All ETDs from UAB

Advisory Committee Chair

Sylvie Mrug

Advisory Committee Members

Edwin W Cook

Sigrid Ladores

Kevin J Leon

Zina Trost

Document Type


Date of Award


Degree Name by School

Doctor of Philosophy (PhD) College of Arts and Sciences


Cystic fibrosis (CF) is a life-threatening chronic illness which requires adherence to a many complex and time-consuming treatments. Adherence is suboptimal in this population. Thus, it is important to consider predictors of treatment adherence. Research has evaluated a number of cognitive factors that are related to adherence in groups with chronic illnesses, including those with CF. The present study considers three factors (health locus of control (HLC), perceived social support, and perceived disease severity) in order to address important gaps in the literature evaluating their relationships with treatment adherence. For each of these factors, the current research is limited by the use of mostly cross-sectional designs. The current study extended these literatures by using a longitudinal design to evaluate the relationships between these factors and treatment adherence both concurrently and over time. The first manuscript evaluated the relationship between HLC dimensions (Internal, Powerful Others, and Chance) and adherence to common treatments for CF both concurrently and over a 3-month period. This manuscript addressed inconsistencies in the literature by evaluating the effects of HLC on adherence with perceived disease severity as a moderator. The results of the first manuscript indicated that perceived severity interacted with Internal and Chance HLC to predict concurrent adherence to some treatments. The second manuscript evaluated the concurrent and longitudinal relationships between three sources of perceived social support (family, friends, and significant other) and treatment adherence directly and indirectly with depressive symptoms as a mediator. Previous research in this area has largely evaluated the effect of social support derived from family on adherence in children and adolescents. Contrary to the hypotheses, greater social support from all sources was related to lower adherence in some treatment domains (corticosteroid, diabetes, and nutritional adherence) over time. Further, the results of this study did not support an indirect effect of social support on adherence through depressive symptoms. The third manuscript evaluated direct linear and quadratic effects of perceived and clinically measured severity on treatment adherence concurrently and over a 3-month period. The direction of linear relationships observed between each severity measure and adherence were not consistent in the present study, nor were the relationships between severity and adherence consistent with the results of other studies. However, perceived severity was quadratically related to adherence to corticosteroids in the present study. Further, percentage predicted forced expiratory volume in one second (%FEV1; a clinical measure of pulmonary exacerbation) was quadratically related to nutritional recommendations and diabetes treatments. For these associations, greater adherence was observed at low and high levels of severity. Clinicians who care for individuals with CF should consider different cognitive factors that may be related to treatment adherence when trying to encourage adherence. Specifically, they may benefit from understanding the potential relationships between HLC, perceived disease severity, and adherence. Further, clinicians can work with patients to identify and address potential barriers to adherence (including number of people in the household, social engagement, etc.) using cooperative strategies. Mental health services should also be made available to individuals receiving care for CF in order to address psychosocial adjustment and depression-related non-adherence. Future studies should continue to evaluate the relationships between treatment adherence and the cognitive factors addressed in the current study (HLC, depressive symptoms, and perceptions of social support and disease severity) using more robust measures of adherence. Additionally, future studies should evaluate these relationships in larger, more representative samples.