Table 5.

Multinomial logit model predicting whether former U.S. immigrants aged 50 years and older chose economics, family problems in Mexico, migration problems, difficult to be in U.S. or other reasons^a over illness as the main reaon for last returning from the U.S. (N=883)

Outcome	Economic reasons vs. illness	Missed family vs. illness	Family problems in Mexico vs. illness
Variables	OR [95% CI]	OR [95% CI]	OR [95% CI]
Timing of lives
Returned at age 50 and older	0.57 [0.15,2.22]	0.27** [0.11,0.68]	0.17** [0.05,0.65]
In in U.S. at least 10 years	0.16* [0.03,0.77]	0.45 [0.15,1.40]	1.04 [0.18,5.85]
Interplay of human lives and historical time
Current age (natural logarithm)	2.05 [0.01,384.50]	2.41 [0.052,110.70]	23 [0.21,2498.70]
Linked lives
Married/In union when last returned	0.10 [0.01,1.01]	0.16+ [0.02,1.29]	0.090* [0.01,0.82]
No. living children	0.93 [0.78,1.10]	0.99 [0.86,1.14]	1.1 [0.95,1.27]
Human agency
U.S. citizen/Legal Permanent Resident (imputed)	1.46 [0.35,6.06]	2.52 [0.73,8.71]	0.66 [0.10,4.20]
Completed elementary school	1.12 [0.27,4.64]	0.87 [0.26,2.96]	3.13 [0.67,14.70]
Male	10.5* [1.19,92.90]	1.35 [0.32,5.64]	0.64 [0.13,3.06]
Outcome	Migration problems vs. illness	Difficult to be in U.S. vs. illness	Other vs. illness
Variables	OR [95% CI]	OR [95% CI]	OR [95% CI]
Timing of lives
Returned at age 50 and older	0.41 [0.12,1.43]	0.30* [0.10,0.86]	0.43 [0.17,1.08]
In in U.S. at least 10 years	1.32 [0.34,5.09]	0.25* [0.075,0.83]	0.49 [0.16,1.46]
Interplay of human lives and historical time
Current age (natural logarithm)	52.2 [0.77,3524.40]	25.8 [0.39,1722.50]	475.1** [9.34,24173.70]
Linked lives
Married/In union when last returned	0.081* [0.01,0.65]	0.16+ [0.02,1.25]	0.15 [0.02,1.14]
No. living children	0.9 [0.77,1.04]	0.95 [0.82,1.11]	0.83** [0.73,0.95]
Human agency
U.S. citizen/Legal Permanent Resident (imputed)	0.074** [0.01,0.43]	1.77 [0.47,6.65]	1.37 [0.44,4.25]
Completed elementary school	2.54 [0.57,11.40]	1.69 [0.47,6.17]	1.57 [0.47,5.25]
Male	4.55 [0.82,25.30]	1.34 [0.28,6.42]	1.51 [0.36,6.31]

Source: Authors’ calculations using the 2012 Mexican Health and Aging Study (MHAS). Estimates weighted using MHAS-provided sampling weights. Every column represents a comparison of the characteristics of immigrants who returned to Mexico for that particular reason compared to those who returned due to illness. All columns represent comparisons within the same model.

Notes: Standard errors are clustered at the household level.

^aThe reference category is missing family.

^*p<0.05

^**p<0.01

^***p<0.001