Skip to main content
. Author manuscript; available in PMC: 2017 Jul 1.
Published in final edited form as: Disabil Health J. 2016 Feb 13;9(3):524–532. doi: 10.1016/j.dhjo.2016.01.011

Table 2.

Mixed-Effects Regression Model of 2001–2012 ADL Scores of Adult Mexicans Aged 60 or Older (top) and Marginal Effects of Selected Covariates (bottom)

Variables Model 95% C.I.

Fixed Portion
 Time 0.035*** 0.025–0.045
Socioeconomic Variables
 Age
 Age Centered −0.009*** −0.012 – (−0.005)
 Age Centered Squared 0.001*** 0.001 – 0.001

 Gender
 Female 0.024* 0.001 – 0.047

 Marital Status
 Married −0.003 −0.027 – 0.021

 Education
 Years of Schooling −0.002 −0.045 – 0.001

 Number of Limitations
 ADL Score at Baseline 0.746*** 0.732 – 0.760

 Attrition
 Dead 0.084*** 0.046 – 0.122
 Number of Contributions −0.007 −0.041 – 0.027

 Interactions
 Female & Time 0.015*** 0.006 – 0.025
 Education & Time −0.001* −0.002 – (−0.001)
 Marital Status & Time −0.009 −0.018 – 0.001

 Constant 0.047 −0.059 – 0.152

Random-Effect Parameters
 Variance (Time) 0.122 0.107 – 0.141
 Variance (Constant) 0.011 0.009 – 0.014
 Covariance (Time, Constant) 0.012 0.010 – 0.013
 Variance (Residual) 0.241 0.230 – 0.252

LR Test vs. Linear Regression
 Chi-Square 2706.66***

 Number of Observations 13,988
 Respondents 5,752
 Average Number of Observations per Respondent 2.4
Variables Marginal Effects 95% C.I.

Time
 2 Years 0.03*** 0.01 – 0.05
 11 Years 0.38*** 0.33 – 0.43

Socioeconomic Variables
 Age Centered 0.003** 0.001 – 0.004
 Years of Education −0.06** −0.001 – (−0.002)
 Female 0.07*** 0.04 - 0.11
 Married −0.02 −0.05 – 0.01

Note:

*

p = .05;

**

p = .01;

***

p = .001.

Marginal effects with the rest of the covariates at their means.

Source: Author’s calculations using data from the Mexican Health and Aging Study2729.