Evidence Based Laboratory Medicine: principles to outcomes
Professor Christopher P Price,
Director of Clinical Research, Bayer Diagnostics

There is considerable debate over what constitutes an appropriate test, with many claims that too much testing is performed. However systematic reviews on the use of diagnostic tests indicate that only a small proportion of studies meet established quality criteria, whilst an even smaller proportion actually address the important issues of establishing clinical and/or economic benefit. An appropriate test is one in which the result provides an answer to a clinical question. The answer will enable a decision to be made, and an action taken, which leads to an improvement in the care of the patient and/or a more efficient use of healthcare resources.

The culture of evidence based practice underpins these objectives in that it promotes the use of best evidence in making decisions on the care of patients. The establishment of a good portfolio of evidence presents many challenges for the laboratorian; it begins with an explicit recognition of what question the test result is seeking to answer. A further major challenge is to ensure that action is taken upon receipt of the result, that is, an intervention is initiated; invariably the outcome depends on both test result and intervention. There are also important criteria that must be recognised in the experimental design of a study in order to minimise the amount of bias that might be introduced in to the results of the study; for example, an appropriate patient cohort, number of patients etc.

These points will be illustrated with examples taken from the literature and from some current developments in Laboratory Medicine.