Reply to the Letter to the Editor Obesity Epidemic: Is Its Impact on Total Joint Arthroplasty Underestimated? An Analysis of National Trends.

To the Editor,

We would like to thank Sloan and colleagues for their interest in our paper. We agree with the authors of the letter to the editor that the low prevalence of obesity estimated by Odum and colleagues using National Inpatient Sample (NIS) is likely due to the poor sensitivity of obesity coding in the NIS database [5, 7]. As they point out, the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database provides BMI values and can reliably estimate the prevalence of obesity in patients who underwent total joint arthroplasty (TJA). Despite its size, the NSQIP database only includes surgeries from participating hospitals, and is not considered to be nationally representative [1]. Moreover, the NSQIP database includes patients from countries other than the United States, and the prevalence of obesity from this source may not be applicable to the US population. As our study was aimed at evaluating the obesity trends specifically in the US population, and a database was not available to directly estimate the prevalence of obesity at a national level, we employed indirect mathematical models to obtain national estimates.

Sloan and colleagues also note that the rate of increase in the prevalence of obesity in TJA in the NSQIP databases is lower than the rate of increase observed in our study. While the obesity estimates in our study are subject to changes in relative risks, the annual changes in our estimates are most sensitive to changes in the obesity prevalence in the US general population [4]. Therefore, the higher growth rate of obesity observed in our study can be partly explained by the fact that our study used earlier years of data (1998 to 2011), and the rate of increase regarding the prevalence of obesity in the US general population was higher during that period. For example, the prevalence of obesity in the US population (as reported in the Behavioral Risk Factor Surveillance System [2]) increased by 51.9% (from 18.3% in 1998 to 27.8% in 2011) during the time period used in our study compared to 11.9% (from 26.7% in 2008 to 29.9% in 2016) during the time period referred to by Sloan and colleagues.

Although our study employs multiple assumptions and has its limitations, the estimates reported in our study are comparable to the estimates based on BMI in multiple single institutional studies suggesting that our method was able to provide a fairly accurate assessment of national obesity trends [3, 6]. However, further studies using different methodologies and multiple databases are encouraged to improve our understanding on the impact of obesity on TJA.