RESUMO
Innovative approaches are needed to improve cost-effective antibiotic use in hospitals. We established an algorithm to provide a recommendation for both a single oral and a single parenteral antimicrobic, including dosage, to be placed in the bacteriology laboratory report for therapy guidance of 174 episodes of lower respiratory infection in a group of "program" patients with no evidence of infection at other sites. The initial recommendation was based on the Gram-stained direct smear of secretions. The recommendation was updated when preliminary culture results were available, and again with final identification and susceptibility data. We compared the severity of illness, length of stay, and cost of therapy for 68 episodes in program patients who received the recommended therapy with 111 episodes in a control group of patients who received antimicrobial therapy but for whom no recommendations were reported. There was a significantly greater use of the recommendations for sicker patients. Antimicrobial therapy, in general, was more often used for the program patients than for the controls. There was no difference in length of stay or therapy cost. We recommend that this approach be used in settings where there is more problematic use of antimicrobics.