OBJECTIVES: This study measured the association between hospital characteristics and changes in hospital charges for major joint replacement (MJR). METHODS: RESULTS: The analysis included 3,016 hospitals.  In multivariate models, many hospital characteristics were significantly associated with the percent change in charges. Hospital type and census region were significantly related to percent change in charges.  Specifically, Government and Non-Profit hospitals experienced smaller charge increases than Proprietary hospitals (-5.5% and -4.5% respectively, p<0.0001).  Compared to the Northeast, the West, South, and Midwest experienced 5.5%, 4.4%, and 3.3% increases in charges on average (p≤0.001), while the Midwest, West, and South were not statistically significantly different from each other (p>0.05). Also, a one-day increase in the average length of stay was associated with an average charge increase of 6.7% (p<0.0001). Overall, the final model accounted for 9% of the total variance. CONCLUSIONS: Hospital characteristics, such as the region and type, as well as changes to average length of stay between 2008 and 2011 were strong predictors of percent change in average charge for MJR.  While several other factors were also statistically significantly related to change in charges for MJR, small effect sizes were not practically meaningful. The small percent of variance explained by the model leads us to conclude there are other factors not captured by Medicare data that are additionally responsible for growth in healthcare costs.