Imaging and Special Studies | ||

<<Transfusions |
Decision Making>> |

The value of diagnostic testing is overestimated. For many conditions the likelihood of a disease in the population selected at random is small, on the order of one in a thousand. Arguably that likelihood is increased if the sample is instead the population of individuals who have self selected or been referred to an orthopedic office. All diagnostic tests have both false positive and false negative results which by traditional statistics or Bayesian inference can be calculated to define the likelihood of the patient having the disease if a given diagnostic test is positive. The strength of evidence provided by a diagnostic test is going to be influenced by the baseline likelihood of disease in the sampled population. The more specialized the provider in dealing with a specific set of problems the higher the likelihood that a larger proportion of that population has definable pathologies of a certain category. A diagnostic test may be exceedingly unhelpful, not to mention cost ineffective when used as a population screen and yet be reasonably useful in the hands of a subspecialist who may order the test. The likelihood that a positive result is a true positive is increased in the latter case. While it may be true that for an especially rare condition the likelihood in the face of a positive test result may still be less than 1 in 2, the fact of a positive diagnostic test may have increased the likelihood that the disease is present by a factor of 10 relative to the general population of patients in that office. Thus for the specialist and the patient the chances for a successful outcome by virtue of the accurate identification of the pathology in question has been improved. It should be emphasized however that the uncertainty in by no means eliminated. Bayesian Inference