Forensic mental health is an area of research, care and policy that the public most often hears about in the wake of tragedy – usually when a person living with severe mental illness is the perpetrator of a violent offense. These moments prompt intense debate about the association between mental illness and violence, the adequacy of public safety, and the provision of evidence-informed care (1, 2). Yet examples from around the world show that political responses often take the form of knee-jerk investments in control policies and public safety infrastructures, despite a lack of evidence for their effectiveness (3, 4). This is frequently done at the expense of upstream strategies: prevention programs, community and family support, opportunities for social and economic participation, and the broader work of reducing marginalization and poverty.

As of 2025, forensic mental health systems remain structured around individualized, deficit-focused models shaped by colonial logics, that is, institutional norms rooted in Western notions of knowledge, expertise, and authority. These frameworks pathologize, marginalize, and punish, leaving unexamined the structural conditions that shape both risk and recovery. Do our systems set people up for failure?

In this paper, we call for a fundamental reorientation of forensic mental health research: shifting our attention toward systemic resilience, contextualized care, prevention, and community-grounded support. To ground this proposal, we first analyze the historical and conceptual legacy that shaped the field. Then, drawing from Andrews and Bonta’s Risk-Needs-Responsivity model (5, 6), Bronfenbrenner’s ecological systems theory (7), and Haldane et al.’s Health Systems Resilience Framework (8), we propose a new conceptual framework centered on systemic accountability. By opening new research avenues and reframing the knowledge base, we aim to inform practice and policy in ways that address both system-level risks and opportunities alongside individual-level needs, ultimately contributing to the prevention of violence and criminal justice involvement among people living with mental illness.

The bibliometric analysis shows that, despite decades of research, forensic mental health remains overwhelmingly focused on the individual: assessing their level of risk, diagnosing their deficits, and guiding their rehabilitation or containment. As a result, remarkably little is known about the systemic and structural factors that shape trajectories into, through, and beyond forensic systems. Few empirical studies, with some exceptions (38–42), have examined how policies, service design, institutional cultures, or broader sociopolitical forces contribute to violence or criminalization, or how these same forces affect recovery and reintegration.

There is a pressing need for a conceptual framework that extends beyond the individual to account for the interplay between personal, systemic, and structural factors. Such a framework could reposition responsibility not only on persons navigating these systems, but also on the societal and governmental decisions that shape their contexts. It could help shift the emphasis from managing risky individuals to interrogating risky systems; that is, systems that may reproduce stigma, exclusion, or punitive responses under the guise of care, thus creating risk within individuals.

We build on these three complementary models – the Risk-Need-Responsivity framework, Bronfenbrenner’s ecological systems theory, and health systems resilience approach – to propose a conceptual framework that expands how risk is understood and studied in forensic mental health. Figure 2 illustrates this framework across three interconnected levels (person, system, and structure), showing how each level aligns with the principles of awareness, focus, and transformability. These principles operate differently at each level, and the following section describes how they are operationalized in practice.

The original Risk-Need-Responsivity model (5) centers primarily on the individual and their immediate context. It involves risk awareness through assessing an individual’s likelihood of reoffending to calibrate intervention intensity (risk principle), focus by addressing criminogenic needs such as substance use or antisocial traits (need principle), and transformability through adapting interventions to the person’s learning style, culture, and identity to maximize engagement (responsivity principle). This layer is the person-level risk.

We build on this by proposing a first expansion: system-level risk (see Figure 2). While the individual’s risk remains the focus, this level acknowledges that risk is profoundly shaped by the institutions and systems surrounding them. Health, social, justice, and policing systems – including their structure, governance, and resource allocation – directly influence a person’s ability to access timely, appropriate care. These factors form part of the exosystem in Bronfenbrenner’s model and are key contributors to institutional and systemic risk. Drawing on the health systems resilience framework proposed by Haldane et al. (8), we can examine how risk is generated through breakdowns in governance, service delivery, workforce capacity, health information systems, financing, and community engagement. For example, insufficient staffing, poor continuity of care, and underrepresentation of marginalized communities in the workforce may erode trust and compromise safety. At this level, the principles of risk awareness, focus, and transformability that we derived from the Risk-Need-Responsivity model are operationalized as follow:

A second, more transformative shift occurs when considering structure-level risk – the macrosystemic conditions that shape how health, social, and justice systems operate. Structural risk refers to the deep-rooted political, historical, and ideological forces, such as colonialism, racism, and neoliberalism, that influence how institutions are designed, resourced, and governed. These forces manifest in laws, policies, professional norms, and dominant assumptions about safety, control, and accountability. Forensic systems embedded in structurally unjust contexts may consistently reproduce inequity, even when frontline practices are well-intentioned. At this level, the three principles are again re-operationalized:

To summarize, person-level risk corresponds to the microsystem – the characteristics of the individual and how these traits influence their own risk; system-level risk reflects how health, social, policing, and justice systems (i.e., exosystem institutions) shape individual risk; and structure-level risk operates at the macrosystem level, influencing how institutions function through broader ideologies, histories, and policies such as colonization and systemic racism.

To illustrate how systems and structures can actively produce risk, consider the case of a family member wanting to bring in home-cooked food for a service user during an event of cultural significance (e.g., Ramadan). At the person level, this may be a culturally responsive intervention. It supports the person’s spiritual and cultural needs, promoting a pro-social identity and perceived social support (49), all factors known to reduce risk and recidivism (49–52). Yet, at the system level, such needs are often deprioritized because they fall outside the dominant framing of “criminogenic needs” (53, 54). As a result, rules around food safety, security, and professional boundaries may be enforced without flexibility, even when such flexibility would enhance care and reduce risk. At the structural level, blanket policies that prohibit outside food reflect broader norms that prioritize institutional control over cultural responsiveness (55). These norms often trace back to colonial legacies and assumptions about what counts as legitimate care (56).

From this perspective, risk is no longer an individual attribute, but a product of how systems distribute or withhold care, safety, and opportunity. This reframing shifts forensic mental health from managing “risky individuals” to interrogating and transforming “risky systems.” Any serious attempt to reduce risk must therefore account for system-level vulnerabilities and capacities: how care is coordinated, who is excluded, what assumptions underpin service design, and whether systems can learn and adapt under pressure.

This paper introduced the Risk in Systems Framework to analyze how institutional rules, policies, discourses, and resource flows produce or exacerbate risk. By addressing structural and systemic conditions, the framework redirects attention from individual deficits to the environments that sustain them.

As a conceptual model, the framework now requires empirical testing. Advancing it will involve operationalizing system- and structure-level risk, examining their effects on care pathways and outcomes, and comparing how these dynamics unfold across jurisdictions. This work calls for coordinated international collaborations and access to diverse forensic, social, and policy environments.

Expanding this research agenda also means drawing on disciplines not grounded in individual-level intervention. Public health, urban planning, geography, architecture, and public policy offer conceptual and methodological tools for analyzing how systems distribute care, safety, and opportunity. Architecture and design can illuminate how built environments, particularly in institutional settings, can affect risk of violence and aggression (57, 58). Geography helps explain spatial marginalization (59) while urban planning shows how neighborhood features such as access to green space and opportunities for social connection influence mental health and risk of violence (60, 61). Engineering may help develop technologies that support service delivery (62), whereas public policy can assess the impact of legislation or funding structures on care pathways and outcomes (38, 63). Finally, health economics and health services research can help evaluate the effectiveness, efficiency, and equity of different models of care. While there is growing interest in rethinking models of care in forensic mental health (33, 34), the current knowledge base remains largely descriptive, with little empirical data – especially comparative or cross-jurisdictional (3, 64, 65) – available to guide reform (66). Together, these fields can help expand the forensic mental health research agenda beyond individual-level treatment toward systemic and structural transformation.

However, broadening disciplinary lenses is not enough. Addressing structural risk requires questioning the hierarchies of knowledge embedded within Western academic traditions (53, 67). Dominant models of evidence – assessment tools, randomized controlled trials – often exclude Indigenous and relational knowledge as well as lived experience (56). Upholding Western post-positivist ideals as gold standard for informing policy and practice invalidates alternative ways of knowing and narrows what counts as meaningful data, reliable knowledge, or effective intervention (68–70). In the same vein, forensic mental health research has only minimally engaged with participatory or community-partnered methodologies (71), and even these efforts are often confined to tokenistic consultation. As a result, priorities defined by service users, their chosen support networks, and their communities are often dismissed as anecdotal, unscientific or a “nice to have” (70, 71). Without deeper epistemological shifts – both in what is considered valid knowledge and who are considered legitimate knowers – the field cannot meaningfully study or intervene on the structural risk produced by our societies’ dominant values and ideologies.

Forensic mental health research and systems have reached a tipping point (66, 72). The field cannot continue to focus almost exclusively on the individual and expect meaningful breakthroughs. As Jamie Livingston notes, “striking a balance between doing research that benefits risk assessment/management and public safety while not exacerbating stereotypes about, and stigma for, people with mental health and substance use issues, including those navigating the forensic mental health system, is a principal challenge in the forensic mental health scholarly space” (personal communication, 2025). Recognizing risk as a property of systems rather than individuals shifts responsibility toward the structures that generate vulnerability. This opens the possibility of forensic mental health systems grounded in dignity and shared responsibility.