Informing Correctional Officer Discretion: A Co-Response Model and the Legal Vulnerabilities Inherent in Prison Work

I draw data from an ethnographic experience of participating in correctional officer training at the Correctional Service of Canada (CSC) to explore the position of prisoner health in informing correctional officer discretion. I unpack how through training CSC holds recruits accountable for their actions, reactions, and discretionary behaviors, while also structuring recruit decision-making by enforcing a model that promotes a co-response between health care and security actors in prison. I speak to correctional officer legal vulnerabilities, the value of documentation as a means to rationalize actions, and make recommendations for future research, policy, and training practices.


Introduction
Correctional officers (COs) are legally and institutionally held accountable for their actions, which are informed by discretion, when following policy and when acting outside of the parameters of policy. In Canada, examples dispatch police alongside mental health professionals when calls for services from persons with mental health needs occur (Borum et al., 1998;Deane et al., 1999;Finn & Sullivan, 1989;Zealberg et al., 1992). In Memphis, Tennessee, the Crisis Intervention Team (CIT) was developed, aiming to provide specialized training related to responding to mental health calls for police Compton & Chien, 2008). In Canada, police services, like those in the U.S., the U.K., and Australia, have progressed toward forming co-response teams that combine health care and security responses (Baess, 2005;Cotton & Coleman, 2010;Koziarski, 2018;Morgan & Paterson, 2019).
Although programs vary across Canada, co-response practices and policies are positioned as more ethical and humane approaches to reducing the criminalization of mental health disorders (Bailey et al., 2018;Puntis et al., 2018). The approaches also acknowledge that police and mental health professionals, including physicians, psychiatrists, social workers, and nurses, need to work together to optimize responses. Moreover, researchers find co-response approaches reduce arrests and hospitalizations of persons with mental health needs and helps connect persons with essential health and social services (Baess, 2005;Helfgott et al., 2016;Lamb et al., 1995;Puntis et al., 2018;Shapiro et al., 2015;Steadman et al., 2000). Other researchers have criticized the co-response approach for being a reactive rather than proactive solution and for being short sighted (e.g., failing to account for the effectiveness of the cross-system service) (Coleman & Cotton, 2016;Puntis et al., 2018;Shapiro et al., 2015;Wood & Watson, 2017). Co-response models have been neither document nor studied in the context of correctional services in Canada. However, similar to the essence of a co-response model, CSC introduced the EIM, which promotes, even instructs, a co-response approach to responding to prisoners with health needs within the institution.

The Engagement and Intervention Model (EIM)
Correctional Services Canada transitioned from the Situation Management Model (SMM) to the EIM. The SMM was a model of incident management that relied rather heavily on "use of force" designed nearly 20 years ago. More recently, the EIM was developed in response to the limitations with the SMM including in the options for how situations with prisoners and criminalized persons were managed both in institutions and in the community (Correctional Service Canada, 2018, p. 3). Specifically, the Special Report by the Office of the Correctional Investigator, entitled Fatal Response: An Investigation into the Preventable Death of Mathew Ryan Hines which was tabled May 2, 2017 in parliament, likely had a part in informing the model change, and thus this new approach to situations within correctional services.
The EIM model is in many ways a framework for assessing risk and actions. The model considers "both Security and Health (physical and mental) components, in balance, when assessing a person and situation" (Correctional Service Canada, 2018, p. 4). Central to the EIM model (see Image 1), is staff using non-physical and de-escalation responses when responding to criminalized persons, including those undergoing physical or mental health challenges. Thus, the model intends to add more options than those (e.g., use of force) provided in the SMM when employees are responding to a situation. The EIM model, in theory, requires staff to think critically and to include a healthcare response when necessary-it is a correctional variant of a policing co-response model. The model first mandates staff "to assess the situation", before "select[ing] the engagement and intervention strategy", and then to "engage, intervene and reassess" as the situation evolves (p. 42). The selection of the response is particularly invaluable. The EIM splits security and health responses, each representing part of the full circle (see Image 1), which signals to COs that they are to call on healthcare staff if the situation presents as health-versus only security-related. The purpose of the model, according to CSC, is to "assist front-line staff in recognizing, responding and addressing situations of medical emergency and/or acute mental health distress" (p. 12), while creating a multidisciplinary team response to situations that includes both health care staff and security staff. As such, the model encourages staff to consider all options and supports that, in security situations, COs increase their use of force when increased risk is posed to the prisoner, staff, or public (e.g., low risk requires verbal intervention, moderate risk includes controlled non-intervention or restraints, and high risk can require batons or firearms). On the literal flip slide of the model, as health risk increases either for the prisoner, staff or public, the severity of the healthcare intervention intensifies (e.g., low risk calls for staff presence and verbal intervention, moderate risk may call for medication and restraints, and high risk demands a transfer to the community hospital and the calling of paramedics). Often, there is need for a mixed response, incorporating both healthcare and security staff to work together to overcome the security and health risks.
The model is person focused, placing the criminalized person in the situation at the center, as such it prioritizes relationships and information sharingcreating awareness of factors that contribute to or may compromise the environment in which the situation is occurring. The model is also riskbased, but it includes the risk posed to the prisoner as well as the risk posed to the safety and security of staff, the institution, and the public.
It counters prior models in that it includes the risk posed to the prisonerthus creating a space for the reconceptualization of risk as tied to legal vulnerabilities and officer discretion.

Current Study
In the current study, I examine how training informs discretion, and, ultimately, decision-making and show how discretion underpins the legal vulnerabilities potentially experienced by COs during an ethnographic experience of the three stages of correctional training program (CTP). I interpret discretion as learned from all sources within the ethnography-the data from which the study is drawn-ranging from informal processes of socialization to the training manuals I studied and analyzed. Always within the framework of discretion, I explore the co-response model employed by CSC and how CSC embeds prisoner health awareness into CO training through the EIM. The EIM is central or informs arguably all aspects of COR training. Finally, I study the legal vulnerabilities and need for documentation tied to the CO occupation, while illustrating that COR training includes more than use of force, the training also promotes practices of dynamic security that prioritize communication skills.

Method
I engaged in an ethnographic study across three stages of the CTP, stage I and II occurring online prior to arriving for stage III, which is the 14-week in person component of the training. During stage III, I lived in a dorm alongside those in my cohort, engaging in all components of the training, including the testing on specific skills and the decision-based training (DBT) scenarios that are enacted (including in mock prison units) at the academy. More information on the details of the ethnographic study are published elsewhere (|Ricciardelli, 2021(|Ricciardelli, , 2022; however, I provide a brief overview of the study processes here to detail how I was "living with and living like those who are studied" (Van Maanen, 1988, p. 2). 2 Central to the current study, is showing how COR training informs their discretion and thus decision-making, prepares CORs for legal vulnerabilities, and promotes awareness of the healthcare needs of those in custody, while prioritizing dynamic security based on communication skills over the use of force. With these objectives in mind, I participated in CTP, openly and without deception-trainers and CORs knew I was a professor, but I was treated as a recruit (e.g., I was reprimanded, earned a 'strike', yelled at). I did have two exceptions to the academy rules, the first is I was able to have my phone on my person during training and I was able to enter the trainer office spaces at the academy, which is generally a restricted space for recruits. These exceptions, to the best of my knowledge, did not affect my training experience or that of my fellow cohort of 24 recruits. I tried to always maintain the lens of a researcher while experiencing training; including during the in-person component which included fitness logging, instructional hours running from 8am to 4pm Monday through Friday, and hours of 'homework', studying for testing, and practicing learned skills. For the most part, the physical skills did not come naturally to me, but I enjoyed learning the diverse techniques employed in CSC and the nuances of the diverse posts COs will hold when working in any institution.
At the Academy, recruits start their day marching (i.e., parade) to class together, eating upwards of three meals a day together (breakfast before class, lunch between classes, and a really early dinner), training together, practicing together, and being reliant on each other for studying and, in some cases, testing. I experienced decision-based training scenarios, was pepper sprayed, participated in learning techniques of de-escalation, shot firearms, and engaged in all components of classroom learning. Maintaining a researcher lens that also recognizes the emotional work that is training, I remained aware that everyone at the academy, from the trainers to the recruits to management, construct together the dynamics and orientation of the training experience (Charmaz, 2006). I arrived for stage III at the National Training Academy in June 2019 with my particular cohort graduating and earning their badges in September 2019. The graduation ceremony was held in the gym at the now closed Kingston Penitentiary in Ontario, representing the successful complete of the training and the transition from a COR to CO ready for institutional deployment. I experienced the emotional ups and downs of the CTP experience, where I was both an active participant and a direct observer (Snow et al., 1986). I was "hanging out" both in a professional sense as a recruit and after classes and meals with my cohort in spaces affiliated (e.g., dorms and houses) and not affiliated (e.g., restaurants, shopping facilities) with the academy (Browne & McBride, 2015). My field notes included interpretations of lessons, interactions, emotions, and observations. I also created what I refer to as 'training notes', in which I documented training content, study notes, and descriptions of study experiences. I reviewed the notes, analyzed the content, seeking themes across lessons and practices as well as addressing apparent inconsistencies. My ethnographic experience is also informed by a simultaneous longitudinal study that includes interviews conducted with 354 CORs (as of December 2020) and interviews with trainers across the training academies. At the time of the ethnography, there were two academies, the National Training Academy in Ontario and a satellite academy in Prince Edward Island, due to COVID-19 additional satellite training academies are now operational. COVID-19 impacted the study in that I have yet to complete the primary worker training (e.g., for those working in women's institutions) or participate in CTP stage III with other cohorts and lead trainers as I had intended.

Results
During recruit training, informing officer decision making and discretion, at least in my experience, is the EIM, which I demonstrate is, in its essence, a co-response model (Bailey et al., 2018;Puntis et al., 2018). Moreover, at CTP, recruits learn that they must assess and understand their environment and that whenever risk escalates in the prison environment (from low to moderate to high risk situations), they are to modify their response, re-access, and intervene accordingly. Thus, recruits are to follow the EIM and explain the reasoning behind their actions or the intervention strategy selected. Counter to prior researcher findings suggesting that use of force dominants in training (Klassen, 2018), I found that although use of force is a notable part of the training, the emphasis in training is "dynamic security", communication, and de-escalation of situations, as well as learning when to use a co-response model that includes healthcare and security. Particularly in use of force situations, recruit training informs officer decision-making and provides a framework for documenting and explaining, even justifying, actions. Within this context, I reveal the learning processes informing officer decision-making in responding to calls for intervention or service within the institution, the role of discretion as informed by training, and the emphasis on legal vulnerabilities that underpin training practices.

Communication Versus Force
Training emphasizes de-escalation over the use of any force, even when learning how force is used. In learning firearms, for example, the emphasis is not on use of force. Instead, the emphasis is safety (e.g., five and seven point safety checks, firearm control, safe direction, finger off the trigger), the components of the different firearms (e.g., including comprehensive inspections of dummy and active rounds and fire arms), and how to ensure one is not left vulnerable (e.g., that the prisoner does not gain control of the firearm during an escort (see: CBC News, 2018)). The instructors explain repeatedly that "99 percent of the time," de-escalation will prevail through dynamic security. Nevertheless, CORs train on diverse firearms, work to create muscle memory, and test on firearm theory, safety, and shooting accuracy within timeframes. CORs spend much time learning how to manage common stoppages, malfunctions, and immediate actions, but the emphasis is always safety, not use of force despite learning actions that underpin use of force.
Discussions centered on justifications for use of force, which is one of many options. For instance, CORs learn that verbal warnings, alone or combined with the threat of force, will often deter the need to use force by de-escalating. Trainers emphasize ensuring clear warnings were given (e.g., "Inmate, drop the knife or I will shoot"), so there could be no confusion about what the consequence would be if the prisoner did not adhere.
Even beyond firearms, direct orders, based on communication skills, precede any potential use of force. For instance, CORs learn that if first officer on scene during an incident, they are to issue a direct order using communication skills. The order may or may not have a direct consequence, but if there is a consequence, the consequence must be specific and clear (e.g., "pass the contraband or we will search the cell"). CORs are to express their actions with clarity, using verbal communication as a means to de-escalation when possible before progressing to any use of force (e.g., arrest and control, OC spray). In discussing and learning how to use force, the underlying message centers on dynamic security and modes of deescalating, rather than physical force. The movement to provide more alternative to the use of force in CSC is also evident in the EIM and co-response with health care to mental (and physical) health crises.

The EIM as a Co-Response Model
CSC engrains in recruits, through their training, the EIM. CSC introduces the EIM in stage I, prior to recruits ever arriving at the Academy, in a comprehensive and explanatory manner through the online training portal. Specifically, recruits learn the nuance of the model, which "is surrounded by an outside ring that denotes the principles through which all decisions in CSC are made, within our legal and policy framework: Leadership, preservation of life, Interdisciplinary Teamwork, CSC Mission and Values, and Necessary and Proportionate" (Correctional Service Canada, 2018, p. 8). Here, modules present the co-response structure of the model. Recruits learn that when responding to a prisoner in a health crisis, they are to include and work collaboratively with healthcare in their response, rather than perform a security response in isolation.
Next, in stage II, CORs complete a 'homework' assignment that involves the application of the EIM (Correctional Service Canada, 2018). The specific assignment asks recruits to review the "Introduction to the EIM" from CTP Stage I, with an emphasis on situational assessment and risk evaluation, before responding to the question: "Why is it essential for a Correctional Officer/Primary Worker (CO/PW) to be highly skilled in assessing a situation and evaluating the level of risk?" 3 The co-response structure of the model informs the assignment, as risk evaluation must be contextualized within the framework of health and security, with recruits learning to employ the least aggressive (e.g., 'necessary and proportionate') means possible to deescalate a potential situation and to work in partnership with institutional healthcare when appropriate.
In Stage III, when taking up the Stage II assignment in class, the optimal responses refer to recruits being able to maintain self-control when in trying circumstances and to determine if any hostility, fear, or aggression is at play through assessing kinesics, proxemics, and para-verbal communication.
Maintaining composure is key to correctional work, COs always perform composure in front of an audience of prisoners, staff, management, and on camera. Recruits learn to assess for prisoner level of compliance, if in an altered level of consciousness (i.e., using substances), as well as any and all situational factors (e.g., time, available staff for support, level of containment, self-injurious or suicide-related behaviors). Here, recruits are to engage healthcare if their assessment includes prisoners undergoing health-related challenges (e.g., substance use, self-injury), thus centralizing healthcare and keeping prisoner health at the fore. Recruits are to assess and reassess situational factors as the incident progresses, working to determine and re-determine the ever-changing level of "risk" the situation produces, and an effective intervention strategy that is proportional to that demanded by the situation.
To evaluate risk, recruits are to use the EIM with the ultimate goal of reducing the level of risk, albeit physical, emotional, or social, while managing their own legal risk. Recruits, prior to arriving at the Academy, learn that they are to approach a situation using the EIM and optimize their actions toward decreasing risk, decreasing emotional responses (e.g., performing composure and not reacting visibly in terms of facial emotions), and increasing rationality. Here, recruits learn to be aware of their own interpretations of situations and to demonstrate confidence by responding with strategies that reflect and are proportional to the level of risk posed (low, moderate, or high).
In stage III, trainers teaching the EIM provide depth in how to apply, practically and theoretically, the model. The practical application of the model, trainers explain, includes a CO or primary worker onsite being assigned the role of the 'sector coordinator' who, remaining under the direction of the 'Correctional Manager', refrains from physical involvement in the situation.
The sector coordinator is responsible to "effectively oversee and coordinate activities" (p. 18), "consider the use of partners to de-escalate the incident" (p. 18), as well as monitor "the physical and mental health state of the offender(s) during the response" (p. 17)-again illuminating the co-response component of the model and highlighting prisoner health. The sector coordinator is to observe the situation and focus on the well-being of the parties involved as well as the threat posed to the security and safety of prisoners, staff, and the institution. Although the sector coordinator is not responsible for coordinating healthcare responses, their specific focus on the state of the criminalized person demonstrates the person-centered approach underpinning the EIM. The sector coordinator is to keep an eye on the involved persons and can encourage a change in approach if/as necessitated by the unfolding situation. Thus, within the EIM not all COs are active participants in the de-escalation of the situation, some are assigned specific roles and have directed responsibilities during the situations (e.g., sector coordinator, perhaps operating a camera).

Discretion and Decision Making
The application of the EIM is a central focus of training; intended to informing COR decision making and thus discretion. Early in stage III of CTP, the in-person on site component of the training, trainers conduct exercises in class where scenarios are read and each COR is to categorize the prisoner's behavior (e.g., cooperative, verbal resistance, passive resistance, active resistance, assaultive behavior). The COR next indicates where their response to the incident falls on the EIM (e.g., health or security, degree of support/force required). In this context, CORs use discretion informed by the EIM to determine their course of action toward situation resolution. Trainers and CORs discuss the diversity of responses to situations, illuminating the different ways a situation can be assessed and interpreted but reinforcing the centrality of the EIM as guiding principles for behaviors. The EIM is then used by each COR as they explain (or in some cases justify) why they engaged in a specific course of action (e.g., can the use of force be justified? How? Why? Can the inclusion or exclusion of healthcare be explained?).
Trainers also use video footage in class, repeatedly, to demonstrate situational assessments and EIM based actions and explanations. The video footage demonstrates the potential volatile situations that can arise during CO work; for instance, responding to between prisoner altercations, prisoner self-harming, and so on. Trainers play the videos and the class works through the situational assessments, using EIM to explain the actions of those involved. Particularly of note, remains the need to intervene with force if a prisoner is harming, either another prisoner, staff, or self.
Decision-based training (DBT) scenarios provide opportunities to apply the EIM to a situation in "real time". CORs are divided into smaller groups and enter into a "scenario" enacted by the trainers where they must demonstrate their use of EIM. Following DBTs are debriefing sessions and the writing of SORs. In training, including in DBTs, the norm among CORs becomes to assume risk because of the liability and occupational obligation of protecting self, prisoners, and society. Although the EIM informs the type of response necessary, the DBTs demonstrate to CORs what can go wrong with inappropriate situational assessments. For instance, failing to assume risk or the underestimation of risk can leave COs and those involved in the situation more vulnerable, susceptible to both additional harm because the situation was not resolved or assessed with success. Thus, the training primes CORs to understand that risk potentiality always requires recognition when making any situational assessment. Here, field notes reveal learning about a "correctional" versus "civilian" filter when determining risk during prison work. The idea here is that a "civilian filter" may produce a false sense of safety, whereas the correctional filter encourages the interpretation of observations at low, medium, versus high risk.
Of note, during training, it is not uncommon for CORs, who are primed to be entering risk situations, to look for suspicious behaviors where none exist -thus creating risk potentiality that otherwise would be null. In this sense, training does encourage CORs to be primed to observe risk when interpreting observations of prisoners, a factor that could inform officer discretion during DBTs. Each DBT is followed by the submission of a Statement/Observation Report (SOR) that speaks to the course of action employed and the thought processes underpinning their actions. The SOR provides a document trail outlining the incident and the officer's actions, but also has the latent function of enforcing accountability on the part of the responding officer(s) as will be discussed.
The fact that COs can become physically vulnerable during their occupational work is recognized during training as is the role of discretion in the promotion of safety. For instance, in learning how to manage situations like hostage taking and forcible confinement trainers teach CORs survival techniques to help mitigate threat to their well-being. CORs learn that each situation is unique, meaning discretion is essential which training tries to inform through emphasis on informing COR decision making, In hostage taking, for instance, CORs learn that they will need to work within the confines of the situation, always trying to maintain self-control, to stay relaxed, and to look for opportunities for escape. Lessons include the value of always answering honestly-tell the truth-if a question is posed by the hostage taker, to agree with their perspective, to let the prisoner know they are listening, to speak when spoken too, but to speak lower and slower and to ask if the prisoner's needs have been met (medication, food). They are to avoid arguing, gestures that may be perceived as threatening, and to refrain from escape attempts unless they are sure to be successful. CORs learn to be flexible in their response and how they act within the situation, thus discretion is essential. Trainers warn that if an officer does something, intentionally or not, that seems to elevate the prisoner's response they are to change their actions immediately. Moreover, they remind that truth telling is a way to build report and trust, explaining that the prisoner may ask simple questions to which they already know the answers to determine if the hostage is open to telling the truth. Thus, officer discretion is paramount in resolving or gaining control of the situation at play.

Legal Vulnerabilities
The legal vulnerabilities that follow COs as they perform their occupational responsibilities are reiterated consistently across training, and the messaging shows that making informed decisions that are justifiable is a key way to protect oneself, particular in light of the law, and to ensure officer discretion supports rather than antagonizes the situations that unfold. For instance, in the first weeks of training, the focus is on law and policy which includes when to use force for those in a position of authority, limits on the degree of force that may be used when exercising authority, and how the use of force is justified (i.e., to prevent escape) (Correctional Service Canada, 2018). Moreover, CORs learn that unreasonable and excessive uses of force by an officer is criminal-an aspect of the job that informs officer discretion and is embodied within the EIM, where justification and instruction for use of force is detailed.
Shedding light on legal vulnerabilities is that clear direction while issuing an order to a prisoner is always essential (e.g., "to your knees") as without it, if an officer acts but the prisoner was unclear on what is expected of they, on the stand an officer would not be able to defend their actions; they would be negligent. The use of force, all force (e.g., arrest and control, firearms) is always taught within the context of justification, proportionality, and to be documented through SORs as a central information-sharing process. Here, an underlying theme across all training is the need to be able to explain courses of action-a form of informed decision-making that rests on discretion, and articulate that explanation through information sharing via a SOR.
Field notes reveal ample discussion underpinned by the existence of legal vulnerabilities as are experienced by officers who fail to explain their actions, particularly when using force to deescalate a situation. Examples include rather simple comparisons of outcomes for two COs, who both "gas and cuff" (e.g., use oleoresin capsicum (pepper) spray and handcuffs) in response to a prisoner's behavior, whereas one explained their course of action and accompanying situational assessment process, and the other did not. The outcome is they who failed to explain may be held accountable or liable if a mishap or adverse outcome were to come from the use of force. However, if the actions are explained, informed decision-making demonstrated, and a situational assessment process that aligns with the EIM articulated, the CO reduces their potential vulnerability if a mishap were to occur.

Discussion
In the current article, I reflected on data from my ethnographic experience of the CTP to explore three components of training: the co-response model that is the EIM; the learning of discretion, particularly within the context of legal vulnerabilities; and the fact that training prioritizes dynamic security and communication skills over the use of force. Incorporating all facets of training in my analyses, I reflect on my experiences.
The CTP training, particularly in stage III, drives home to recruits the value of communication skills, de-escalation techniques, and the peacekeeping role of COs when working within an institution. For instance, central to using the EIM is to engage in situational assessment (to coordinate actions with health care when necessary) and always to act in ways that reaffirm the peace on the unit-to nullify the impact of a potential unfolding situation. As CORs experience during DBTs, how they assess and then approach a situation will affect the outcome. For instance, in some DBTs, CORs approached situations with a faulty risk assessment, either viewing the risk as too high or too low. Both situations result in poor outcomes, where the prisoner in question can endure unnecessary harm if the CO(s) under-assesses risk and situations can escalate unnecessarily if a risk is over-assessed. The question, remains, however, if approaching a situation with the assumption of higher risk (due to caution) escalates or deescalates the situation; securing versus compromising safety. In DBTs, CORs see that not intervening with adequate force or health care can lead to a prisoner succumbing to injury or experiencing more dire outcomes. Thus, informed and appropriate risk assessment that includes using the necessary force, always proportional to the situation, and ensuring processes for assessing force are compatible and understanding of the greater situation at play. Whenever possible, the EIM prioritizes verbal skills over force, ensuring officers are aware of the full range of possible behaviors that can be used to neutralize an arising challenge. The emphasis, then, is not use of force, but instead communication. A challenge arises in that effective communication does not come naturally to everyone, and, with this in mind, it would be advantageous to CSC to have more training designed to teach communication skills, including de-escalation, and to find ways COR communication skills can be accessed during training, thus helping to build each CORs' related toolbox of skills.
Given much of COR training is designed to teach COs how to assess and reassess situations, thus perform risk assessments, CORs learn to operationalize and strive to nullify risk using the EIM. The EIM is, in essence, a risk assessment co-response model, that informs how risk is managed by COs. Based on dynamic security practicesrelational practicesthe EIM prioritizes using an interdisciplinary team, with shared accountability between security and health staff, to work together to maintain the safety and security of incarcerated persons. Although a co-response model, the role of the EIM as a risk assessment tool is to outline courses of actions that correspond to situations of either a potential breach in security or a potential health crisis. The model is a tool for determining risk-the risk posed to the prisoner(s), staff, and the institution. However, how COs grow to understand risk during their training was outside of the scope of the current analyses, and requires additional interview-based research directly with CORs and COs to fully illuminate the nuances tied to risk in prison. Future research that explores how COs and CORs conceptual risk, that includes problematizing interpretations of risk, is necessary to better understand how models, such as the EIM, inform public safety work in practice. How CO's frame risk will inform how they assess risk in situations. Specifically, their orientations to risk, or risk potentiality, will inevitably affect interpretations of colleagues and, possibly, relationships with prisoners and other staff. To this end, greater insight into how risk is learned, understood, and acted upon by COs is necessary within a prison context, particularly the examination of how CORs apply the EIM once deployed to their institution, to shed light on both CO and prisoner safekeeping strategies and the parameters of institutional safety.
As much as the EIM is a risk assessment tool, the EIM is also the first documented co-response model used in Canadian federal correctional services. Co-response models are purported as more ethical and humane approaches to responding to the criminalization of mental health disorders (Bailey et al., 2018;Puntis et al., 2018). The EIM as well supports an interdisciplinary collaborative approach when responding to prisoners in diverse states of health crisis (e.g., mental and physical health crises). However, if the co-response approach results in prisoners obtaining greater access to health services and supports post incident, as researchers have found when using a co-response model in policing (Helfgott et al., 2016;Lamb et al., 1995;Puntis et al., 2018;Steadman et al., 2000), remains an area necessitating future inquiry. How COs working in prisons of diverse security classification use the EIM, if they do at all, in practice (outside of training), and how the EIMs informs officer discretion (and thus official decision making) are other areas for future inquiry.
The EIM training inadvertently keeps the health of the prisoner central when choosing a course of action in that the EIM incorporates healthcare into all decision-making, ensuring health, particularly prisoner health, is a point of assessment instead of only assessing security threat and risk of harm. In this way, the EIM is a progressive model, and although requiring further investigation into how exactly the EIM is affecting the management of situations in practice, the EIM may warrant application in provincial and territorial carceral facilities. Keeping prisoner health central to situational assessment ensures that persons in custody receive the healthcare attention necessary to manage the averse symptoms tied to compromised mental or physical health. Particularly in the case of prisoners experiencing mental health crises the EIM, in theory, encourages the collaborative response with healthcare, bringing in persons trained in mental health and working in health care, to help manage the averse symptoms presented while also supporting the healthcare staff with possible security (e.g., the COs perform security functions) if threat is detected. Thus, the model informs institutional safety, as well as the safety and health of the prisoner and staff.
Discretion, which ultimately leads to officially documented decisionmaking including in SORs, underpins the application of the EIM and, basically, all components of CO work. Recognizing COs are responsible for enforcing the rules and regulations governing prison living, and thus prison society, discretion is invaluable for peacekeeping and for the safety and security of the institution and all those living and working within. Discretion shapes the daily operation of prisons and relationships between prisoners and staff and between staff members. In prison, COs perform discretion and witness the discretion exercised by their colleagues, thus CORs need to be comfortable making decisions and exercising discretion with an audience of both staff and prisoners who will all in turn scrutinize their actions-particularly in cases of adverse outcomes. COs in using discretion must remain aware that their decision-making will impact both prisoners and colleagues; populations with whom officers must maintain longer term relationships. The need to continue working or supervising those involved in any situation may create additional pressures when discretion is used as the outcomes will have consequences well beyond the immediate situation at play. Perhaps in part for this reason, discretion is central to CO training, informed by the context (e.g., modules covered), lessons by trainers (e.g., informal and formal instruction), and the parameters of the occupation (e.g., law and policies). CORs learn discretion, recognizing that managerial scrutiny is to be anticipated in correctional work, particularly in instances where force is used or the prisoner's well-being is compromised through the process of resolving a situation. Training informs CO discretion, which in turns informs, at least in part, their official decision-making, that ultimately must be recorded in an SOR.
CORs learn to document their courses of action through the SOR. The SOR has multiple functions, acting as a statement of explanation of actions, an opportunity to defend one's action through the articulation of thought processes informing actions and thus decision-making, and as a means to hold an actor accountable for the outcome of their actions. COs are held accountable for their actions, in terms of their continued employability and within systems of justice. Although never referred to as 'legal vulnerabilities', the message that an officer could be investigated for their actions and even charged was evidenced in formal and informal socialization and curriculum during stage III of CTP. The significance here is that the precariousness of the CO is known but under-recognized or perhaps never formally documented. How this potential legal vulnerability impacts discretion and official decision making, particularly in use of force assessments and situations, is an area worthy of further investigation. For instance, when does an officer recognize their legal vulnerability? Does their legal vulnerability actually redirect their actions to appease legalities or are actions focused on optimal responses to incidents? Do officers recognize their vulnerability when in use of forces situations? Further, to inform training, research is necessary that examines how prisoners experience use of force, which could affect the way force is used. Prisoners voices and experiences are absent in informing training.
Recognizing that CO are legally vulnerable in their occupational work, it may be helpful to inform CORs during training about the investigation process, as there is much probability that COs at some point in their occupational tenure will be investigated or participate in an investigative process. Training that includes an introduction to investigative processes will further drive home, through formal socialization and learning, the legal vulnerabilities of the COoccupation and reinstate the importance of documentation and policy adherence for officers when performing their duties.
Escalating situations, however, are also always potentially psychologically traumatic event exposures, thus post incident, the well-being of those involved must always be prioritized to proactively minimize the experience of adverse psychological outcomes among staff and prisoners. The fact that correctional workers are exposrd to potentially psychologically traumatic events at rates that far exceed those among the general population (Carleton et al., 2019), as well as the correlation found between trauma exposure, correctional work, and mental health disorders (Carleton et al., 2018;Carleton et al., 2020;Ricciardelli et al., 2020), supports that the mental health and well-being of correctional staff be prioritized and addressed post incident.

Limitations
The current study is limited in that I did not include interviews with either CORs or COs to learn how the EIM and thus discretion is applied once in the institution. Moreover, I have yet to study how CORs are on boarded into their positions once working in a prison and how that ranges across different prisons, including those of diverse security classifications. Additional limitations include that I have only participated in one instance of CTP, led by two specific trainers, a limitation which is somewhat resolved in that modules are taught by different and varying trainers. Nevertheless, the need to reengage in CTP to see how the teachings may come across differently from lead trainers is warranted as well as participation in primary worker training to understand differences between how a CO working in men's versus women's institutions are trained.

Conclusion
COs are vulnerable in their occupation legally and trained to explain their actions to manage their legal vulnerabilities. CSC trains recruits in assessing and responding to a range of situations, including those that are escalating and violent, and teaches CORs to have a justification for their actions, decisionmaking, and thus discretion within the framework of their training, specifically the EIM. The training, perhaps unintentionally, recognizes how legally vulnerable COs are in their work environment, evidenced in that underpinning training is an intervention co-response model shaping COdiscretion thus decision making. Moreover, the training prepares COs in the administrative component of the job, which also serves to create a document trail outlining events and explaining courses of action as recognized in the moment. In essence, the same discretion that offers space for COs to protect themselves, prisoners, the institution, and ultimately society also makes COs vulnerable; exposing them to stigma, unfair media coverage, and situation in which they may even be brought to court.