Assessment and Diagnosis: Building and Testing a Differential
Build the differential before you order
Assessment items are scored as a reasoning process, not a recall list. Start with a problem-focused history and physical that produces a ranked differential, then choose the test whose result will actually change management. Know test characteristics cold: a highly sensitive test is best for ruling out (a negative result is reassuring), a highly specific test is best for ruling in. Likelihood ratios move you between pretest and posttest probability, which is why a D-dimer is useful in a low- or intermediate-probability patient and near-useless in a patient who is already likely to have pulmonary embolism.
Numbers with context
Expect arterial blood gas interpretation (compensation, anion gap, delta gap), lactate and lactate clearance, renal function (creatinine is blind to muscle mass, so calculate eGFR), liver panel patterns, and drug levels. Always interpret against the person: a normal white count does not exclude sepsis in an older or immunocompromised patient, and a normal heart rate does not exclude a surgical abdomen in a patient on a beta blocker.
Documentation and cognitive bias
The medical decision-making note should show data reviewed, differential considered, and risk. Items test anchoring on an initial label, availability after a recent similar case, and premature closure. Ask two questions every time: what else could this be, and what is the worst thing this could be?
Worked rule of thumb
A 48-year-old with pleuritic chest pain, heart rate 104, recent immobilization, no hemoptysis: Wells criteria place her in the likely group, so she goes to CT pulmonary angiography rather than D-dimer. Reverse the picture - no risk factors, low Wells score - and a negative age-adjusted D-dimer excludes pulmonary embolism without radiation.