Studies using large primary-care databases have linked higher continuity with fewer emergency admissions and with better control of chronic disease, and at least one analysis has reported lower mortality among patients who see the same doctor repeatedly. Correlation is not causation, and sicker patients may find it harder to move; but the pattern is consistent across countries.

Systems differ in how much choice they permit, and the differences have consequences. Allowing patients to consult specialists directly, as in Germany, is popular, but the result is fragmented care: a cardiologist, a dermatologist and a family doctor may each prescribe a different drug without knowing what the others have done. Requiring referral through a named general practitioner, as in the Netherlands and, since the mid-2000s, in France, preserves a coordinator but limits the patient's freedom. Neither arrangement is obviously superior; what matters is whether somebody is responsible for the whole patient.

There is also a question of fairness. The patients most likely to exercise choice are those who are articulate, mobile, well informed and confident in dealing with institutions. Those least likely to do so are the elderly, the poor, the housebound and those with mental illness — precisely the groups whose care is most likely to go wrong. Choice schemes can therefore widen the gap they were meant to close, and they can destabilise services: a popular practice gains patients faster than it can recruit doctors, while a failing one loses them without losing the obligation to look after the patients who remain.

None of this amounts to an argument for taking choice away. The ability to leave a doctor one does not trust is a genuine safeguard, and it gives practices a reason to respond to complaints. But it is worth noticing what kind of freedom this is. It is not the freedom of a shopper comparing brands; it is the freedom of a patient to end a relationship that has broken down, and its value depends on the alternatives actually being available and on the switch being quick and unashamed. A right that exists only on paper, or only for those who can navigate the system, is not much of a right at all.

The conclusion I would draw is that continuity, not choice, should be the organising principle of primary care. Every patient should have a named doctor who takes responsibility for their care and who is not changed without good reason, and every patient should be able to change that doctor promptly. Choice exercised once, when trust has failed, is worth more than a menu of options that nobody has the time or knowledge to use. Governments tempted to advertise choice as a substitute for investment in staff should be honest: the freedom to choose a doctor is not much use if there are not enough doctors to choose from.