Examining the Effects of Treatment and Patient Characteristics on Patient Profiles of Depression, Anxiety and Functionality after an Acute Coronary Syndrome

Abstract

Depression and anxiety are linked to higher mortality rates and diminished quality of life in patients with an acute coronary syndrome (ACS). These psychiatric disorders are very prevalent in ACS patients, but they are still under treated. The objectives of this study were to identify symptomatology profiles of depression, anxiety, and functional impairment among ACS patients, describe change over time in their symptomatology, and determine if patients receiving cognitive behavioral therapy (CBT) showed signs of remission earlier than patients that received the usual care. In addition, the association between symptomatology and age, gender and length of stay in the hospital was examined. One hundred ACS patients with scores on the Hospital and Anxiety Depression Scale (HADS) indicatingmild to severe depression and/or anxiety at one month post-hospital discharge were enrolled in a randomized trial of CBT and were followed for changes in anxiety and depression at two, three, and six months post-baseline. Latent transition analysis was used to identify symptomatology profiles. Generalized estimating equations were used to predict symptomatology profile from treatment, gender, age, and length of hospital stay. The latent transition model specifying a three-class solution was selected to identify and describe symptomatology profiles. One class was characterized by patients with both psychiatric disorders and impaired functionality; the second class by patients with psychiatric disorders but normal functionality and the third class by patients with anxiety but without depression, having normal functionality. Results indicated that patients who received CBT reported higher rates of symptomatology remission between baseline and month-two followup, compared to patients in the control group. Further, women showed a better response to CBT than men. The use of this statistical approach clarifies periods during which an intervention is most likely to be effective. Abstract presented at the 42nd Annual Meeting of the Society for Epidemiologic Research, Anaheim, California, June 23-26, 2009.