The rise in antibiotic-resistant bacteria has focused attention on the selective pressure caused by widespread use of antibiotics for conditions in which antibiotics are not effective. In particular, patients seem confused about the normal presentation and course of an uncomplicated upper respiratory tract injection (URTI). Although discoloured nasal discharge is a normal self-limited phase of a viral URTI, and randomised placebo-controlled trials have shown no effect of antibiotics on purulent rhinitis or discoloured nasal discharge, patients are confused about the meaning of discoloured nasal discharge. Nearly 80% of adult patients believe antibiotics are effective for an URTI with discoloured nasal discharge. In this confusion about the normal presentation of respiratory infections being transmitted by the primary care physician to the patient? The purpose of our study was to examine primary care physicians’ understanding of the normal presentation and appropriate treatment regimens of URTIs, and acute bronchitis. Questionnaires were sent to 250 primary care physicians (general practitioners, family physicians, and general internists) in each of four states: Minnesota, South Carolina, Texas, and Oregon. The physicians were randomly selected from the American Medical Association’s Physician Masterfile. A total of 211 were returned and not able to be used because the physician either moved, died, or retired from practice. 214 completed questionnaires were returned and used for analysis. We assessed knowledge and current practices of treatment of URTIs with two scenarios representing the normal presentation of an uncomplicated URTI, varying only in the inclusion of discoloured nasal discharge. Similarly, knowledge and practice for acute bronchitis was based on two scenarios representing the normal presentation of acute bronchitis, varying only in the inclusion of purulent sputum. When the scenario presented an adult with symptoms of an URTI, physicians were more likely to diagnose sinusitis than an URTI when the nasal discharge was described as discoloured (table). Antibodies were prescribed seven times more frequently when the nasal discharge was discoloured. Even respondents who labelled both scenarios as an URTI were 5·6 times more likely to use an antibiotic when the nasal discharge was discoloured than when it was clear (28% vs 5%, p<0·001). In the second scenario which presented an adult with a cough following URTI symptoms, when the sputum was described as discoloured physicians were more likely to label the illness as bronchitis and prescribe antibiotics (table). Interestingly, physicians were more likely to use a bronchodilator for cough with clear sputum. In both the clear and purulent sputum scenarios, bronchodilators have been shown to have some asymptomatic benefit for patients with acute bronchitis regardless of sputum colour while antibiotics are of questionable effectiveness. This small study shows that physicians assign a great deal of predictive value to the colour of nasal discharge or sputum despite evidence that the appearance of these secretions alone is not predictive of bacterial infection or the effectiveness of antimicrobial therapy. Patients appear to have learned this interpretation of discoloured secretions from physicians and may bring expectations of receiving antibiotics for illnesses into the physician-patient encounter. These misperceptions may contribute to the inappropriate prescribing of antibiotics and may be a factor in the development of antibiotic-resistant strains of common pathogens.