Fig. 3.

Adjusted β and 95%CI for OBS and serum klotho level, stratified by several key risk factors. Models were adjusted for age and sex, race, marital status, education status, and poverty income ratio, daily energy intake, whole grain, milk, red meat, cured meat and egg intake. In each stratified analysis, the stratification variable was excluded in the adjustments. Likelihood ratio tests were used for assessment of interaction, and two-sided P values (unadjusted for multiple comparisons) are reported.