Figure 1. Biobehavioral Research Framework.

The interaction between symptoms and the appropriate handling of the temporal nature of symptoms in response to pathophysiological and situational factors are major omission from many symptom models. Using latent growth mixture modeling, common and distinct trajectories of change among symptoms and biomarkers over time will be identified in response to ventricular assist device implantation (Figure 1-A). Identified symptom and biomarker trajectories (C₁ and C₂) may have different intercepts (i), slopes (s), and non-linear patterns of change (q) over time. Our framework was developed to codify the clinical relevance of differential responses to ventricular assist device implantation in the context of both patient-oriented and clinical outcomes. As such, associations between symptoms and biomarkers and the outcomes of health-related quality of life (Figure 1-A) and 6-month clinical event-risk (Figure 1-B) will be quantified. Our framework also includes predictors of favorable and unfavorable symptom and biomarker responses. We will explore a number of factors in order to identify patients at greater risk for symptom burden, worse pathogenic responses, poor health-related quality of life, and/or greater event-risk (Figure 1-C). Figure 1A-C match with our identify, associate, predict statistical approach to specific aims 1 and 2. To effectively capture interactions among symptoms and indices of pathogenesis, patterns of association among multiple symptoms and among multiple pathogenic biomarkers over time will be quantified using parallel process modeling in a symptom biochemistry element (Figure 1-D). In sum, our biobehavioral research framework has elements of symptom science, heart failure pathophysiological research, symptom biology/biochemistry, and clinical and patient-oriented outcomes research.