Skip to main content
Diagnosing Dizziness in the Emergency Department: Physicians May Rely Too Heavily on Symptom Quality. Results of a Multicenter, Quantitative Survey
Annals of Emergency Medicine (2007)
  • Paris B Lovett

Study Objectives: The textbook approach to diagnosing dizziness relies heavily on initially classifying the patient’s qualitative complaint as vertigo, presyncope, disequilibrium, or ill-defined dizziness, with each “type” indicating a narrow spectrum of possible causes. Specifically, vertigo is said to imply a vestibular problem, presyncope a cardiovascular problem, disequilibrium a neurologic problem, and illdefined (nonspecific or vague) dizziness a psychiatric or metabolic one. Although this “quality-of-symptoms” approach is cited frequently in medical textbooks and peerreviewed medical literature, it is unknown whether emergency physicians generally use this approach clinically, nor to what extent it might influence their diagnostic reasoning or management. The relevance of this question is heightened by recent evidence suggesting that emergency department dizzy patient reports of dizziness type are unreliable.

Methods: Our goal was to quantify emergency physicians’ self-described practice in the diagnostic assessment of dizziness. We hypothesized that most would endorse the “quality-of-symptoms” approach, and that doing so might be associated with “risky” diagnostic reasoning or management strategies. The study design was an anonymous, internet-based survey conducted at 17 academic-affiliated emergency departments. Eligible participants included attending and resident emergency physicians. We assessed the ranked relative importance of symptom quality, timing, triggers, and associated symptoms to diagnosis of the dizzy patient. We also assessed level of agreement with each of 20 statements about the diagnostic assessment of dizziness in clinical practice using a 7-point Likert scale from strongly agree to strongly disagree. Logistic regression was used to assess the relationship between “quality ranked first” and Likert responses to clinical practice questions.

Results: There was an 82% (n 415/505) response rate. 93% (95% CI 90-95%) agreed that determining dizziness type is very important, and 64% (95% CI 60-69%) ranked “quality” the most important diagnostic feature in the assessment of a dizzy patient. The three other dizziness attributes were each raked “most important” by five-fold fewer physicians (associated symptoms 13%, triggers 12%, timing 11%). In a multivariate model, those ranking symptom quality most important more often reported risky clinical reasoning that might predispose to misdiagnosis (e.g., in a patient with persistent, continuous dizziness -- who could have a cerebellar stroke– these physicians reported feeling reassured that a normal head CT indicates the patient is safe to go home: OR 2.43, 95% CI 1.23-4.77).

Conclusion: Our study demonstrates that physicians report taking a quality-of-symptoms approach to diagnosis of dizzy patients in the emergency department. Those who rely heavily on this approach may be predisposed to high-risk downstream diagnostic reasoning. Other clinical features (timing, triggers, and associated symptoms) appear relatively undervalued. Educational initiatives merit consideration.

Lovett P, Newman-Toker DE, Stanton V, Hsieh Y, Camargo Jr. CA, Edlow JA, Goldstein JN, Abbuhl S, Lin M, Chanmugam A, Rothman RE/The Johns Hopkins University, Baltimore, MD; University of California, San Francisco, San Francisco, CA; Massachusetts General Hospital, Harvard Medical School, Boston, MA; Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA; New York Presbyterian Hospital-Columbia University Medical Center, New York, NY; Massachusetts General Hospital, Boston, MA; University of Pennsylvania School of Medicine, Philadelphia, PA; San Francisco General Hospital, San Francisco, CA

  • dizziness,
  • emergency,
  • stroke,
  • migraine
Publication Date
September, 2007
Citation Information
Paris B Lovett. "Diagnosing Dizziness in the Emergency Department: Physicians May Rely Too Heavily on Symptom Quality. Results of a Multicenter, Quantitative Survey" Annals of Emergency Medicine Vol. 50 Iss. 3 (2007)
Available at: