Background: A previous study from Germany showed high antibiotic use in university hospitals, particularly in intensive care units (ICU) and hematology-oncology services, but there has been no information about recent antibiotic use in non-university hospitals. In the present study, we collected data from 40 non-university regional general hospitals located in the southwestern part of the country, and analyzed use density in the medical and surgical services of these hospitals. Materials and Methods: Hospital pharmacy records for the calendar years 2001 and 2002 were evaluated. The number of defined daily doses (DDD, definition according to the WHO/ATC 2001 index) and prescribed daily doses (PDD) per 100 patient days (DDD/100 or PDD/100, respectively) were calculated to compare antibiotic use densities in medical and surgical services. Data for surgery included various subspecialties and gynecology. Results: Antibiotic use in the participating hospitals increased minimally between 2001 and 2002 both in medicine as well as in surgery. Use density in internal medicine (ICU areas excluded) in the year 2002 ranged between 13.5 and 93.7 DDD/100 with a weighted mean of 49.9 DDD/100 (corresponding to 28.6 PDD/100, respectively). Values for surgery were lower with a weighted mean of 43.4 DDD/100 (corresponding to 26.1 PDD/100, range, 10 to 65.4 DDD/100), respectively. Hospital size was not a strong predictor of use density, while large differences were observed between intensive care areas and normal wards. Mean use densities in intensive care areas in 2002 were 105.6 DDD/100 (or 49.7 PDD/100) in medical intensive care units, 116.9 DDD/100 (or 61.2 PDD/100) in surgical intensive care units, and 112.7 DDD/100 (or 66.7 PDD/100) in mixed, interdisciplinary intensive care units. Betalactams made up > 50% of all PDDs, while fluoroquinolones were the second most frequently prescribed drugs (15% of all PDDs). Fluoroquinolones were usually given orally. Overall glycopeptide and aminoglycoside use was < 1 PDD/100. Conclusion: This recent data from a large regional non-university acute care hospital sample confirms that hospital antibiotic use density largely depends on patient care areas and less on hospital size. Surprisingly low use was observed for glycopeptides and aminoglycosides. The data may be useful, as a benchmark for further pharmacoepidemiologic evaluation and focused drug use control interventions.