Survival analysis in social neuroscience and public health: A research exemplar from the field of cognitive epidemiology

Abstract

Survival analysis is a group of methods designed to track the association between an exposure variable and the probability of some discrete outcome over time. Some examples of applications of survival analysis include the time to relapse among current smokers who are abstaining, tracking medication nonadherence over time among those living with diabetes, and the prediction of mortality as a function of cognitive ability. We use this latter example to illustrate one approach to survival analysis from the field of cognitive epidemiology. In this chapter we seek to answer the question: Is some aspect of executive functioning associated with time until mortality? To achieve this goal, we utilize one of the most important models in biostatistics and public health, the so-called Cox proportional hazards model (Cox 1972). This model will allow us to investigate the survival association of interest while adjusting for important confounding variables. Through this example, we hope that readers will be convinced of the importance of considering the time until an event occurs, as opposed to a binary version of the event, due to the gain in information that the former approach provides.