Background: Recent years have seen the establishment of bariatric surgery credentialing processes, center-of-excellence programs, and fellowship training positions. The effects of center-of-excellence status and of the presence of training programs have not previously been examined. The objective of this study was to examine the effects of case volume, center-of-excellence status, and training programs on early outcomes of bariatric surgery. Study Design: Data were obtained from the Nationwide Inpatient Sample from 1998 to 2006. Quantification of patient comorbidities was made using the Charlson Index. Using logistic regression modeling, annual case volumes were analyzed for an association with each institution's center-of-excellence status and training program status. Risk-adjusted outcomes measures were calculated for these hospital-level parameters. Results: Data from 102,069 bariatric operations were obtained. Adjusting for comorbidities, greater bariatric case volume was associated with improvements in the incidence of total complications (odds ratio [OR] 0.99937 for each single case increase, p = 0.01), in-hospital mortality (OR 0.99717, p < 0.01), and most other complications. Hospitals with a Fellowship Council-affiliated gastrointestinal surgery training program were associated with risk-adjusted improvements in rates of splenectomy (OR 0.2853, p < 0.001) and bacterial pneumonias (OR 0.65898, p = 0.02). Center-of-excellence status, irrespective of the accrediting entity, had minimal independent association with outcomes. A surgical residency program had a varying association with outcomes. Conclusions: The hypothesized positive volume-outcomes relationship of bariatric surgery is shown without arbitrarily categorizing hospitals to case volume groups, by analysis of volume as a continuous variable. Institutions with a dedicated fellowship training program have also been shown, in part, to be associated with improved outcomes. The concept of volume-dependent center-of-excellence programs is supported, although no independent association with the credentialing process is noted.