Of Hoofbeats, Horses, and Zebras: A "Conversation" with Drs. William Dock, Theodore Woodward, and Alvan Feinstein

It all began, so they say, with Theodore Woodward, "Don't look for zebras on Greene Street." This aphorism has since been passed on to generations of residents and medical students at the University of Maryland School of Medicine and beyond. A zebra is a "medical condition whose appearance at a particular time and place, in a particular person, is both unexpected and astonishing. The 'condition' may be an item from the history, a physical finding, a laboratory test, or a diagnosis." This maxim nudges clinicians to consider competing or paradoxical hypotheses when assessing patients for common, unexpected, and rare diseases. Whereas errors are unavoidable, considering zebras (uncommon diagnoses) helps avoid "not-to-miss" diagnoses, patient harm, and the associated regret of failure. Perhaps there's also the tug to "bag the big one." On the one hand, pursuing zebras may result in overestimating probability assessments with the potential for harming patients and wasting resources, while applying the science of formal medical decision-analysis lessens the chances of overzealous workups. Unfortunately, conducting studies to evaluate the utility of clinical findings for rare diseases is difficult. The key is to identify symptoms and signs, preferably in combination, with high specificity and sensitivity. On the other hand, assuming horses (common diagnoses) for too long also has its risks. Both the failure to recognize when the current therapeutic plan is ineffective and to appreciate the diagnostic value of the history, physical, and routine laboratory tests can delay the search for alternative hypotheses and ultimately efficacious treatment. Sir William Osler admonished us over a century ago, "Use the knife and cautery to cure the moral necrosis which you will feel in the posterior parietal region, in Gall and Spurzheim's centre of self-esteem, where you will find a sore spot after you have made a mistake in diagnosis."