Clinicians must integrate a huge variety of clinical data while facing conflicting pressures to decrease diagnostic uncertainty, risks to patients, and costs. Deciding what information to gather, which tests to order, how to interpret and integrate this information to draw diagnostic conclusions, and which treatments to suggest is known as clinical decision making.
When evaluating a patient, clinicians usually must answer the following questions:
In straightforward or common situations, clinicians often make such decisions reflexively; diagnoses are made by recognizing disease patterns, and testing and treatment are initiated based on customary practice. For example, during a flu epidemic, a healthy adult who has had fever, severe myalgia, orbital pain, and harsh cough for 2 days is likely to be recognized as another case of influenza and provided only appropriate symptomatic relief. Such pattern recognition is efficient and easy to use but may be subject to error because other diagnostic and therapeutic possibilities are not seriously or systematically considered. For example, a patient with that flu pattern and decreased oxygen saturation might instead have COVID-19 or have bacterial pneumonia and require antibiotics. Clinicians must be aware of potential biases that can be introduced into the diagnostic process (1).
In more complex cases, a structured, quantitative, analytical methodology may be a better approach to decision making. Even when pattern recognition provides the most likely diagnostic possibility, analytic decision making is often used to confirm the diagnosis and exclude potential disease mimics. Analytic methods may include application of the principles of evidence-based medicine, use of clinical guidelines, and use of various specific quantitative techniques (eg, Bayes theorem).