We retrospectively collected data from one community managed care organization on all ambulatory care patients initially diagnosed with pneumonia or acute bronchitis from October, 1, 1992, to March 31, 1993, and from November 1, 1993, to January 31, 1994. We considered treatment to be successful when patients did not return for any related service within 15 days of initial diagnosis. We identified 2,490 episodes of illness, 85.7% which were acute bronchitis and 14.3% which were pneumonia. Overwhelmingly, physicians approached these conditions empirically (no diagnostic test); just 8.6% of patients had a diagnostic test during the 15-day episode of illness. Two-hundred twenty-nine of the episodes (9.2%) were apparently related to initial diagnoses, as they occurred during the 15-day period. More branded prescriptions (vs. generic) were dispensed during these related episodes. One patient was hospitalized and 19 patients used the emergency room either for first or subsequent visits. Empiric treatment is associated with effective diagnosis and therapy in ambulatory care patients with acute bronchitis and pneumonia. It remains unclear, however, if this strategy is the most cost-effective or if it leads to the most effective utilization of services.