A Retrospective Estimate of Ear Disease Detection Using the "Red Flags" in a Clinical Sample.

The sample was a convenience clinical sample chosen to contain many disease diagnoses, so it does not represent the population prevalence of ear disease. Only 251 of 307 recruited patients were analyzed, because information had to be within 90 days of the neurotologist exam. No separate power analysis was performed for this analysis. Red flag symptoms were derived from retrospective case history and examination data, so the symptoms had to be interpreted. Some red flag definitions were ambiguous, and different interpretations changed sensitivity and specificity in preliminary analyses. The neurotologist diagnosis is not a strictly defined reference test. The 104 diseases were not independently evaluated as meaningful contraindications for hearing device acquisition and use. Positive and negative predictive values were illustrative because the true prevalence of these ear diseases is unknown.

The supplied article text does not report a conflict-of-interest declaration. The metadata lists NIH and non-U.S. government research support.

Do not conclude the AAO-HNS red flags are a better screening tool because they caught 98% of disease cases. They screened clear only 20% of people without disease, and the study sample was selected to contain many disease diagnoses, so this does not show how they perform in people seeking hearing aids.