Impact of initial antibiotic choice on clinical outcomes in community-acquired pneumonia: analysis of a hospital claims-made database.

BACKGROUND Much controversy exists regarding the initial choice of antibiotics and selected outcomes for patients with community-acquired pneumonia (CAP). METHODS The investigators analyzed a hospital claims-made database to assess the impact of initial antibiotic choice on 30-day mortality, total hospital costs, and hospital length of stay (LOS). Fine risk groups allowed for stratification for variations in the severity of illness. Patients were divided into five monotherapy groups (ie, ceftriaxone, "other" cephalosporins, fluoroquinolones, macrolides, or penicillins) and four groups that received dual therapy (ie, the agents listed above, except macrolides) plus macrolides. Patients also were stratified by age (ie, > 65 years of age and < 65 years of age). Severely ill patients were excluded. RESULTS Overall, 44,814 persons met the criteria for inclusion. Among monotherapy patients, those who received macrolides had the least mortality but were the least ill. Patients who received dual therapy generally had shorter LOSs, lower total hospital charges, and decreased mortality compared with those who received monotherapy. Differences among dual-therapy regimens regarding outcomes studies were noted. Patients who were < 65 years of age had lower mortality rates, shorter LOSs, and lower hospital charges than did the more elderly patients. Within this group, those who received dual therapy had better outcomes than those who received monotherapy. CONCLUSIONS We confirmed the value of dual therapy employing macrolides as a second agent in decreasing mortality from CAP, and we provided similar data regarding shorter LOSs and lower hospital charges. This appears to hold for a younger population. Differences among dual-therapy regimens (all employing macrolides) appear to exist and may be clinically relevant.