Preventing Aggressive/Violent Behavior: A Role for Biomarkers?

Abstract

The burden of aggressive/violent behavior Despite a declining rate in the past 10 years, aggressive/violent behavior remains among the main causes of worldwide premature mortality [1]. Associated direct (medical and nonmedical expenditure) and indirect (longterm societal implications) socio-economic costs are also substantial [2,3]. These have been estimated in 2001 at approximately 5% of the gross national product of industrialized countries, and as much as 14% of the gross national product of low-income countries [3].In this context, policy makers have attempted to enact interventions that could prevent the development of aggression and violence in at-risk populations (i.e., abused children and troubled youth, alcohol and substance abusers, prisoners). These public health approaches that include, among the others, education, behavior change, policy, engineering and environmental support, have proven to be effective in reducing the rates of aggression/violence as well as its associated economic costs [4].Nonetheless, there is still a substantial proportion of individuals who enact aggressive/violent behavior over the lifetime course that is not targeted by these interventions. Further, such interventions are implemented only once aggressive behavior has already manifested itself.By definition, primary preventive strategies intend to avoid the manifestation of a specific disease or event before it occurs. To be effective, preventive approaches need to identify risk factors that can determine the onset of an illness or of an event in an exposed population.In the case of aggressive/violent behavior, there is a multitude of determinants (clinical, environmental and biological) that set the liability threshold for its manifestation. It is plausible that predictive models integrating different levels of information (from phenotypic to 'omics' data) might help identifying with higher accuracy individuals at risk for aggression/violence compared with those relying only, for instance, on clinical data. However, the inclusion of biological markers (henceforth biomarkers) in such models can be achieved only if they are identified in longitudinal prospective studies and are replicated in independent samples. Importantly, this process could also lead to a more precise understanding of the neurobiological underpinnings of aggression/violence.Here, we critically discuss the recent advancements in the identification of biological markers of aggression and the eventual barriers toward their implementation in predictive models. To this end, we will first briefly present the current knowledge on the neurobiological make up of aggression/violent behavior. Next, we will discuss how the phenotypic heterogeneity of aggression might have hindered the identification of reliable biomarkers. Finally, we will present data on available biological markers of aggression that might be suitable for implementation in predictive models.