A longitudinal analysis of body mass and morbidity: evidence from the health and retirement study

TitleA longitudinal analysis of body mass and morbidity: evidence from the health and retirement study
Publication TypeThesis
Year of Publication2014
AuthorsBoydstun, J
AdvisorCossman, JS
Number of Pages193
Date Published2014
UniversityMississippi State University
Thesis TypePh.D.
Accession Number1566481933
KeywordsDisabilities, Health Conditions and Status, Methodology, Public Policy

This dissertation aimed to investigate the direct and indirect pathways between body mass and three indicators of morbidity - health conditions, disability, and self-rated health status - and how these associations differed across birth cohorts. Using data from the Health and Retirement Study, a nationally representative longitudinal sample of adults age 51 or older, I used latent growth curve modeling to investigate the association of body mass on morbidity for five cohorts. Latent growth curve models allow for changes in body mass, health conditions, disability, and self-rated health over several waves of data to be assessed. To determine if the influence of education on health has declined, I also stratified each cohort by educational attainment groups. The results show that starting body mass is positively associated with starting health conditions; no other pathway examined was significant for all cohorts. In the three youngest cohorts, higher starting body mass was associated with worse self-rated health. Changes in body mass over time were not significantly associated with changes in self-rated health over time for the two youngest cohorts, perhaps indicating that adults at older ages do not perceive higher body mass as detrimental to overall well-being. Education was a significant predictor of body mass and self-rated health for all cohorts. When cohorts were stratified based on educational attainment, adults with at least some college had fewer health conditions than their less educated peers in all cohorts except the youngest cohort. The relationship between the rate of change in body mass and the rate of change in health conditions was not significant for the most educated group in the youngest cohort. However, higher starting body mass resulted in worse self-rated health regardless of cohort or education. Despite inconsistent findings, the results potentially support an expansion of morbidity because the onset of illness does seem to be occurring at younger ages. The effect of education on body mass and health was mostly consistent across cohorts. Stratifying the data by educational attainment suggests that the youngest and best educated cohort has not benefited from the increased level of education relative to the older cohorts in the study.


Copyright - Copyright ProQuest, UMI Dissertations Publishing 2014 Last updated - 2014-10-10 First page - n/a

Endnote Keywords

Public health

Endnote ID


Short TitleA longitudinal analysis of body mass and morbidity: evidence from the health and retirement study
Citation Key6226