Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Health Outcomes and Policy Research

Research Advisor

Ben Zarzaur, M.D., M.P.H.


Robert J. Nolly, M.S., D.Ph. Esra Ozdenerol, Ph.D. George Relyea, M.S. Junling Wang, M.S., Ph.D.


aging, functional status, injury, mediation, trajectory


With nearly 40 million Americans being treated for nonfatal injuries in emergency departments annually, the burden of nonfatal injury in the United States is extensive. The longterm functional consequences of these injuries can be enduring and far-reaching, especially for older adults. Although studies have reported that persistent functional deficits exist after injury, less information is known about long-term recovery patterns and the factors that influence functional outcomes.

The primary aims of this study are to (1) classify differences in long-term, longitudinal changes in functional limitations within the injured population, (2) identify individual characteristics that predict recovery after injury and (3) assess whether medical care use mediates the relationship between long-term changes in functional limitations and significant predictors of outcomes after injury, specifically insurance status. Longitudinal survey data from the Health and Retirement Study (HRS) was analyzed using group-based trajectory modeling and mediation analysis.

The trajectory analysis identified five distinct functional trajectories with the following characteristics: Trajectory 1- consistently low functional limitations scores (18.9%), Trajectory 2- increase in functional limitations after injury followed by a gradual, but not complete recovery (46.3%), Trajectory 3- increase in functional limitations followed by further decline in functioning over time (10.5%), Trajectory 4- increase in functional limitations after injury followed by a gradual, complete recovery (13.4%), and Trajectory 5- consistently high functional limitations scores (10.8%). Regression analyses showed that women, individuals with multiple health conditions, and individuals with no insurance and public insurance were more likely to belong to trajectories with poorer functional outcomes.

The mediation analysis found that public insurance was associated with increased functional limitations relative to private insurance. The total effect of public insurance on functional limitations was partially mediated by medical care use. Doctor visits was the only significant medical care use mediator for individuals with public insurance. The relative total and direct effects of being uninsured on functional limitations were not significant. However, the indirect effect of being uninsured on functional limitations was significant, indicating that medical care use may suppress the effect of being uninsured on functional outcomes. Prescription drug use was the only significant mediator of the effect of not having insurance on functional status.

These results illustrate that distinct courses of recovery after injury in the older adult population exist. Furthermore, personal characteristics of individuals can be used to predict functional trajectories. This study also demonstrated that insurance status is a significant predictor of both functional outcomes and medical care use after injury. Insurance status was found to exert its effect on health outcomes both directly and indirectly through medical care. Ultimately, the findings from this study can be used to improve the understanding of how individuals' functional outcomes differ after injury and the causal processes that determine these outcomes. This knowledge may lead to tailored policies and treatments that improve quality of life after injury.




Two year embargo expired May 2016