Academic Medicine:
doi: 10.1097/ACM.0000000000000385
Letters to the Editor

Back to Bayesics

Sinha, Pranay; Yoo, Alexander Dong Hyung

Free Access
Article Outline
Collapse Box

Author Information

Fourth-year student, University of Virginia School of Medicine, Charlottesville, Virginia; pranay2000@gmail.com.

Fourth-year student, University of Virginia School of Medicine, Charlottesville, Virginia.

Disclosures: None reported.

Back to Top | Article Outline

To the Editor:

The decline in physical examination skills among medical students and resident physicians is concerning.1 The devolution in these skills goes hand in hand with widespread doubts about the utility of the physical exam in modern medicine. Reflecting on the underpinnings of diagnosis and clinical reasoning offers insight to reestablish the exam’s importance in modern medicine.

Bayesian analysis is the bedrock of medical diagnosis and reasoning.2 The raw material of Bayesian analysis are data, which sway the individual towards or away from a conclusion. A Bayesian approach requires one both to efficiently collect data and to know how to weigh the collected clues. Consequently, we have a twofold suggestion based on the principles of Bayesian analysis for resurrecting physical observation.

First, teach trainees to collect data efficiently and thoroughly. Bayesian analysis relies upon specific and accurate information to argue for or against an eventuality. Unfamiliarity makes for inefficient and unhelpful examination. Goal-oriented physical diagnosis and observation skills must be emphasized early and continually in medical education. Otherwise, students run the risk of missing management-changing information in practice, or acquiring it expensively through advanced imaging and lab tests.

Second, teach students to weigh their observations. Vague qualitative comments about the efficacy of physical signs (“Egophony suggests consolidation”) fail to provide information needed for interpretation in complex scenarios. Therefore, educators should give reasoned, quantitative, and evidence-based information about observations (“Egophony has a positive LR of 4.1 for pneumonia in patients with fever, cough, sputum production, or dyspnea”).3

A caveat needs to be offered to all students: Clinical findings must not be interpreted out of context. The context (e.g., history and epidemiology) helps formulate a pretest probability of disease. Indeed, a patient with a low pretest probability of disease who demonstrates physical findings with high sensitivities and specificities for the disease is still unlikely to harbor the illness (low posttest probability). For instance, one should not rush to report smallpox when confronted with a diffuse vesicular rash.

A combination of detailed observation and rigorous evidence-based reasoning creates a brand of physical diagnosis that is needed in 21st-century medicine.

Pranay Sinha

Fourth-year student, University of Virginia School of Medicine, Charlottesville, Virginia; pranay2000@gmail.com.

Alexander Dong Hyung Yoo

Fourth-year student, University of Virginia School of Medicine, Charlottesville, Virginia.

Back to Top | Article Outline

References

1. Corbett EC, Elnicki DM, Conaway MR. When should students learn essential physical examination skills? Views of internal medicine directors in North America. Acad Med. 2008;83:96–99

2. Gill CJ, Sabin L, Schmid CH. Why clinicians are natural Bayesians. BMJ. 2005;330:1080–1083

3. McGee S Evidence Based Physical Diagnosis. 20123rd ed Philadelphia, PA: Elsevier/Saunders

© 2014 by the Association of American Medical Colleges

Login

Article Tools

Share