BACKGROUND: A previous study (Davey et al, Value in Health 1999, 2: 333-341) suggested that prior community antibiotic therapy is associated with reduced in-hospital mortality. OBJECTIVES: To confirm the original finding in a larger study and to test the hypothesis that patients who received prior antibiotic treatment would also have lower hospital length of stay. METHODS: Patients with hospital discharge codes including lower respiratory infections (LRTI) or chronic obstructive airways disease (COAD) were identified at three acute hospitals from ICD9 codes in the Scottish Morbidity Records (SMR) for 1993 to 1995. Patients with episodes starting >48h after admission were excluded because they were likely to have nosocomial pneumonia. Prior antibiotic treatment was defined as any treatment in the 28 days before hospitalisation. RESULTS: There were 1285 patients with LRTI alone and 1603 with COAD of whom 366 also had LRTI. Antibiotic treatment occurred before hospitalisation in 35% of patients with LRTI alone and 41% of patients with COAD. Patients with LRTI alone who received community antibiotics were less likely to die in the 50 days after admission: relative risk 0.73 (95% CI 0.61-0.87) unadjusted and 0.83 (0.69-1.00) adjusted for age, sex and social deprivation. There was no difference in 50 day mortality between COAD patients who did or did not receive community antibiotics. Age was the only variable that was significantly associated with length of stay in the LRTI patients. Length of stay was not different for patients who received community antibiotics, either in the entire cohort (p=0.25) or in the 832 patients who were discharged alive (p=0.81). CONCLUSION: This study cofirmed that community antibiotic treatment is associated with reduced 50 day mortality for patients admitted with LRTI, but not COAD. However, there was no significant association between community antibiotic treatment and length of hospital stay.