Document Type


Date of Degree

Summer 2016

Degree Name

PhD (Doctor of Philosophy)

Degree In


First Advisor

Christensen, Alan J.

First Committee Member

Lutgendorf, Susan

Second Committee Member

O'Hara, Michael

Third Committee Member

Pagedar, Nitin

Fourth Committee Member

Vander Weg, Mark


Although unintentional weight loss (UWL) and depressive symptoms are critical outcomes following diagnosis and treatment for head and neck cancer (HNC), there is a limited understanding of how they influence one another over time. As part of a large, prospective study on HNC outcomes, growth curve modeling was used to evaluate 564 patients’ trajectories of depressive symptoms and percentage UWL and analyze longitudinal associations between these variables across the first year following HNC diagnosis. The hypothesized temporal precedence model was not supported—pretreatment depressive symptoms predicted neither total percentage weight loss at 6 months (t(561) = -1.50, p = .13), nor rates of curvilinear change in percentage weight loss over time (t(561) = 1.38, p = .17). The opposite temporal precedence model also lacked support—early weight loss predicted neither level of depressive symptoms at 6 months (t (432) = 0.24, p = .81), nor rates of linear change in depressive symptoms over time (t (432) = 1.31, p = .19). Instead, a pattern of concurrent covariation emerged—changes in depressive symptoms over time were associated with concurrent changes in UWL (t (1148) = 2.05, p = .041) and changes in UWL over time were associated with concurrent changes in depressive symptoms (t (556) = 2.43, p = .015). That is, to the extent that depressive symptoms increased on a monthly basis, patients lost incrementally more weight than was lost due to the passage of time, and to the extent that weight loss increased on a monthly basis, depressive symptoms also increased.

Together, these bidirectional results depicted an ongoing transactional interplay between depressive symptoms and UWL across time, such that changes in either variable resulted in deviations from the average trajectory of the other variable. Patient-reported pain and eating abilities emerged as potential mechanisms through which these variables influence one another. The results have important clinical implications, indicating that ongoing screening and treatment for depression and weight loss throughout the first year after HNC could benefit patients’ psychological and nutritional outcomes alike.


depressive symptoms, head neck cancer, weight loss


viii, 95 pages


Includes bibliographical references (pages 85-95).


Copyright 2016 Julia Rose Van Liew

Included in

Psychology Commons