Skip to Main Content

Clinical Scenario

CLINICAL SCENARIOS

Consider the following diagnostic situations:

  1. A 43-year-old woman presents with a painful cluster of vesicles grouped in the T3 dermatome of her left thorax, which you recognize as shingles from reactivation of herpes zoster.

  2. A 78-year-old man returns to your office for follow-up of hypertension. He has lost 10 kg since his last visit 6 months ago. He describes reduced appetite but otherwise has no localizing symptoms. You recall that his wife died a year ago and consider depression as a likely explanation, yet his age and exposure history (ie, smoking) suggest other possibilities.

Two Complementary Approaches to Diagnosis

The first case in the opening scenarios illustrates the rapid, nonanalytic approach that expert diagnosticians use to recognize disorders they have seen many times before (ie, pattern recognition) and that is particularly relevant to the diagnostic properties of aspects of the physical examination.1-6 The second case illustrates a more challenging circumstance in which simple pattern recognition fails, so expert diagnosticians slow down and toggle to a more analytic mode of diagnostic thinking.7,8 This includes the probabilistic approach to clinical diagnosis that uses evidence from clinical research—the focus of this chapter (Figure 16-1). Using this probabilistic analytic approach, expert diagnosticians generate a list of potential diagnoses, estimate the probability associated with each, and conduct investigations, the results of which increase or decrease the probabilities, until they believe they have found the best answer to fit the patient's illness.9-14

FIGURE 16-1

Pattern Recognition vs Probabilistic Diagnostic Reasoning

Applying the probabilistic approach requires knowledge of human anatomy, pathophysiology, and the taxonomy of disease.11,12,14 Evidence from clinical research represents another form of knowledge required for optimal diagnostic reasoning.15-17 This chapter describes how evidence from clinical research can facilitate the probabilistic mode of diagnosis.

Clusters of Findings Define Clinical Problems

Using the probabilistic mode, clinicians begin with the medical interview and physical examination, which they use to identify individual findings as potential clues. For instance, in the second scenario, the clinician noted a 10-kg weight loss in 6 months that is associated with anorexia but without localizing symptoms. Experienced clinicians often group findings into meaningful clusters, summarized in brief phrases about the symptom, body location, or organ system involved, such as “involuntary weight loss with anorexia.” These clusters, often termed “clinical problems,” represent the starting point for the probabilistic approach to differential diagnosis (see Chapter 17, Differential Diagnosis).11

Clinicians Select a Small List of Diagnostic Possibilities

When considering a patient's differential diagnosis, clinicians must decide which disorders to pursue. If they considered all known causes to be equally likely and tested for them all simultaneously (the “possibilistic” list), unnecessary ...

Pop-up div Successfully Displayed

This div only appears when the trigger link is hovered over. Otherwise it is hidden from view.