A cross-sectional study to determine the prevalence, distribution and antimicrobial profile of Salmonella serotypes isolated from food items and apparently healthy supermarket butchery workers was undertaken from September 2003 to February 2004. A total of 1200 food items consisting of chicken carcass (208), pork (194), mutton (212), minced beef (142), cottage cheese (190), Tilapia fish meat (128) and ice cream (126) were purchased in Addis Ababa. Additionally sixty-eight stool samples were analyzed. Chicken carcass, pork, mutton and minced beef samples were collected from 32 randomly selected supermarkets while cottage cheese, fish and ice cream samples were collected from open markets (3), fish shops (6) and pastry shops (17) in Addis Ababa, respectively. Out of the food items, 7.8% were positive for Salmonella and of sixty-eight stool samples five gave positive result (7.4%). About 14% of chicken carcass, 11.3% of pork, 10.8% of mutton, 8.5% of minced beef, 2.1% of cottage cheese, 2.3% of fish and none of the ice cream yielded Salmonella. Salmonella was recovered from samples taken from 21 of the 32 supermarkets considered in the study. On the other hand one open market out of three, two fish shops out of six and none of the 17 pastry shops gave Salmonella positive results. A total of 14 different serotypes out of 98 Salmonella isolates were identified. Salmonella Newport (41.8%) was the most prevalent serotype, followed by S. Braenderup (12.2%), S. Hadar (8.2%), S. Typhimurium (7.1%), S. Dublin (6.1%) and S. Haifa (6.1%). Less commonly isolated Salmonella serotypes included: S. Infantis, S. Kentucky, S. Bovismorbificans, S. Anatum, S. Zanzibar, S. Kottbus, S. Saintpaul and S. 1: 9, 12:-. Salmonella Newport and S. Kentucky were reported for the first time in Ethiopia. Salmonella Newport was isolated from all sample types except ice cream, while S. Braenderup, S. Kottbus, S. Saintpaul were detected only from chicken carcass, pork and minced beef samples, respectively. There was a statistically significant difference in the rate of Salmonella isolation between meats (chicken carcass, pork, mutton and minced beef) and the rest of the samples (cottage cheese, fish and ice cream) (Pearson’s X2 = 37.569, df = 1, p-value = 0.000). The level of antimicrobial x i resistance was significantly higher for chicken carcass and pork isolates as compared to other samples (p = 0.003). Assay of antimicrobial resistance revealed that 32.7% of Salmonella isolates were resistant to one or more of the 24 antimicrobials tested. Generally resistance for 13 different antimicrobial drugs was recognized. The most common resistance was to streptomycin (24/32, 75%), ampicillin (19/32, 59.4%), tetracycline (15/32, 46.9%), spectinomycin (13/32, 40.6%) and sulfisoxazole (13/32, 40.6%). All the three Salmonella Kentucky isolates showed resistance to at least 8 antimicrobials, which includes: ampicillin, amoxicillin/clavulanic acid, ciprofloxacin, nalidic acid, spectinomycin, streptomycin, sulfisoxazole and tetracycline. Out of the 12 Salmonella Braenderup isolates, 10 (83.3%) showed multidrug resistance to ampicillin, spectinomycin, streptomycin, sulfisoxazole, sulfamethoxazole/trimethoprim, amoxicillin/clavulanic acid and trimethoprim. Among the 8 S. Hadar isolates 7 (86.5%) showed antimicrobial resistance of which three isolates showed resistance to streptomycin and tetracycline, two isolate showed resistance to tetracycline and the other two for streptomycin. All the 6 S. Dublin isolates were resistant to carbadox (100 %) while one was additionally resistant to tetracycline. All the 6 S. Haifa strain isolated were resistant for at least ampicillin, streptomycin and tetracycline. Up to ten different antimicrobial resistances pattern was observed. Antimicrobial resistance was most common among Salmonella isolated from chicken carcass (18/29, 62.1%) followed by pork (5/22, 22.7%). Multiple antimicrobial drug resistance was observed in 23 Salmonella isolates (23.5 %). The detection of 7.4% Salmonella carriers’ supermarket workers shedding S. Newport, the most prevalent serotype, suggests possible linkage and potential source of infection. The findings of the present study ascertain that Salmonella serotypes were widely distributed particularly in supermarket meat samples and significant proportion have developed resistance for routinely prescribed antimicrobial drugs both in veterinary and public health sectors. This poses considerable health hazards to the consumers unless prudent antimicrobial usage, adequate heat treatment, improvement of standards of hygiene and development and enforcement of suitable legislation, which safeguard consumers, are urgently instituted. Key words: Prevalence, Salmonella, serotype, antimicrobial resistance, food items, supermarket, personnel, Addis Ababa