College of Science and Health Theses and Dissertations

Date of Award

Fall 11-22-2016

Degree Type


Degree Name

Master of Arts (MA)



First Advisor

Leonard A. Jason, PhD

Second Advisor

Molly M. Brown, PhD


Cognitive behavioral theories of chronic fatigue syndrome (CFS) assert that cognitions and behaviors perpetuate the fatigue and impairment that individuals with CFS experience (Wessely, Butler, Chalder, & David, 1991). Vercoulen and colleagues (1998) utilized structural equation modeling to empirically develop a cognitive behavioral model of CFS. The resulting model indicated that attributing symptoms to a physical cause, focusing on symptoms, and feeling less control over symptoms were associated with increased fatigue. Additionally, individuals who attributed symptoms to a physical cause reported lower activity levels and more fatigue and impairment. However, in an attempt to replicate this model, Song and Jason (2005) demonstrated that the model displayed inadequate fit statistics for a well-characterized group of individuals with CFS; the model resulted in appropriate fit for individuals with chronic fatigue from psychiatric conditions. Despite uncertainty surrounding the model’s validity, it continues to be cited to support the application of cognitive behavioral and graded exercise therapies to individuals with CFS (White et al., 2011). The current study utilized second-stage conditional process modeling (i.e., moderated mediation) to reexamine the behavioral pathway of the Vercoulen et al. (1998) model. This pathway is characterized by the association among causal attribution for symptoms, activity level, and fatigue and impairment. The use of a large sample allowed for a robust examination of the pathway, and moderators isolated potential factors that contributed to previous studies’ discrepant results. Findings were generally inconsistent with the Vercoulen et al. (1998) model. Results indicated that individuals did not reduce their activity level due to illness beliefs. Although activity level and impairment were significantly correlated, this correlation decreased as case definition stringency increased. Furthermore, a canonical correlation analysis demonstrated that activity level, impairment, and fatigue could be conceptualized as indicators of illness severity. Rather than implicating activity level as the cause of fatigue and impairment, the relation among these variables may be due to their shared association with the latent construct of illness severity. This study represents the second attempt to replicate the Vercoulen et al. (1998) model; neither the Song and Jason (2005) nor the current study resulted in findings consistent with the original model. As this model provides the theoretical foundation for cognitive behavioral and graded exercise treatments for ME and CFS, these failed replication attempts support patient-expressed concerns about the appropriateness and efficacy of these treatments.

SLP Collection