OBJECTIVES: Antibiotic overuse and misuse are major factors contributing to the development of resistance. Therefore, information on the trends and patterns of antibiotic consumption is essential to use suitable strategies that aim to reduce antibiotic use. METHODS: A one year survey (January-December 2004) was conducted at 30 well stocked retail pharmacies under five municipal wards of Delhi, India. Data collected by recording the monthly (bulk) purchase data of all pharmacies and by doing 15-25 ‘exit-interviews’ per pharmacy per month of patients purchasing antibiotic. These two methodologies were used to validate the results because there is no database available for antibiotic consumption for outpatients in India and in many other developing countries. Results were analyzed by calculating the number of defined daily doses (DDD) per 1000 population (for bulk purchase data) and DDD/1000 patients for ‘exit interview’ data, using WHO anatomic therapeutic chemical classification and DDD measurement methodology. RESULTS: Bulk purchase data and exit interview data (month wise and total consumption for one year) showed that consumption (DDDs) of fluoroquinolones (32805/1000 patients, 1347/1000 population) was highest followed by penicillins (14463/1000 patients, 415/1000 population), cephalosporins (12888/1000 patients, 380/1000 population) and macrolides (11085/1000 patients, 336/1000 population).  A detailed analysis for the pattern of consumption of different members from each class of antibiotic was done. Highest consumption of different members from these four classes of antibiotics was -  Fluoroquinolones (J01MA) - levofloxacin, ciprofloxacin, ofloxacin and gatifloxacin; Penicillins (J01C) – amoxicillin+clavulanic acid, amoxicillin, amoxicillin+cloxacillin; Cephalosporins (J01DA) – cefixime, cefuroxime, cephalexin, cefadroxil; and Macrolides (J01FA) – roxithromycin, azithromycin. We noted a growing peak in the consumption of antibiotics in the month of February and March and a little increase in the consumption during August. CONCLUSIONS: These data provide a useful methodology for surveillance of antibiotic use and for assessing public-health strategies that can be aimed to reduce antibiotic use and containment of antibiotic resistance.