Derivation of Patient Decision Tool for Sore Throat

Document Type

Conference Proceeding

Publication Date


Publication Title

Acad Emerg Med


Background: Most diagnostic support tools are designed for clinicians. Although web or mobile based symptom-based tools are widely available, few if any have been validated. Particularly in under-served communities, such tools could reduce healthcare cost if they are able to guide patients to avoid ED or urgent care when unneccessary.

Objective: To derive a patient decision tool for sore throat and assess patients' capacity for using such a tool.

Methods: A prospective observation study inclusive of patients presenting to an urban ED and urgent care center. We included ages 2 to 50 years with complaints of sore throat and upper respiratory complaints. We excluded patients outside this age range or if they or their guardian were unable to provide informed consent. We collected extensive symptom and clinical information that would be available to any patient at home. The primary outcome was streptococcal pharyngitis confirmed by throat culture or a related complication such as peritonsillar abscess. We also queried patients on their capacity and willingness to use such information at home through a web-based score tool to guide their care. Analysis consisted of multivariable logistic regression modeling to derive a score tool and to determine its diagnostic test characteristics for predicting the diagnosis.

Results: We enrolled 176 patients, 64% female, 88% African American, and 31% < 18 years. There were 46 (26%) patients that had strep pharyngitis. The most common alternative diagnosis was a viral upper respiratory infection (67%). A derived score tool had an area under the curve of 0.76 (95%CI 0.68-0.84). Patients scoring < 5 (n=76, 43%) on a 0 - 20 scale had a NPV of 89% (95% CI 80-95%) for streptococcal pharyngitis. 52% of patients stated they had no access to a clinic to seek medical advice for their current medical problem. Nearly all participants (98%) stated that they have phone internet access, 76% routinely use mobile apps, and 95% affirmed that they would use a mobile tool designed to help them choose the appropriate level of medical care.

Conclusions: This study indicates that a patient administered score tool for sore throat may perform similar to existing physician administered score tools. Patient willingness to use such a tool to guide their choices for medical care is high.



First Page


Last Page


This document is currently not available here.