Evaluation of the physician's ability to recognize the presence or absence of anemia, fever, and jaundice

Abstract

Objective: The evaluation of the patient through a comprehensive history and physical examination is considered the cornerstone of medical diagnosis, but many studies suggest that physicians have inadequate physical examination skills. It is unknown whether these skills are reliable and whether they can be adequately acquired through training. The objective of this study was to evaluate the ability of the clinician to detect the presence and discriminate the extent of clinical anemia, fever, and jaundice in an ED or hospitalized patient.

Methods: This was a prospective observational study of a convenience sample of patients presenting to the ED or admitted to the hospital who had a rectal temperature measurement within 30 minutes prior to the observation, serum hematocrit measurement on the day of observation, or serum bilirubin measurement one day prior to the day of observation. Observers' (emergency medicine attending physicians', resident physicians', and rotating medical students') estimated serum hematocrit, rectal temperature, and serum bilirubin values were obtained after each observation. Sensitivity, specificity, positive predictive value, negative predictive value, and mean absolute difference between actual and estimated values were calculated for each observer.

Results: The physicians detected the presence or absence of anemia, fever, and jaundice in patients with sensitivities and specificities of approximately 70%. Their predictions varied from the measured value on average by 6.0 +/- 4.6% for serum hematocrit, 1.3 + 1.1 degrees F for rectal temperature, and 3.4 +/- 5.3 mg/dL for serum bilirubin. Observer accuracy decreased when evaluating patients with high and low measured values.

Conclusions: The ability to correctly perform and interpret the physical examination appears to be independent of the observer level of training, patient ethnicity, or patient gender. The examination for pallor, warmth, and jaundice is unreliable in predicting the corresponding laboratory or electronic measurement. Certain anemic, febrile, or jaundiced patients may not be reliably detected solely by a focused physical examination.