Monitoring antibiotic consumption is a valuable tool which has been increasingly used in the last years due to the current concern with the emergence of resistant microbial strains. The present study aimed at monitoring antibiotic consumption, evaluating the economic impact of hospital antibiotic prescription and assessing the relationship between the prescribed antibiotics and the indications for either prophylactic or therapeutic use. This was a longitudinal pilot-study for which data were collected in six privately managed public hospital units during the month of May 2004, with a resulting sample of 1,122 admitted patients. We observed a prescription incidence rate of 76.9%, corresponding to a total of 1,154 dispensed antimicrobials, with a mean 71.2% of these antimicrobials being dispensed for the prophylaxis of surgical site infection (SSI). The mean cost of antibiotic courses was higher in cases of "suspected infection" (9.09 euro) or "confirmed infection" (8.74 euro) and lower in cases of "prophylaxis" (5.67 euro), a finding which is explained by the shorter mean duration of the later. There was a considerable variation among the different hospital units regarding the type of antibiotic compound that was used for SSI prophylaxis, with a mean duration of antibiotic use of 2.61 days for this indication and about half of the prophylactic regimens lasting longer than 24 hours, a fact that suggests an insufficient observation of the current recommendations for antibiotic use in SSI prophylaxis. This finding indicates the need for an investigation on the actual existence of local recommendations for SSI prophylaxis in individual hospital units and also for the evaluation of the compliance of practicing surgeons with eventually existing recommendations.