Skip to main content
. Author manuscript; available in PMC: 2012 Jan 1.
Published in final edited form as: J Am Geriatr Soc. 2010 Nov 4;59(1):18–25. doi: 10.1111/j.1532-5415.2010.03169.x

Table 2.

Univariate Associations Between Covariate Measures and Repeated Body Mass Index Measurement over Time Using Mixed-Effect Modeling

Parameter Estimate (Standard Error Estimate) P-Value
Variable Main Effect Interaction with Time (Slope)
Age −0.200 (0.029) <.001 −0.012 (0.003) <.001
Female 0.837 (0.347) .02 −0.113 (0.031) <.001
Education −0.028 (0.057) .63 0.009 (0.005) .07
Alcohol
 Current −0.482 (0.304) .11 −0.005 (0.038) .89
 Past −0.361 (0.288) .21 −0.009 (0.031) .77
 Never Reference Reference
Smoking
 Current −1.291 (0.355) <.001 −0.018 (0.040) .65
 Past −0.365 (0.301) .22 0.015 (0.030) .60
 Never Reference Reference
History of
 Cancer −0.020 (0.235) .93 0.0048 (0.035) .89
 Hypertension −0.143 (0.283) .61 −0.102 (0.040) .01
 Diabetes mellitus 0.610 (0.214) .004 0.049 (0.030) .10
 Heart attack 0.152 (0.212) .47 −0.003 (0.030) .91
 Stroke −0.062 (0.229) .79 −0.021 (0.039) .60
 Depression 0.207 (0.233) .37 0.007 (0.039) .85

Each model includes one covariate, main effect of time and time squared, and an interaction term between the covariate and time.