Graduate School of Biomedical Sciences, Clinical & Population Health Research Program
Depression; Anxiety Disorders; Cognitive Therapy; Models, Statistical; Acute Coronary Syndrome; Academic Dissertations; Dissertations, UMMS
Life Sciences | Medicine and Health Sciences
The purpose of the current project is to illustrate the application of advanced statistical techniques to address research questions about depression and anxiety in patients with an acute coronary syndrome (ACS). The first study, using data from 100 patients who were randomized into a clinical trial of cognitive behavioral therapy, used bivariate mixed models to determine trajectories of depression and anxiety after an ACS, to examine the effects of cognitive behavioral therapy (CBT) on depression and anxiety, and to determine if anxiety and depression symptoms change at the same rate with CBT treatment as indicated by joint modeling of these two psychiatric disorders. The findings suggest that depression and anxiety are highly correlated and persistent in patients with an ACS both at baseline and over time. The intervention used in the present investigation does not appear to uncouple the association between anxiety and depression, suggesting that CBT has comparable effects on both psychiatric disorders.
The second study used latent transition analysis to identify symptomatology profiles of depression, anxiety, and functional impairment in patients with an ACS, describe changes over time (two, three and six-month follow-up) in patient’s acute symptom profiles, and determine if patients receiving CBT showed signs of remission in depression, anxiety and impaired function earlier than patients that received usual care. A three-class model was selected to identify and describe these acute symptom profiles. One class was characterized by patients with both psychiatric disorders and impaired function, the second by patients with psychiatric disorders but normal function, and the third by patients with anxiety but without depression, and having normal function. There was moderate improvement in depression, anxiety and functional status for control patients, but this improvement was less evident than in the treatment group. Women showed a better response to CBT than men.
The third study used latent class and latent transition analysis to determine symptom profiles of depression and anxiety in patients with an ACS using the Hospital Anxiety Depression Scale; a secondary study goal was to examine the effects of age and gender on these symptom patterns. A two-class model was selected to describe depression and anxiety symptomatology profiles. Class I (76% of patients at baseline) was labeled as “severe depression and some anxiety” whereas Class II (24% of patients at baseline) was labeled as “mild depression and distress anxiety”. More than 70% of older patients continued to have severe depression and anxiety at follow-up and a large proportion of these patients who reported mild depression and anxiety at baseline showed worsening of symptoms at follow-up. The current study demonstrates that patients with depression and anxiety after an ACS can be identified on the basis of the symptoms that they present. This is particularly important to identifying individuals at potential risk for developing clinical complications after an ACS.
Tisminetzky, Mayra, "Modeling Co-Occurring Depression and Anxiety in Patients with an Acute Coronary Syndrome: A Dissertation" (2009). University of Massachusetts Medical School. GSBS Dissertations and Theses. Paper 421.