Clinical judgment predicts culture results in upper respiratory tract infections.

BACKGROUND We wanted to describe the natural history, familial transmission, microbiology, and accuracy of clinical judgment of potential pathogens of respiratory tract infections in a community family practice.

METHODS The study was a prospective case series in which consecutive patients requesting treatment for respiratory tract infections were evaluated after nurse triage during 3 fall-spring months in a solo family practice in suburban Cleveland, Ohio. According to the physician's usual practice, patients were classified into high-, medium-, and low-risk groups for bacterial illness based on their clinical signs and symptoms. Cultures were performed and sensitivities were determined for pathogens from the infected throat, nasopharynx, conjunctiva, or other sites. Patient symptoms and well-being were scored at the initial visit and at 3, 7 and 14 days later.

RESULTS There were 111 illness episodes in 86 patients; 94% had cultures taken, of which 38% grew a potentially pathogenic bacteria, most commonly group A streptococci, Branhamella catarrhalis or Staphylococcus aureus. The physician's judgment of bacterial infection was associated (P < .001) with having a positive culture (sensitivity 53%, specificity 78%, positive and negative predictive values 60% and 73%, respectively). A positive culture was associated with 2 of 16 signs or symptoms: purulent discharge from any site or a red swollen eye. There was no association of treatment status with clinical outcomes during 2 weeks of follow-up observation.

CONCLUSION Infection with a potentially pathogenic bacteria is difficult to determine solely by clinical signs and symptoms, but clinical judgment is associated with positive culture results. The effect of selective treatment of upper respiratory tract infection based on clinical signs and symptoms and patient and family culture results remains to be determined, but using clinical judgment could result in more selective antibiotic use than found in current practice patterns.