Project Summary
Older United States veterans are a population at risk for disability due to their early life experiences with
military service and increasing age, and they may experience more physical health difficulties compared to
civilians. Military service was a normative part of early adult life for today’s older U.S. population. The
number of veterans aged 65 and older in the U.S. in 2015 was 9.3 million, and that number is projected to
rise as veterans of the Vietnam War continue to age. Despite the lived experience and substantial number
of older veterans, the details of their late life disability experience are not well understood. One major
obstacle in the field is that non-random selection into the military biases the effects of early life military
service on long term disablement processes and mortality, making it difficult to accurately pinpoint the
mechanisms explaining the mixed findings in the literature. It is unclear if the key lies in the difference in the
type and severity of chronic conditions, the nature of disablement and its progression over time, the
mortality risk related to disease and disability, specific service related experiences, or a combination of
these. To address these issues, the proposed project will first account for non-random selection forces into
military service using propensity score matching. Then, I will determine how chronic conditions influence
late life disability processes and mortality among veterans and civilians at the population level, and how
service related experiences of veterans may fuel disparities. Data for this project come from the nationally
representative, longitudinal Health and Retirement Study, which includes information on chronic conditions,
disability, mortality, pre-service (childhood) characteristics, and past military service, including the 2013
Veteran Mail Survey (VMS) data supplement on veteran military experiences. After non-random selection
into the military is accounted for, Aim 1 will assess how chronic conditions differentially impact the process
of disablement of veterans over time compared to civilians using latent growth curves (LGC), including
analysis of effects on timing and accumulation of disability using latent growth curves with random onset
(LGCRO). Aim 2 will determine the overall effect and timing of chronic conditions and disability on mortality
in later life for veterans and civilians by treating a binary mortality variable as a distal outcome of the
chronic condition and disability trajectories. Aim 3 will re-examine the components of the disablement
process most salient to veteran/civilian disparities (identified in Aims 1 and 2) to determine if service related
experiences are implicated in poor health processes among older veterans. This will be accomplished by
comparing the veteran subsample in the VMS to a matched, random sample of civilians within the core
HRS, using LGCs and traditional regression methods. Results from the study will provide insight into the
disablement process, mortality, and the factors contributing to health disparities, with the intent to inform
intervention development and policy decisions that can improve the overall health equity of U.S. veterans.