http://bmj.com/cgi/content/full/324/7336/0/g

BMJ 2002;324 ( 2 March )

Editor's choice

Diagnosis, diagnosis, diagnosis

What is it that doctors offer that other professionals cannot? "Diagnosis, diagnosis, diagnosis," answered a former chief medical officer for England. A nurse might one day transplant a heart, a technician anaesthetise a patient, and a pharmacist control a patient's complex drug regimen. But doctors are needed for diagnosis. The doctors' doctor is not a heroic surgeon cutting into a diseased brain but the doctor, probably a physician, who using almost magical skills can make the diagnosis when all about her are lost in a profusion of radiographs, printouts, and misleading conjecture. And quite right toobecause diagnosis is hard.

Diagnosis is also hard to study, and only a small fraction of studies in medical journals give reliable information on the usefulness of diagnostic tests. The Cochrane Collaboration, which is doing such a splendid job of sorting out the chaos of "the medical literature," has concentrated on treatments and stayed away from diagnostic methods. This is partly because it's much less clear how to analyse systematically studies of diagnostic methods.

To contribute to the debate on diagnostic studies BMJ Books has published The Evidence Base of Clinical Diagnosis, which is edited by André Knottnerus, and we are currently serialising the book in the journal. This week Dave Sackett and Brian Haynes, the godfathers of evidence based medicine, describe the four phases of diagnostic studies (p 539). (Sackett, in characteristic style, declares his competing interests as having been "wined, dined, supported, transported, and paid to speak by countless pharmaceutical companies for over 40 years." Many doctors could say the same.)

In a phase I study you ask if patients with a condition have a different result of a test from those who don't. Sackett and Haynes use the example of measuring B type natriuretic peptide in patients with left ventricular dysfunction. Phase I studies don't usually include patients who might or might not have the diseasethe very group in whom diagnosis is important. A phase II study changes the direction of interpretation and asks if patients with an "abnormal" result are more likely to have the condition than those with a normal result. A phase III study arrives in the real world and asks if an abnormal test will make the diagnosis in patients in whom it is clinically sensible to suspect the condition. This is where diagnostic methods often fail, and measurements of B type natriuretic peptide, which looked so promising in phase I and II studies, don't look useful in the phase III study. A phase IV study, the ultimate test, asks whether patients given a diagnostic test have better outcomes than those not given the test. (Patients care less about diagnosis and more about outcomes.) A randomised trial is usually necessary, and such trials are rare.

An editorial discusses the huge difficulty of diagnosing "funny turns" (p 495). A review suggested that over a third of children diagnosed by a Leicester doctor as having epilepsy did not have the condition at all. The editorial examines how such a problem might be avoided in the future.

Footnotes

To receive Editor's choice by email each week subscribe via our website: www.bmj.com/cgi/customalert

© BMJ 2002

Related editorials in BMJ:

The misdiagnosis of epilepsy .

David Chadwick and David Smith
BMJ 2002 324: 495-496. [Full text]  

Other related articles in BMJ:

EDUCATION AND DEBATE
Evidence base of clinical diagnosis: The architecture of diagnostic research.

D L Sackett and R B Haynes
BMJ 2002 324: 539-541. [Full text] [extra: List of DLS's potential conflicts]  

ALL INFORMATION, DATA, AND MATERIAL CONTAINED, PRESENTED, OR PROVIDED HERE IS FOR GENERAL INFORMATION PURPOSES ONLY AND IS NOT TO BE CONSTRUED AS REFLECTING THE KNOWLEDGE OR OPINIONS OF THE PUBLISHER, AND IS NOT TO BE CONSTRUED OR INTENDED AS PROVIDING MEDICAL OR LEGAL ADVICE.  THE DECISION WHETHER OR NOT TO VACCINATE IS AN IMPORTANT AND COMPLEX ISSUE AND SHOULD BE MADE BY YOU, AND YOU ALONE, IN CONSULTATION WITH YOUR HEALTH CARE PROVIDER.