OBJECTIVES: To evaluate the clinical and economic outcomes based on bariatric surgery type. METHODS: 2005 and 2006 National Inpatient Sample (NIS) datasets were utilized as the data source.  The bariatric population was defined by ICD-9 codes (44.38, 44.39, 44.68).  A diagnosis of morbid obesity was required for inclusion; records with a diagnosis of a stomach or intestinal cancer were excluded from the analysis.  RESULTS: Each NIS dataset contains approximately 8,000,000 hospital discharge records; the bariatric surgery sub-population contains 33,938 records.  The sample population primarily consisted of females (82%) with a mean age of 43 years.  Most of the procedures were performed at urban hospitals (53%), and the most common payer was private insurance (75%).  Clinical and economic outcomes were compared by surgery type; the surgery types were: open gastric bypass (OGBP), laparoscopic gastric bypass (LGBP) and laparoscopic banding procedure (LB).  The most common procedure was the laparoscopic gastric bypass procedure (n = 26,465).  Length of stay (in days) differed significantly between procedures (OGBP = 3.8, LGBP = 2.5, LB = 1.6, p <0.0001).   Total charges differed significantly between procedures (OGBP = $ 40,770, LGBP = $ 37,306, LB = $ 26,859, p <0.0001).  The complication rate was low (<2% for all procedures); mortality was also low (<0.5% for all procedures).  A multivariate model was used to assess the relationship between procedure type and predictor variables (age, gender, hospital location, year, and race); race, gender and hospital location were statistically significant. CONCLUSIONS: The laparoscopic banding procedure was associated with the least cost and shortest length of stay.  Mortality and complication rates were low for all surgery types.  Females were more likely to receive the laparoscopic gastric bypass; males were more likely to receive laparoscopic banding.