Extract

We thank Drs. Keevil and Khaw (1) for their comments on our article (2) regarding on the possibility of overadjustment in the association between obesity and handgrip strength.

We agree with Drs. Keevil and Khaw that adjusting for baseline body weight when modeling the effect of obesity history on handgrip strength using analysis of covariance is not without problems. However, unlike in the example by Drs. Keevil and Khaw based on cross-sectional data (1), we utilized retrospective data on obesity and the explanatory variable was obesity duration not body mass index (BMI) itself.

Drs. Keevil and Khaw justify their criticism about overadjustment by the fact that earlier studies (and their data on EPIC Norfolk) have shown positive association between BMI and handgrip strength (3,4). BMI is a measure of the shape of the body, not the size of the body, but it correlates closely with body weight—a measure of body size (5). Consequently, BMI typically correlates positively with higher absolute grip strength. Therefore, muscle strength as a measure only makes sense when body size is taken into account. That is why, for example, weight lifting competitions have different weight categories.

You do not currently have access to this article.