Antibiotic Prescribing for Respiratory Tract Infections and Encounter Length

TO THE EDITOR: The use of telemedicine is promising and will probably increase access to health care, particularly in rural areas. Martinez and colleagues (1) found that the average duration of telemedicine encounters was shorter when antibiotics were prescribed. However, we do not believe that the results of their study, which analyzed patient visits through the direct-to-consumer telemedicine platform, can be extrapolated to the reality of primary care. The results were adjusted for the data registered during the visits, but some important characteristicssuch as how serious clinicians judged the infection to be or whether the patient demanded antibioticswere not reported. The authors showed that nearly 50% of the patients using the telemedicine platform were diagnosed with sinusitis and that antibacterials were prescribed in more than 90% of these cases. The epidemiology of the respiratory tract infections seen in the telemedicine system clearly differs from that of most of these infections treated in office visits. It also differs from that of the general population of the United States, because sinusitis is present in fewer than 15% of the respiratory tract infections seen in the community (24). That clinicians probably mislabeled the diagnosis to better justify antibiotic prescribing for infections that are mostly self-limiting may partially explain the discrepancy between sinusitis being diagnosed in 50% of telemedicine encounters but fewer than 15% of in-person encounters. In fact, differentiating bacterial sinusitis from a mild episode of acute rhinosinusitis without examining a patient is difficult. Ashwood and associates (5) recently showed that the convenience of direct-to-consumer telehealth may drive many patients who would not have sought care for an illness had telemedicine been unavailable (5). As such, patients who decide to talk to a clinician by telephone or live video are not replacing those who would attend an in-person office visitinstead, most are new users. This point is crucial: The patients analyzed by Martinez and colleagues would seldom have made an appointment to see their physicians, probably because their symptoms were milder. Extrapolating their results to the reality of primary care thus may be misleading.