Marital Status, Marital History, Body Weight, and Obesity

Abstract

Marital status and marital history are associated with health. Marital history can be represented by the marital trajectory components of timing, transitions, sequence, and duration. We examined whether marital trajectory components add insights beyond marital status in predicting body weight in a retrospective analysis of 3,011 adults. Marital status findings revealed that married men were heavier than separated/divorced men, and never married women were heavier and more often obese than married women. Marital history findings showed that after adjusting for marital status, trajectory measures of age at first marriage, second marriage or second divorce, experiencing widowhood, and duration of separation/divorce were not clearly associated with body weight or obesity. Body weight and obesity appear to be associated with current marital status but not marital history.

KEYWORDS:

• body weight
• marital history
• marital status
• marital trajectories
• obesity

ACKNOWLEDGMENTS

This analysis was supported by the National Research Initiative of the USDA Cooperative State Research, Education, and Extension Service (CSREES) grant 2005-35215-15752. We thank Edward A. Frongillo and Francoise Vermeylen for assistance.

Notes

Sample data are weighted.

Numbers in parentheses are standard deviations.

Statistically significant differences are denoted: * = p < .05 and ** = p < .01.

(1) Sub-sample of ‘ever married’ respondents excludes those who were never married, cohabiting or had greater than two marriages.

Sample data are weighted.

Numbers in parentheses are standard errors.

Statistically significant differences are denoted: * = p < .05 and ** = p < .01.

Regression parameter estimates are based upon regressions of five separate imputed datasets, combined using PROC MIANALYZE.

Reference indicates the reference category to which all other categories are compared.

Numbers in parentheses are standard errors.

Statistically significant differences are denoted: * = p < .05 and ** = p < .01.

Regression parameter estimates are based upon regressions of five separate imputed datasets, combined using PROC MIANALYZE.

Reference indicates the reference category to which all other categories are compared.

Sample data are weighted.

Numbers in parentheses are standard errors.

Statistically significant differences are denoted: * = p < .05 and ** = p < .01.

Regression parameter estimates are based upon regressions of five separate imputed datasets, combined using PROC MIANALYZE.

Reference indicates the reference category to which all other categories are compared.

Sample data are weighted.

Numbers in parentheses are standard errors.

Statistically significant differences are denoted: * = p < .05 and ** = p < .01.

Regression parameter estimates are based upon regressions of five separate imputed datasets, combined using PROC MIANALYZE.

Reference indicates the reference category to which all other categories are compared.

Sample data are weighted.

Numbers in parentheses are standard errors.

Statistically significant differences are denoted: * = p < .05 and ** = p < .01.

Regression parameter estimates are based upon regressions of five separate imputed datasets, combined using PROC MIANALYZE.

Reference indicates the reference category to which all other categories are compared.