Course of illness and the development of vascular disease in individuals with bipolar disorder

For over a century, there have been suggestions of a link between what is currently called bipolar disorder and cardiovascular mortality. In the contemporary epidemiological literature, this risk has been confirmed and approximates twice that expected based on age and gender. To date, however, this information has come primarily from clinical samples, which carry considerable risk of selection bias. The studies contained in this dissertation sought to assess this relationship using methods less vulnerable to selection bias and to determine the role that course of illness and treatments for illness may play in the development of vascular disease. In a nationally representative sample, we confirmed a link between mood disorders and vascular disease, which was particularly pronounced in women with bipolar disorder. In subsequent studies, a dose-response relationship between the duration of clinically significant hypomanic or manic symptoms and both cardiovascular mortality and endothelial function was seen. While medication exposure did not appear related to mortality or endothelial function, first generation antipsychotics were associated with arterial stiffness, an effect apparently mediated by elevations in blood pressure. In cross-sectional samples, our data suggests that vasculopathy is not present early in the course of bipolar disorder although is much greater than expected later in the course of illness. This dissertation purports that vasculopathy develops over the long-term course of bipolar disorder, is proportional to symptom burden, and is influenced by health behaviors and treatments. These findings may provide opportunities for clinicians and those afflicted to intervene to address this excess risk of vascular morbidity and mortality.