Domestic violence homicide is among the most difficult forms of human violence to understand, investigate, and prevent. Unlike many criminal acts that arise from transient opportunity or impulsive aggression, intimate partner homicide typically develops over weeks, months, or even years through a progressive pattern of coercion, domination, escalating threats, and increasingly dangerous behavior. Although each case is unique, retrospective reviews consistently reveal that many offenders exhibited recognizable behavioral warning signs before the homicide occurred. The challenge for professionals is not the absence of warning signs, but rather determining which combinations of behaviors signify an elevated risk of lethal violence and require immediate intervention.
There is no single roadmap for understanding the complexity of human behavior in general, and lethal domestic violence in particular. Human behavior is shaped by a dynamic interaction of psychological, biological, developmental, social, cultural, and situational influences that often defy simple explanation. If a reliable roadmap existed, domestic homicide might be prevented with far greater certainty. Unfortunately, the behavioral sciences have not advanced to the point where clinicians, law enforcement officers, or researchers can accurately predict who will commit murder or precisely when escalating anger, desperation, coercive control, or perceived loss will culminate in lethal violence (Sefton, 2013).
The inability to predict violence with certainty, however, should not be confused with an inability to recognize danger. Modern research has demonstrated that many perpetrators of intimate partner homicide follow a discernible trajectory characterized by escalating risk factors, identifiable behavioral “red flags,” and repeated opportunities for intervention. Prior assaults, threats to kill, non-fatal strangulation, stalking, obsessive jealousy, forced sexual behavior, violations of protective orders, access to firearms, suicidal ideation, and increasing social isolation have each been associated with substantially elevated risks of homicide (Campbell et al., 2003; Hilton et al., 2004; Messing et al., 2017). While none of these factors alone can predict homicide, their cumulative presence substantially increases concern and provides professionals with an evidence-based framework for assessing dangerousness.
Accordingly, the objective of domestic violence risk assessment is not to predict the future with absolute certainty but to improve the quality of professional judgment by identifying measurable indicators of escalating danger. Structured professional judgment instruments, actuarial risk assessment tools, psychological autopsies of domestic homicide offenders, and multidisciplinary case reviews have collectively demonstrated that lethal domestic violence is rarely a spontaneous or unpredictable event. More commonly, it represents the final stage of an escalating pattern of abuse in which warning signs were present but not fully recognized, effectively communicated, or acted upon. The continued refinement of risk assessment practices, coupled with coordinated intervention among law enforcement, the courts, behavioral health professionals, and victim advocates, offers one of the greatest opportunities to reduce the incidence of domestic violence homicide.
Instead, researchers and practitioners rely on identifying empirically supported risk factors and behavioral warning signs rather than making definitive predictions. One important method for advancing this knowledge is the psychological autopsy—the systematic retrospective examination of an individual’s behaviors, relationships, mental state, life circumstances, and documented interactions before a violent death. By reconstructing the sequence of events that preceded a homicide, investigators can identify recurring behavioral patterns, commonly referred to as red flags, that were present before the fatal incident.
The psychology underlying this escalation is rooted in coercive control. Many perpetrators do not view their partners as autonomous individuals but rather as extensions of themselves whose behavior must be monitored, controlled, and restricted. When victims assert independence—whether by seeking assistance, filing criminal charges, obtaining a restraining order, or beginning a new relationship—the abusive partner may experience profound narcissistic injury, humiliation, abandonment, or loss of dominance.
The inability to predict violence with certainty, however, should not be confused with an inability to recognize danger. Modern research has demonstrated that many perpetrators of intimate partner homicide follow a discernible trajectory characterized by escalating risk factors, identifiable behavioral “red flags,” and repeated opportunities for intervention. Prior assaults, threats to kill, non-fatal strangulation, stalking, obsessive jealousy, forced sexual behavior, violations of protective orders, access to firearms, suicidal ideation, and increasing social isolation have each been associated with substantially elevated risks of homicide (Campbell et al., 2003; Hilton et al., 2004; Messing et al., 2017). While none of these factors alone can predict homicide, their cumulative presence substantially increases concern and provides professionals with an evidence-based framework for assessing dangerousness.
Accordingly, the objective of domestic violence risk assessment is not to predict the future with absolute certainty but to improve the quality of professional judgment by identifying measurable indicators of escalating danger. Structured professional judgment instruments, actuarial risk assessment tools, psychological autopsies of domestic homicide offenders, and multidisciplinary case reviews have collectively demonstrated that lethal domestic violence is rarely a spontaneous or unpredictable event. More commonly, it represents the final stage of an escalating pattern of abuse in which warning signs were present but not fully recognized, effectively communicated, or acted upon. The continued refinement of risk assessment practices, coupled with coordinated intervention among law enforcement, the courts, behavioral health professionals, and victim advocates, offers one of the greatest opportunities to reduce the incidence of domestic violence homicide.
Although no single warning sign is sufficient to predict homicide, the accumulation of risk factors can substantially increase concern. Research on domestic violence has identified common indicators such as escalating threats, obsessive jealousy, coercive control, stalking behaviors, prior strangulation, access to firearms, repeated violations of protective orders, and recent relationship separation. Examining these factors across many cases allows researchers to identify patterns that might otherwise remain unnoticed.
The practical value of psychological autopsy research lies not in predicting the exact moment violence will occur, but in improving risk assessment and informing early intervention. By recognizing recurring patterns of high-risk behavior, law enforcement, healthcare professionals, courts, and social service agencies can better justify proactive safety planning, enhanced monitoring, coordinated intervention, and, when legally appropriate, measures to contain individuals who present a heightened risk of serious or lethal violence. While behavioral science cannot eliminate domestic homicide, it can improve our ability to recognize danger earlier and intervene before violence escalates to its most tragic outcome.
Sefton, M (2013) Roadmap to Violence: What can we do to predict violence. Blog post: taken 6-30-2026, https://msefton019.wordpress.com/wp-admin/post.php?post=1315&action=edit
Leave a comment