Antibiotic use is the driving force for increasing antibiotic resistance. A large proportion of antibiotics in hospitals are used inadequately. The objective of this study was to evaluate antibiotic use at the Hospital for Infectious Diseases through point-prevalence surveys conducted in 2006, 2008, and 2009. Point prevalence surveys were part of the European Surveillance on Antimicrobial Consumption (ESAC) Hospital Care Subproject and patients' data were collected following ESAC protocol. Additionally, the adequacy of antimicrobial therapy and administration of the first line antibiotic according to the local guidelines were assessed by an infectious disease doctor and a clinical microbiologist. In the study period among the 599 patients admitted to hospital, 352 (58·8%) received antibiotics. Out of 448 antimicrobial treatments, 313 (69·9%) were administered parenterally and 135 (30·1%) orally. Altogether in years 2006, 2008, and 2009 the most commonly prescribed antibiotics were ceftriaxone (19·9%), co-amoxiclav (15·4%), ciprofloxacin (12·3%), narrow spectrum penicillins (6·5%), and penicillinase resistant penicillins (5·6%). Most (82·6%) of the treated infections were community acquired infections. The predominating diagnoses were urinary tract infections and infections with no primary site defined, followed by skin, soft tissue and bone and joint infections. The overall adequacy of antimicrobial therapy was 82% and the first line antibiotic according to the local guidelines was administered with high frequency for central nervous system and cardiovascular infections (100%), and low for ear, nose, and throat infections, urinary tract infections, lower respiratory tract and bone and joint infections (23·0%, 51·6%, 52·5%, 65·0%, respectively) which indicates a significant overuse of antibiotics for diagnoses listed. The results of an individual point prevalence survey provided reliable and representative data for the hospital. Point-prevalence surveys proved to be a valuable method for detecting targets for antibiotic prescribing improvement and they clearly showed that our local hospital guidelines offered too many choices of antibiotic treatment for each clinical indication and needed revision.