Patients seldom present to their doctors with diagnoses—rather, they have symptoms or signs. The major challenge of medicine is being able to talk to the patient and obtain a history, then carry out a physical examination looking for pointers to their likely underlying problem. Our elders and, some would argue, betters in medicine had fewer tests available to them than we have today, and their diagnoses were often made solely from the history and examination. Of course, they would claim that their clinical acumen and skills were greater than ours, and that we rely too heavily on the huge armoury of laboratory and other investigations available today. This, in part, is probably true, but we cannot ignore the fact that advances in science and technology have spawned a bewildering array of very useful and sophisticated tests that help us to confirm our diagnostic suspicions.
By ‘test’ we mean the measurement of a component of blood, marrow, or other body fluid or physiological parameter to determine whether the patient’s value falls within or outside the normal range, either suggesting the diagnosis or, in some cases, actually making the diagnosis for us.
Many measurable body constituents vary throughout life. For example, a newborn baby has an extremely high haemoglobin concentration, which falls after delivery. This is completely normal and is physiological, rather than pathological. A haemoglobin level this high in an adult would be pathological, since it is far outside the normal range for the adult population.
Factors affecting measurable variables
•Physiological conditions (e.g. at rest, after exercise, standing, lying).
•Sampling methods (e.g. with or without using a tourniquet).
•Container used, e.g. for blood sample, as well as anticoagulant.
These are published for most measurable components of blood and other tissue, and we have included the normal ranges for most blood and cerebrospinal fluid (CSF) analytes at the end of the book.
A really good test, and one that would make us appear to be outstanding doctors, would be one that would always be positive in the presence of a disease and would be totally specific for that disease alone; such a test would never be positive in patients who did not have the disorder. What we mean is that what we are looking for are sensitive tests that are specific for a given disease. Sadly, most tests are neither 100% sensitive nor 100% specific, but some do come very close.
Rather than request tests in a shotgun or knee-jerk fashion where every box on a request form is ticked, it is far better to use the laboratory selectively. Even with the major advances in automation where tests are batched and are cheaper, the hospital budget is finite and sloppy requesting should be discouraged.
Outline your differential diagnoses: what are the likeliest diseases, given the patient’s history, examination findings, and population from which the patient come?
Decide which test(s) will help you make the diagnosis: request these and review the diagnosis in the light of the test results. Review the patient and arrange further investigations as necessary.
It is important to remember that tests may often give ‘normal’ results, even in the presence of disease. For example, a normal electrocardiogram (ECG) in the presence of chest pain does not exclude the occurrence of myocardial infarction with 100% certainty. Conversely, the presence of an abnormality does not necessarily imply that a disease is present. This, of course, is where clinical experience comes into its own—the more experienced clinician will be able to balance the likelihood of disease with the results available, even if some of the test results give unexpected answers.
| Sensitivity and specificity | |
| Sensitivity | % of patients with the disease and in whom the test is positive |
| Specificity | % of people without the disease and in whom the test is negative |
This simply does not exist. Talking to patients and examining them for physical signs and assimilating knowledge gained in medical school are absolute requirements for attainment of sound clinical judgement. Those students and doctors who work from books alone do not survive effectively at the coal face! It is a constant source of irritation to medical students and junior doctors, when a senior doctor asks for the results of an investigation on the ward round and you find this test is the one that clinches the diagnosis. How do they do it? Like appreciating good wine—they develop a nose for it. You can learn a great deal by watching your registrar or consultant make decisions. This forms the basis of your own clinical experience.