Health: When doctors haven't the faintest clue

Despite advances in medical science, doctors often have no idea what is wrong. Now they are beginning to admit as much to their patients.

FOR THREE years, David Rogers had suffered with chronic low back pain. He endured it for some time hoping it would go away, before finally deciding to see his GP to find out exactly what was wrong. But when he had his consultation with his doctor, and then a second one with an orthopaedic specialist, both were unable to explain what was wrong, or to give him a diagnosis.

"I was surprised. I'd expected to be told what was causing it and get pills or have surgery or whatever. They just said `it's low back pain; we don't know what's responsible, but try some exercise and, if you need them, here are some painkillers.'"

Most patients believe and expect that a doctor will always be able to tell them what is wrong. The popular perception is that a diagnosis will be produced without fail from a vast and mysterious diagnostic treasure chest.

But for up to 30 per cent of patients, that's not the case. In ER and Casualty there may always be an answer, but in real-life medicine a substantial number of patients never do get a positive diagnosis. They simply have their symptoms controlled, rather than the problem identified and treated.

In the past, patients such as these were often told that there was nothing wrong with them, given a diagnosis based on the doctor's hunch, or informed that their complaint was all in the mind. The golden rule then was that doctors should never say they didn't know, because such an admission would undermine the patient's confidence.

But now a new type of so-called negotiating diagnosis is being tried by some GPs. Patients are given a range of options about what could be wrong with them, and a number of choices on what to do about it.

In most cases diagnosis is relatively straightforward. The easier cases are those where there are classic or specific symptoms, and where blood tests, X-rays, scans, biopsies, and other examinations, tests and exploratory procedures are available.

The worst cases to diagnose are those with non-specific symptoms, where no test is available. These include chronic fatigue syndrome, depression, irritable bowel syndrome, dry coughs, back pain, and the early stages of some acute conditions, including meningitis.

It has been estimated, for instance, that up to one-third of patients with back pain never get a definitive diagnosis. Some acute conditions, such as life-threatening meningitis, are also difficult to diagnose, particularly in the early stages of the disease.

Researchers have found that 10 per cent of people with what's called a "non-productive cough" never find out what is wrong with them. Coughing is a symptom that is commonly associated with serious chest and lung conditions, but other potential diagnoses range from post-nasal drip syndrome to gastro-oesophageal reflux, and even a special type of asthma.

"Patients do have a perception that we will always know what is wrong, but as doctors we know we are dealing with a very inaccurate science, and we use different standards for different circumstances,'' admits Dr Catti Moss, vice-chairman of the Royal College of General Practitioners' Patient Liaison Group. "With a diagnosis we don't ever get the concept of complete definiteness. You can get to a probability of 99.9, but it is still not absolute. A good doctor is one who will not be as definite as the patient wants him to be, and a bad doctor will give a definite diagnosis if there is a 97 per cent probability.''

One of the dilemmas for doctors is that patients expect their visits to the surgery to be productive, and want both a diagnosis and a prescription for a remedy.

"They don't like uncertainty, and general anxiety is worse than specific anxiety. Once they are told what is wrong they are less anxious," says Dr Lance Workman, a psychologist at the University of Glamorgan. "If they don't get a diagnosis there and then, they often assume the worst... irritable bowel syndrome becomes bowel cancer, a sore throat is cancer of the larynx, and an unexplained headache is a brain tumour.''

Researchers at California University, which is hosting a research conference this month, have been looking at the problems of diagnosis. They say that although developments in scanning and testing technology have increased the proportion of definite diagnoses, there is still a huge grey area.

One way of dealing with the lack of a definite diagnosis is to prescribe medication. It would be unethical for a doctor to prescribe a pill knowing that it could have no effect. But it has long been suspected that GPs have been giving out antibiotics for conditions for which such drugs could never be effective.

At her surgery in rural Northamptonshire, Dr Moss has been dealing with the problem in a different way, by using what she calls a "negotiating diagnosis" with many of her patients.

"I reckon to make a working diagnosis in 70 to 80 per cent of the cases," she explains. "Very often that working diagnosis will include two possibilities, and the way I deal with patients is to explain what those possibilities are and what the likelihoods are, and negotiate with them about the next step - what tests are needed, and so on.

"They are happy with that approach; it is less threatening and empowers them to make choices with my support and advice. But it is not an approach that is generally used. Giving options is something that is only now coming in. I remember being told off by my trainer in 1980 for telling the patient I didn't know. Never do that, he said, because you will lose credibility.''

Although that kind of paternalistic approach to patients is fast disappearing, Dr Moss says that doctors themselves need to appreciate that all the patients who turn up at their surgeries do have something wrong.

"There will be something wrong with them, even when we can't find anything. People don't complain maliciously very often - and even when they do, of course, there is something wrong with them."