Outpatient Antibiotic Prescribing Patterns: Is it Time for a Change?


Synopsis: In a survey of physician settings, colds, upper respiratory tract infections, and bronchitis were the most common source of antibiotic prescriptions in the ambulatory care setting. As these infections are usually self-limited and usually of viral origin, this may represent a common and potentially inappropriate use of antimicrobial therapy.

Source: Gonzales R, et al. JAMA 1997;278:901-904.

The threat of antimicrobial resistance is increasing. Since the 1980s, physicians have witnessed the progressive resistance of both H. influenzae and pneumococcus to commonly prescribed antibiotics. This trend has profound implications for public health, with more than 10% of pneumococcus now highly resistant to penicillin. While the source of this problem may be multifactorial, significant evidence now points to the excessive inappropriate use of antibiotics as the major culprit. The study by Gonzales et al provides insight into how antibiotics are really used in the outpatient setting, and the results are not comforting.

Using the National Ambulatory Care Survey, the authors examined the records of 28,787 office visits of 1529 physicians during 1992. The antibiotic prescription rates for acute nasopharyngitis (common cold), acute upper respiratory tract infections, and bronchitis were evaluated. The authors found that bronchitis and upper respiratory tract infection were the second most common source of antibiotic prescription in the outpatient setting overall, exceeded only by sinusitis. These two diagnoses were responsible for 21% of all outpatient prescriptions. Prescription rates for colds, upper respiratory tract infections, and bronchitis ranged from 51% to 66%. Women (OR, 1.65) and those treated in a rural setting (OR, 2.25) were more likely to receive antibiotics, while blacks tended to receive fewer prescriptions. Interestingly, age, ethnicity, physician specialty, or payment type did not affect the likelihood of receiving an antibiotic for a minor upper respiratory tract illness. In multivariate analysis, only rural practice location was independently associated with prescribing patterns.


The prevalence of antimicrobial resistance in both community and nosocomial bacteria is increasing dramatically and has been especially noteworthy for pneumococcus, H. influenzae and Salmonella species. Both the use of antimicrobial agents in animal husbandry and excessive and inappropriate prescribing for inpatients for prevention or colonization have been targeted by the physician community. The data provided by Gonzales et al further add to the conviction that outpatient prescribing patterns must also be modified. This large and broadbased survey reported that colds, upper respiratory tract infections, and bronchitis are the most common sources of misapplication of antimicrobial chemotherapy in the ambulatory care setting. Interestingly, specialty affiliation did not influence this widespread practice, while women and those treated in a rural setting were most likely to be inappropriately treated. The authors did not "drill down" into the database, and we can only assume that most of these infections were viral in etiology and self-limited. This may not have been the case in all situations, as patients with acute exacerbation of COPD often benefit from antibiotic treatment if they present with multiple symptoms.

Why are physicians overprescribing antibiotics? There are no clear answers, but preliminary evidence suggests that doctors are responding to "what is expected" by their patients. We know that even when physicians accept the premise of clinical practice guidelines, they often do not change their practice patterns. My suspicion is that most physicians overprescribe antibiotics and know this is incorrect. To deal with this problem, creative and widespread patient education will have to be provided to change the ingrained public belief in the efficacy of antibiotics for any infection.


1. Doern. Amtimicrob Agents Chemother 1996;40:1208-1213.

2. Anthonisen, et al. Ann Intern Med 1987;106:196-204.