In a 750-bed community-teaching hospital with 3,200 employees, throat cultures were routinely done in hospital personnel complaining of a sore throat. During a 3-month period, 323 employees had throat cultures; only 20 (6.2%) of these throat cultures grew group A streptococcus. The prevalence of positive throat cultures was similarly low in employees (6.2%) and adult patients (7.3%). There was no evidence that employees either had higher prevalence of group A streptococcal pharyngitis or that they spread the infection to patients. It is concluded that routine throat cultures are not warranted in employees complaining of a sore throat, and that throat cultures should be done only selectively in hospital personnel with a high probability of group A streptococcal pharyngitis.