Self and parental report of physician-identified acute otitis media (AOM) in a rural sample

Acute otitis media (AOM) is a leading cause of family medicine consultations. Rates of AOM are traditionally determined by review of medical charts, which can be costly and time consuming. This information can also be obtained directly from patients (or parents) by self-administered surveys or personal interview. To ensure the quality of self-reported AOM as a proxy for physician-recorded diagnosis, we assessed its accuracy compared to medical report documentation. Self (and maternal) reports of AOM at outpatient consultations at family practice clinics and hospital emergency departments were collected prospectively by interview from participants in the Hutterite Influenza Prevention Study. Similar data were also collected by fax requests for medical record information to the medical facilities. We calculated AOM reporting by each data source. Sensitivity, specificity, predictive values and likelihood ratios of self-reported AOM using medical record documentation as the gold standard were determined. Compared to the medical records, the proportion of AOM cases was underestimated by participants (22% versus 16%), but this difference was not significant (p = 07). Self-report of AOM was a very specific measure (93%), but had lower sensitivity (47%) than medical records. Positive predictive value was moderate at 64% but negative predictive value was good at 86%. The positive likelihood ratio was 6.7, while the negative likelihood ratio was 0.57. Self-report of AOM in our sample had high specificity and good negative predictive value. However, reliance on self-report without verification by medical record may result in a number of false negatives, which may affect enrolment eligibility or outcome analyses in medical research.