Assessing Risk Factors for Victims of Violence in a Hospital-Based Violence Intervention Program

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for giving me the opportunity to review this paper.  I found that this was very well written and the paper as a whole had a strong, coherent and consistent argument.  The application of the RNR principles seems novel in the current context and the assessment tool provides a flexible yet systematic approach.  I agreed with the authors that this would be of interest to practitioners and would be very useful in a medical context and I felt that the work progresses with a strong methodology (despite only using descriptive statistics and some tests of association). However, I have some areas which I feel the authors need to address.  Although there are four main things, each of these is relatively minor and I believe the authors could address these relatively straightforwardly. 

1) First, to a certain extent,the medical utility is a potential challenge for the paper as a whole.  Is the nature of the contribution so niche that in fact it lacks wider relevance for the readers of this journal and therefore is in fact less of a contribution than the authors are claiming?  I feel this needs to be established more strongly. For me this would not take much additional writing - instead this would potentially be a reframing of parts of the argument.   While it is up to the authors how they do this, I can see two potential ways to achieve the objective and propose these as suggestions.  The authors could adopt one or both of these or address the importance of establishing a wider/stronger contribution in a different way.  My suggestions are:

a) Expanding the focus in the discussion to consider whether this tool and the application of the RNR principles could apply to other non-forensic but risk related settings (e.g. education, child protection/safeguarding of vulnerable adults, social work). 

and/or b) Establishing a stronger case that the RNR approach adds value alongside existing methods of screening in accident and emergency departments or health/mental health contexts where those experiencing trauma may be seen.  I would suggest this article to get started on that direction. 

Raja, S., Rabinowitz, E.P. and Gray, M.J. (2021).  Universal screening and trauma informed care: Current concerns and future directions.  Families, Systems and Health, 39, (3), 526-534.  Doi: 10.1037/fsh0000585

 

2) My next point for strengthening the work is how do these findings translate to contexts other than the USA? I can potentially see wider applicability but the tone and context of the article is very US-centric. In fact I would go as far as to say that the article is written in a way that seems unaware of a world beyond US borders.  I am sure in reality that this is not true and therefore feel that the authors need to address whether and how these ideas and findings have this wider relevance.

 

3) The third thing which I think the authors need to tighten up on is the supporting evidence for the claims made in the introduction.  While some sections are very well supported, page 3 contains a lot of unsupported claims and pages 6&7 are very reliant on one source to establish and justify the use of the VRRAI.  There is also allusion made to this being one of a suite of possible tools (maybe related to the point I made earlier about how to strengthen the contribution beyond a niche) but limited discussion and citation is given to these. If one of a possible set of tools, why this one ahead of the others and how does this relate to contribution?  I would suggest clarifying this section.

 

4) Finally I had a minor suggestion in terms of presentation but one which the authors may need to think about before deciding how to change.  My background with the RNR is in its original application to forensic psychology risk assessment and I know that static risk absolutely does not indicate low or no priority.  Therefore I was really, really surprised to see this implied in Table 2.  For example, adverse childhood experiences (a static risk factor) are not given low/no priority in the RNR (nor are they given low/no priority in a trauma informed approach...or the crossover between the two - see e.g. Fritzon et al., 2021 - reference below).  I can however appreciate that the context is different here and I can see how in a medical context this might be the case.  However I think the authors need to be much clearer about why they have equated static risk factor with low/no priority. 

 

 

Fritzon, K., Miller, S., Bargh, D., Hollows, K., Osborne, A. and Howlett, A.  (2021).  Understanding the relationships between trauma and criminogenic risk using the Risk-Need-Responsivity Model.   Journal of Aggression, Maltreatment and Trauma, 30, (3),  294-323.  Doi: 10.1080/10926771.2020.1806972

 

In addition to the four things highlighted above, I had two further suggestions which I felt might enhance this or future work (but which are more optional in nature). First, most applications of the RNR in current practice would also include a consideration of strengths based approaches and I noted that the authors did not address protective factors, resilience or character strengths.  A consideration of this would enhance the applicability and sophistication of their approach.  Second, as mentioned above, the authors focussed on descriptive statistics and some simple tests of association rather than considering things like odds or risk ratios.  It may be that the authors do not have access to levels/degrees of exposure because of the binary nature of their data collection.  And a binary coding process would certainly justify the statistics used.  However for future work, being able to calculate this in terms of odds or risk ratios would significantly improve the communication of risk to HVIP practitioners.  This is something for the authors to consider as they progress the work further. 

Author Response

Thank you for giving me the opportunity to review this paper.  I found that this was very well written and the paper as a whole had a strong, coherent and consistent argument.  The application of the RNR principles seems novel in the current context and the assessment tool provides a flexible yet systematic approach.  I agreed with the authors that this would be of interest to practitioners and would be very useful in a medical context and I felt that the work progresses with a strong methodology (despite only using descriptive statistics and some tests of association). However, I have some areas which I feel the authors need to address.  Although there are four main things, each of these is relatively minor and I believe the authors could address these relatively straightforwardly. 

1) First, to a certain extent,the medical utility is a potential challenge for the paper as a whole.  Is the nature of the contribution so niche that in fact it lacks wider relevance for the readers of this journal and therefore is in fact less of a contribution than the authors are claiming?  I feel this needs to be established more strongly. For me this would not take much additional writing - instead this would potentially be a reframing of parts of the argument.   While it is up to the authors how they do this, I can see two potential ways to achieve the objective and propose these as suggestions.  The authors could adopt one or both of these or address the importance of establishing a wider/stronger contribution in a different way.  My suggestions are:

1. a) Expanding the focus in the discussion to consider whether this tool and the application of the RNR principles could apply to other non-forensic but risk related settings (e.g. education, child protection/safeguarding of vulnerable adults, social work). 

and/or b) Establishing a stronger case that the RNR approach adds value alongside existing methods of screening in accident and emergency departments or health/mental health contexts where those experiencing trauma may be seen.  I would suggest this article to get started on that direction. 

Raja, S., Rabinowitz, E.P. and Gray, M.J. (2021).  Universal screening and trauma informed care: Current concerns and future directions.  Families, Systems and Health, 39, (3), 526-534.  Doi: 10.1037/fsh0000585

 

The following section was added to section 1.5 Fundamentals of Risk, Needs and Responsivity in Risk Assessment to strengthen the case that the RNR model adds value to existing risk screening methods, specifically within the context of hospital-based violence intervention programming.

 

The principles of the RNR model bolster traditional risk screening in medical settings, which tends to focus on the presence or absence of risk factors, by considering responsivity. Systemic responsivity, meaning the ability of systems to adequately respond, is especially critical to HVIPs that rely on medical and community systems to address the needs of participants and should consider the capacity and ability of these systems to adequately meet the needs of program participants (Taxman, 2017; Taxman, 2019).

 

2) My next point for strengthening the work is how do these findings translate to contexts other than the USA? I can potentially see wider applicability but the tone and context of the article is very US-centric. In fact I would go as far as to say that the article is written in a way that seems unaware of a world beyond US borders.  I am sure in reality that this is not true and therefore feel that the authors need to address whether and how these ideas and findings have this wider relevance.

 

The following section was added to the discussion:

Firearm violence is not solely an American health issue. Many countries face the same static risk factors identified in the American population; globally, men are more likely to face firearm related mortality than women, with rates often highest among 20-24 years old (Naghavi et al., 2018). The RNR principles can be further used to develop risk assessment tools that consider the dynamic risk factors specific to different cultural contexts.

 

3) The third thing which I think the authors need to tighten up on is the supporting evidence for the claims made in the introduction.  While some sections are very well supported, page 3 contains a lot of unsupported claims and pages 6&7 are very reliant on one source to establish and justify the use of the VRRAI.  There is also allusion made to this being one of a suite of possible tools (maybe related to the point I made earlier about how to strengthen the contribution beyond a niche) but limited discussion and citation is given to these. If one of a possible set of tools, why this one ahead of the others and how does this relate to contribution?  I would suggest clarifying this section.

 

We have revisited the front end of the paper to include notation of sources throughout key area. Further, we are only aware of the VVRAI as a tool being used by HVIP rather than a suite of tools. We have clarified our language in this regard.

 

4) Finally I had a minor suggestion in terms of presentation but one which the authors may need to think about before deciding how to change.  My background with the RNR is in its original application to forensic psychology risk assessment and I know that static risk absolutely does not indicate low or no priority.  Therefore I was really, really surprised to see this implied in Table 2.  For example, adverse childhood experiences (a static risk factor) are not given low/no priority in the RNR (nor are they given low/no priority in a trauma informed approach...or the crossover between the two - see e.g. Fritzon et al., 2021 - reference below).  I can however appreciate that the context is different here and I can see how in a medical context this might be the case.  However I think the authors need to be much clearer about why they have equated static risk factor with low/no priority. 

 

The reviewer makes a great point about how our prioritization could be interpreted differently or implied from our intention of indicating that HVIPs may not be equipped to address some factors in their shorter term responses or due to limitation of resources. We have concluded that these categorical labels are unnecessary and excluded from the text. We appreciate the perspective provided by the reviewer!

 

Fritzon, K., Miller, S., Bargh, D., Hollows, K., Osborne, A. and Howlett, A.  (2021).  Understanding the relationships between trauma and criminogenic risk using the Risk-Need-Responsivity Model.   Journal of Aggression, Maltreatment and Trauma, 30, (3),  294-323.  Doi: 10.1080/10926771.2020.1806972

 

In addition to the four things highlighted above, I had two further suggestions which I felt might enhance this or future work (but which are more optional in nature). First, most applications of the RNR in current practice would also include a consideration of strengths based approaches and I noted that the authors did not address protective factors, resilience or character strengths.  A consideration of this would enhance the applicability and sophistication of their approach.  Second, as mentioned above, the authors focussed on descriptive statistics and some simple tests of association rather than considering things like odds or risk ratios.  It may be that the authors do not have access to levels/degrees of exposure because of the binary nature of their data collection.  And a binary coding process would certainly justify the statistics used.  However for future work, being able to calculate this in terms of odds or risk ratios would significantly improve the communication of risk to HVIP practitioners.  This is something for the authors to consider as they progress the work further. 

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review this interesting manuscript. The aim is to use a risk measure for violence reinjury to identify risky individuals and facilitate appropriate intervention resources for them in a hospital-based violence intervention program in Nebraska. The victim-perpetrator nexus is an important issue, however it is one that needs to be addressed with some sensitivity. Overall, the paper is well-written and presents a case for a promising approach to violence risk prediction, but is not without issues. 

INTRO: This section is concise with the purpose clearly defined. The rationale is provided by the authors and pertinent literature is reviewed. Despite literature support, I am uneasy about the central assumption of victims of violence being framed as ticking timebombs for further violence rather than as survivors of trauma in the first instance. The more obvious imperative for me isn't about increasing marginalisation of survivors, but rather, understanding the impact of the wider cultural tolerances of weapon-readiness on the perpetration of violence in this State. You tend not to see these kinds of services in other developed countries. RECOMMEND: A statement (1) justifying framing people as a source of danger rather than a trauma-informed view of survivor experience, and (2) describing the cultural millieu and the role of weapons in the relevant community.

METHODS: The approach used is described in a succinct and clear way allowing for replication. I may have missed this, but what was the criteria for admission to the program? Did participants opt-in or were they directed to attend? If so, by whom (e.g., physician, social worker) and under what conditions? Also, why these participants? It is not unusual for victims of violent trauma to harbour a distrust of social institutions, not least because they may have not provided safety and protection for service users when needed. Indeed, it is possible that the 'risky' people may not have attended the intervention if left to their own volition. In addition, what is the theory behind this measure? The design of a measure on criminogenic (RNR) principles is based on different assumptions (i.e., recidivism risk) than those that address trauma (i.e., health and well being). Further, a factorial approach to measure design misses context and experience by privileging correlations. Also, criminogenic logic concerns resource allocation rather than human-centred values or community safety. RECOMMEND: (1) Clarity around participant involvement and selection and (2) A more critical discussion of the adopted perspective and approach taken. 

RESULTS: Overall, these are clearly explained and adequately summarised. My above-noted issues notwithstanding, the Tables/findings provide interesting insights about the group and what variables have the strongest grouping.

DISCUSSION/CONCLUSION: Main findings are synthesised and efforts are made to describe applications into intervention design and resourcing. This section is particularly thoughtful in this regard. The limitations of a lack of case management data is on point. Although I am wary of a study that targets individual-level variables without being informed by community or cultural context to make sense of social issues such as violence and the threat of future harm, the conclusions are in keeping with the body of the report. 

GENERAL: Tables to be formatted appropriately (presumably APA 7e... Table 1 needs a caption). Consider geographic relevance in the title ('...Hospital-based violence intervention program in Nebraska'). This gets around readers thinking that this is a culturally-universal study and also highlights the region as a focus of important research developments.          

Author Response

Thank you for the opportunity to review this interesting manuscript. The aim is to use a risk measure for violence reinjury to identify risky individuals and facilitate appropriate intervention resources for them in a hospital-based violence intervention program in Nebraska. The victim-perpetrator nexus is an important issue, however it is one that needs to be addressed with some sensitivity. Overall, the paper is well-written and presents a case for a promising approach to violence risk prediction, but is not without issues. 

INTRO: This section is concise with the purpose clearly defined. The rationale is provided by the authors and pertinent literature is reviewed. Despite literature support, I am uneasy about the central assumption of victims of violence being framed as ticking timebombs for further violence rather than as survivors of trauma in the first instance. The more obvious imperative for me isn't about increasing marginalisation of survivors, but rather, understanding the impact of the wider cultural tolerances of weapon-readiness on the perpetration of violence in this State. You tend not to see these kinds of services in other developed countries. RECOMMEND: A statement (1) justifying framing people as a source of danger rather than a trauma-informed view of survivor experience, and (2) describing the cultural millieu and the role of weapons in the relevant community.

It was not the authors intention to describe the program participants, or any other survivors of firearm injury, as potential dangers. The manuscript has been revised to further clarify the population as at risk for a subsequent violent injury and that efforts must be made to reduce that risk. Additionally, the sentence “Community violence is complex, often the subset of individuals engaging in violent behavior are exposed to violence through direct or indirect victimization (Spano, Pridemore & Bolland, 2012).” To clarify that although there is overlap between victim and perpetrator of violent injury, exposure to violence as a direct victim or through community exposure often predates the perpetration.

METHODS: The approach used is described in a succinct and clear way allowing for replication. I may have missed this, but what was the criteria for admission to the program? Did participants opt-in or were they directed to attend? If so, by whom (e.g., physician, social worker) and under what conditions? Also, why these participants? It is not unusual for victims of violent trauma to harbour a distrust of social institutions, not least because they may have not provided safety and protection for service users when needed. Indeed, it is possible that the 'risky' people may not have attended the intervention if left to their own volition.In addition, what is the theory behind this measure? The design of a measure on criminogenic (RNR) principles is based on different assumptions (i.e., recidivism risk) than those that address trauma (i.e., health and well being). Further, a factorial approach to measure design misses context and experience by privileging correlations. Also, criminogenic logic concerns resource allocation rather than human-centred values or community safety. RECOMMEND: (1) Clarity around participant involvement and selection and (2) A more critical discussion of the adopted perspective and approach taken. 

The following section (2.1) was further clarified.

Participants: Patients who were victims of interpersonal violence (firearm injuries, stabbing, physical assault) admitted to the trauma center and voluntarily enrolled in a hospital-based violence intervention program after meeting with a Violence Intervention Specialist.

We re-emphasize the victim-offender overlap and our focus on community gun violence that HVIPs are intended to address to clarify our perspective as well as added reference to related work (Armstrong & Griffin, 2007) indicating that among high risk individuals, similar sociological factors are related to both victimization and offending.

RESULTS: Overall, these are clearly explained and adequately summarised. My above-noted issues notwithstanding, the Tables/findings provide interesting insights about the group and what variables have the strongest grouping.

DISCUSSION/CONCLUSION: Main findings are synthesised and efforts are made to describe applications into intervention design and resourcing. This section is particularly thoughtful in this regard. The limitations of a lack of case management data is on point. Although I am wary of a study that targets individual-level variables without being informed by community or cultural context to make sense of social issues such as violence and the threat of future harm, the conclusions are in keeping with the body of the report. 

GENERAL: Tables to be formatted appropriately (presumably APA 7e... Table 1 needs a caption). Consider geographic relevance in the title ('...Hospital-based violence intervention program in Nebraska'). This gets around readers thinking that this is a culturally-universal study and also highlights the region as a focus of important research developments.    

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you to the authors for clarifying the key points requested in the review. I am happy that my comments have been addressed.