A systematic review of evidence capturing efficacy of community and school-based approaches to knife crime intervention and prevention programs

The World Health Organisation (WHO) consider violence to be a public health concern (WHO, 2022). The WHO argued that a public health approach is the most effective form of understanding violence intervention and prevention due to the impact it has on health services and the criminal justice system as well as individuals and the wider community (WHO, 2022). This was supported by the UK Government in 2018 when it published their serious violence strategy in response to the increasing issue of violence and knife crime. Serious violent crime (SVC) refers to a category of criminal offences characterised by the perpetration of severe and harmful acts, such as rape or sexual assault, robbery and aggravated assault, which cause significant physical, psychological or emotional harm to the victims (Langton and Truman, 2014). The serious violence strategy argues for a public health and whole system approach to tackle SVC locally (Craston et al., 2020). It is thought that violence is not inevitable; it is a preventable issue that can be avoided with successful intervention strategies implemented within communities (Hassan, 2018). Knife crime is also a persistent issue in the USA as knives are the second most used weapons in homicides after firearms (Harms and Bush, 2022). Violence has an impact on the health services and the criminal justice system, as well as individuals and wider community (Public Health England, 2019). This approach considers the risk and protective factors underpinning violence, informing the understanding of the root causes and situational factors (Haylock et al., 2020). Violence can be reduced and prevented through measures and tailored programmes. This could inform policy changes across all levels; individual, community and society (Craston et al., 2020; WHO, 2022). To support local areas, the government invested funding in 18 violence reduction units across England and Wales as well as pledging to make £200m available over the next decade through the Youth Endowment Fund (YEF) to put interventions at the centre of tackling youth offending and violent crime (Bath, 2019). Further evaluation work, specifically impact evaluations, is required to consider the effectiveness of the intervention and preventative programmes, and whether their outcomes at the individual level are feeding into the bigger aspiration of ultimately reducing knife crime and SVC (Armstead et al., 2018).

Knife crime refers to any criminal activity involving the possession, use or threat of a knife or bladed instrument. This includes offences such as carrying a knife in public without a lawful reason, using a knife to commit assault or robbery and causing injury or harm with a knife. An offensive weapon is defined as any object designed or adapted for causing injury to a person or intended by the person carrying it to be used for such purpose (Cps.gov.uk, 2022). This can include knives, firearms and other sharp or bladed instruments, as well as items that have been modified or used in a threatening manner to cause harm. In the UK, carrying certain types of knives or offensive weapons in public places is prohibited by law, except in specific circumstances outlined by legislation (Cps.gov.uk, 2022). Knife crime and SVC have been on the rise and remains the leading cause of mortality in young people in the UK (Gavine, 2014). Possession and the use of knives, particularly in young individuals, has significantly increased and continues to be a social issue for society (Wortley and Hagell, 2021). Reports for 2020–2021 captured a recorded 41,000 offences involving knives across England and Wales: a 27% increase from 2010 to 2011 (Harding and Allen, 2021). The West Midlands police force alone recorded a 41% increase in knife crime from 2019 to 2020, with the highest rate of 156 offences per 100,000 population (Harding and Allen, 2021). In 2019–2020, 275 knife mortalities were recorded across England and Wales (ONS, 2020). Evidence suggests children as young as six years old are being groomed to carry knives for older individuals (Dodd, 2017).

The surge in violent crime among young people has led to discussions regarding the role of psychosocial maturity (PSM) in shaping adolescent engagement with violence. PSM encompasses three key dispositions – temperance, responsibility and perspective – and is influenced by cognitive competence. Research indicates that as adolescents develop these dispositions and cognitive competencies, they are less likely to engage in criminal and antisocial behaviour (Schaefer and Erickson, 2019; Schubert et al., 2016). However, the progression of ageing does not automatically equate to appropriate levels of PSM. Disruptions such as out-of-home placement during adolescence can negatively impact cognitive growth and, in turn, PSM. Police contact and arrests may further undermine adolescents’ PSM, as these interactions can limit their future opportunities and prospects (Schaefer and Erickson, 2019; Hipwell et al., 2018).

The concept of environmental influence is central to these discussions, as highlighted by the ecological systems theory (Ryan, 2001). This theory posits that the microsystem, comprising a child’s development, family and school, has the most direct impact on young individuals. It is within this immediate environment that they navigate challenges and opportunities. Additionally, trauma and adverse childhood experiences (ACEs) have been identified as factors that can profoundly affect young people’s mental health (Drury et al., 2017; Haylock et al., 2020). Poor mental health, in turn, has been linked to involvement in criminal activities (Drury et al., 2017; Haylock et al., 2020). Understanding these complex interplays is essential for addressing youth violence comprehensively.

Amidst these developmental challenges, the impact of ACEs comes into sharp focus. ACEs encompass a wide range of traumatic events or conditions during childhood, including physical, emotional or sexual abuse, neglect, household dysfunction or exposure to substance abuse and mental health issues among caregivers. Research has consistently shown that individuals who have experienced ACEs are at an increased risk of engaging in violent behaviour during adolescence and adulthood (Drury et al., 2017). These experiences can disrupt the development of cognitive competencies, exacerbate behavioural problems and erode PSM, making individuals more susceptible to violence involvement (Haylock et al., 2020; Dong et al., 2004). For further details regarding the link between ACEs and perpetrators of crime, see (Drury et al., 2017).

Moreover, ACEs have far-reaching consequences for mental health and well-being. Children exposed to ACEs are more likely to develop mental health disorders, including depression, anxiety and post-traumatic stress disorder (Drury et al., 2017; Boduszek et al. (2012). These mental health challenges not only impair individuals’ ability to cope with stress and adversity but can also lead to a higher propensity for engaging in violent behaviours as a means of coping or self-defence (Haylock et al., 2020). Thus, ACEs are integral to the complex interplay of factors that influence youth violence.

Understanding these multifaceted dynamics, from PSM and cognitive competence to ACEs and mental health, is essential for devising effective prevention and intervention strategies that holistically address youth violence and its root causes. The integration of these factors into a comprehensive public health approach is key to reducing violent crime and promoting the well-being of young individuals and communities alike.

This systematic review specifically focuses on knife crime interventions. While resources such as the YEF evidence gap map and the YEF toolkit encompass literature on knife crime, they provide limited evidence of knife crime interventions specifically. Therefore, this review aims to address this gap in the literature by examining the effectiveness of interventions targeting knife crime and SVC. By conducting a comprehensive analysis of available evidence, this review seeks to provide insights into the limited number of interventions and preventive programmes that have been evaluated and synthesise the available body of evidence. Other knife crime intervention papers, such as Browne et al. (2022) consists of a broader focus on knife crime such synthesising the available evidence for characteristics and factors of knife crime. Our review captures a narrow focus into the limited impact evidence available for knife crime interventions specifically.

This systematic review aims to synthesise the available evidence regarding knife crime responses, interventions and prevention programmes. Previous reviews and systematic reviews (e.g. Browne et al., 2022; Howard et al., 1999) have not included evidence from the grey literature, in particular, government commissioned reports, and have included more around characteristics and risk of knife crime perpetrators. Some reviews, for example, the YEF toolkit and evidence gap map (YEF, 2022) have included grey literature such as charity websites and reviews, however, these reviews have been more widely focussed on serious violence crime and weapon use rather than specifically considering the evidence for interventions and preventions to reduce overall knife-related crimes. This is of particular importance, given the funding for interventions on educational tools and teachable moments in health-related settings. Therefore, this systematic review includes academic and full grey literature evidence, specifically focusing on capturing intervention and prevention programmes and efforts with a niche focus on knife crime/knife carrying interventions only.

This review aimed to explore:

The purpose of this systematic review is to capture and synthesise the available evidence regarding community and school-based prevention and interventions for knife crime. This systematic review encapsulates the existing empirical evidence and grey literature of knife crime specific interventions under the themes of: impact evaluations, in-school interventions, community, family and health interventions (see results below). A thorough screening process was undertaken with three researchers, applying inclusion and exclusion criteria to the results gathered from searches (see full list of search terms in  Appendix).

The primary search engine used for the initial scoping was Google Scholar as it provides a vast array of literature from many different sources include grey literature. However, extensive searches of databases were also conducted. Searches were completed in February 2023. The scoping exercise informed the final set of search terms, including the removal of terms that were leading to irrelevant papers and sources.

Once duplicates had been removed, all of the collated papers were screened for relevance by titles and abstracts. The remaining papers were screened using full text and applying the inclusion and exclusion criteria below (see Table 1). Papers focusing specifically on domestic violence (DV), violence against health-care workers and school staff were excluded, as they were not specifically related to responses to knife crime in communities. After removing duplicates, all collected papers underwent initial screening based on titles and abstracts to assess relevance. Subsequently, the remaining papers underwent a full-text review with all members of the research team, during which the inclusion and exclusion criteria (detailed in Table 1) were applied. Papers focusing on DV, violence against health-care workers and violence against school staff were excluded as they did not directly address responses to knife crime in communities. The entire research team participated in the screening process, with each member reviewing approximately one-third of the remaining papers. To ensure consistency and reliability in the screening process, a rigorous moderation approach was implemented. This involved periodic cross-checking and discussion among team members to resolve any discrepancies or uncertainties in the screening decisions. Difficult to assess papers, such as those with limited information on methodology, data collection or outcomes (e.g. Cogan et al., 2021), were flagged for discussion between reviewers and moderators to determine appropriate scoring which is reflected in Table 3.

The below inclusion and exclusion criteria (see Table 1) was applied during the screening and selection stage to ensure include evidence focussed on knife crime responses in young adults and children. In this review, interventions conducted before 1990 were excluded due to legislative changes introduced by the Offensive Weapons Act 1996. This act specifically addresses the possession of knives on school premises, making it an offence for any individual to have such articles on school grounds. Additionally, the Offensive Weapons Act 1996 imposes age restrictions on the sale of knives, making it illegal to sell knives to individuals under the age of 16 (Cps.gov.uk, 2022). These legal provisions may have implications for interventions targeting knife crime, particularly those implemented within educational settings and, therefore, guided the exclusion of pre-1990 interventions from our analysis. However, it is important to acknowledge that this exclusion may limit the historical context of our findings, and this decision was made to prioritise the relevance and methodological rigour of the research considered.

The research team used the mixed methods appraisal tool (MMAT) (2018) by Hong et al. (2018) to consider and rate the quality of each of the final set of papers included in the systematic review. This was to ensure rigour when considering the quality of the evidence alongside the contribution to efficacy. The primary appraiser conducted an appraised each paper, adhering to the MMAT criteria tailored to different study designs (e.g. qualitative, quantitative, mixed methods). These criteria encompassed various dimensions of study quality, including the appropriateness of research questions, data collection methods and statistical analyses. To ensure consistency and reliability, a second researcher moderated the reviewed papers and any discrepancies or disagreements in the assessments were resolved through discussion and consensus between the two appraisers. For example, there was limited information in regard to methodology, data collection and data outcomes for Cogan et al. (2021)’s study was a short-form report displayed as a poster for a conference with limited information. This resulted in difficulties when appraising this specific study, as the appropriate information was not available.

The remaining papers were synthesised using an approach similar to that proposed by Whittemore and Knafl (2005) and Wilkinson and Caulfield (2020) of data reduction, data display, data comparison and verification of conclusions. This approach was deemed most suitable by the team, given the inclusivity of combining papers using diverse methodologies (e.g. interviews and survey data with quantitative data). This procedure allowed for the process of identifying patterns, which were then categorised to form the overarching themes for discussion.

The PRISMA diagram demonstrates that 5,430 publications were found from the search terms. With 2,273 duplicates removed and 2,960 excluded, and after screening full text articles, 100 articles did not meet the inclusion criteria; therefore, 13 were included in this review. Table 2 shows that, of the 13 papers included in this review, 12 papers were from UK and one study was from the USA (please see Figure 1).

Table 3 Critical appraisal of studies below details the critical appraisal of the papers included in the final synthesis of the systematic review.

This systematic review included 13 papers, of these papers, three were community-based, four were school-based, one was simulation-based and five in “other” interventions. As a result of the search and inclusion and exclusion criteria, no hospital-based interventions were matched the criteria to be included in this review.

A community-based study conducted by Williams et al. (2014) evaluated the community initiative to reduce violence (CIRV) in Glasgow. This study demonstrated some positive outcomes for reducing violent offending and weapon carrying. The intervention offered statutory services based on the needs of participants such as housing, social services, health and education services. Third sector parties also provided services such as job-readiness training, diversionary activities and personal development. Both categories of services were offered on the basis whereby if participants breached the “no violence, no weapons” they would lose access to these services. Despite the incentives, the results of this study found that there was no reduction in the amount of physical violence detected and observed. This may be due to the evaluation lacking sensitivity to capture changes as no process or outcome evaluation was implemented. Furthermore, it is recognised by the authors that a process evaluation would have been beneficial to analyse the effectiveness of the CIRV.

The second community-based intervention this review analyses is the Kinsella (2011) report. This report encompasses several interventions across the UK and consists of The Young Disciples project in Birmingham, The Adventure Club in Liverpool, Unity Nottingham/Kicks, The Goodies and Hoodies project in Cardiff, The Tees Valley project in Stockton, The Safe Passage-Catch 22 Nottingham programme The Parent Intervention Takes a Stand. The Kinsella report highlights the positive initiatives taking place across the country. However, none of the initiatives included were formally evaluated; therefore, it is not possible to determine the level efficacy, if any, these initiatives possess. Also, the reported initiatives did not include whether each initiative had objectives and if they were achieved. The evidence gap and scarcity of robust evaluations are common in non-academic literature when investigating knife crime interventions. Thus reiterating the need for process evaluations and feasibility studies to enhance robustness and to capture the impact of these initiatives.

Bleetman et al. (1997) also evaluated a community intervention – “Operation Blade”, a community-based intervention using hospital data in Strathclyde, Glasgow. The operation blade initiative commenced in February 1993, which included five phases alongside a prominent media campaign. These phases included a knife amnesty programme, an intensified stop and search campaign, enhanced safety measures in public entertainment venues, such as closed-circuit television, metal detectors, improved lighting and steward training. Additionally, the initiative involved engaging in discussions with knife retailers and addressing secondary school students on the topic. The researchers analysed the quantity and type of attack victims before and after operation blade was put into effect. Upon arrival at the department, patients were triaged by a senior nurse into different treatment categories based on factors such as injury history, injury mechanism, patient appearance and recorded vital signs. Those with penetrating chest and abdomen injuries, signifying a major injury mechanism, were directed to the resuscitation room. The study also compiled a list of stab victims who required treatment in the resuscitation room from January 1992 to August 1994, serving as an indicator of the frequency of severe knife-related assaults during that period. The findings showed that, compared to before operation blade’s implementation, there were no differences detected in the quantity or kind of assault victims visiting the hospital. Although, operation blade initially decreased the number of serious stabbings over a period of 10 months, the numbers subsequently increased thus indicating the need to have a higher level of embedment of interventions. It is concluded that there is a need for continual press, police campaigns, public health and education programmes for improved embedment.

In terms of school-based interventions, a USA study by Crawford and Burns (2016) investigated the impact of school security, law enforcement and environmental factors that reduce school violence. The findings demonstrated that when the school implemented a heavier law enforcement presence, violence also increased. The authors deemed that it is important to thoroughly investigate and capture the dynamics and culture of the school to best formulate an intervention. While this study capture knife carrying and knife violence, there was a heavier emphasis on gun violence, therefore, any adaptations and interpretations from the Crawford and Burns (2016) to UK schools should proceed with caution.

The second study in school-based interventions found was the evaluation by England and Jackson (2013) whereby the researchers aimed to measure the efficacy of workshop sessions delivered by nurses in Liverpool-based schools. The sessions were rated highly by staff and pupils; however, no comparison group was implemented and a self-report survey was the only measure of efficacy used. This decreased the robustness of the evaluation as only the perceived attitudinal changes were collated.

The third school-based intervention included in this review is the Gilbert and Sinclair (2019) workshop-based evaluation, which provided anti-knife crime sessions being delivered in secondary schools where an impact evaluation of “The Devastating After Effects” was conducted. The workshop involved the dramatisation of the consequences and dangers of knife crime and includes facilitated discussions with students. Self-report measures were used pre- and post-intervention. It was found that a significant number of participants reconsidered their likeliness to carry a knife and their perception of its effectiveness for personal safety and, an increased understanding that knife usage affected not only the individual but also the victims. Moreover, more participants expressed willingness to intervene if someone they knew carried a knife. However, limited conclusions can be drawn as self-report measures were used and due to the anonymous nature of this study, we are unable to delve deeply into the underlying factors driving the observed attitudes. Any access to respondent background information is limited to gender alone, preventing the establishment of causation or correlation with regard to these attitudes (Gilbert and Sinclair, 2019). Finally, Gavine (2014) conducted a process evaluation of the Medics Against Violence (MAV) programme in schools in the West of Scotland. The decision to develop a school-based programme was based on literature supporting such interventions and MAV aims to reduce attitudes towards violence and develop empathy. Although MAV lacks an explicit theoretical basis, it aligns with the theory of planned behaviour and the social development model. A research team at the University of St Andrews conducted a process evaluation and it was concluded that MAV was a successful programme in educating young children in schools surrounding the risks and consequences of using knives, as well as improving the participants’ attitudes towards violent, antisocial behaviour (Gavine, 2014). There were 136 health-care professionals which consisted of doctors, dentists and small number of paramedics and nurses were recruited through hospital email and poster campaigns and attended training sessions at the Royal College of Surgeons in Glasgow. New volunteers were paired with experienced volunteers for their initial sessions. The programme consisted of a 50-min session within the personal and social education curriculum and was delivered in various schools in Scotland. During the session, students watched an 18-min film called “Your Choice”, which illustrated the physical and psychological impacts of violence on victims, offenders and their families. The film avoided graphic content and instead included interviews with individuals affected by violence, including a paralysed young man, a mother who lost her son to murder and a young man incarcerated for murder. Subsequent discussions explored topics related to victimisation and perpetration of violence, concluding with strategies for staying safe during social activities.

Aside from its primary goal of reducing pro-violent attitudes, the programme also aimed to enhance empathy among participants by demonstrating and discussing the impact of violence on victims and their families. Empathy is a crucial socio-emotional skill for young people’s social development. The researcher emphasised that building empathy is seen as a protective factor against violence, prompting various violence prevention programmes to focus on developing this skill in young individuals. Interventions based in schools with the backing of health-care professionals can be beneficial as first responses as well as educating people on the medical consequences of knife crime and SVC, which could inevitably reduce the prevalence of SVC and knife-related injuries.

Tribe et al. (2018) developed a mobile simulation workshop to educate young individuals about knife crime’s medical and legal consequences. The workshop involved a sequential simulation, discussions on knife-related attitudes and group activities. The qualitative data showed that participants retained information, reported decreased likelihood of engaging in knife crime, increased willingness to assist victims and knowledge of appropriate emergency responses. The immersive nature of the simulation, accessibility of the workshop, safe environment and, audio-visual elements contributed to its effectiveness. However, limitations included indirect data collection in the first workshop, inability to directly compare data between different time points and unexpectedly well-integrated participants in education or the workplace.

Hopkins et al. (2020) focused on mapping interventions across the East Midlands to understand the landscape of efforts to reduce knife crime in the region. The agencies involved recognise the importance of understanding what works and have established outcome indicators to evaluate their effectiveness. Examples of interventions include mentoring and support in custody and aftercare to offenders, hotspot policing in high rates of violent crime areas, employment support and coaching, sport interventions, mentoring at arrest, trauma therapy, mentoring delivered in schools, music-based interventions, focused messaging regarding risks of knife crime, police cadet programmes, educational anti-knife crime workshops and, parenting/family/carer programmes. To measure outcomes the report presented “key examples of indicators” such as reductions in serious violence offending, Reductions in youth violence improved feelings of community reassurance, securing employment, increases in knife seizures, among others. However, the lack of precise data presented in the report significantly undermines the validity of the “positive outcomes” and results in a low score on the MMAT critical appraisal tool (Hong et al., 2018). In terms of key implications, Hopkins et al. (2020) suggested that the focus of evaluation should be redirected towards examining the impact of the programme overall compared to individual interventions. Hopkins et al. (2020) also highlights the importance of examining the differences between violence reduction partnership (VRP) areas and non-VRP areas, how their practices differ and how the partnership has been developed to plan for a sustainable legacy that can be embedded into the area after VRP’s are dissolved.

The Ward et al. (2011) study investigated the Tackling Knives Action Programme (TKAP). Phase II had both local and national components. Locally, each area received £4–5m per year from the government to develop their strategies. They focused on high-risk areas for youth violence, using the funds for enforcement (like more stop and searches), engagement (such as additional evening activities) and prevention (like anti-knife media campaigns). At the national level, a central budget supported projects across all TKAP areas in England and Wales. These projects aimed to increase data sharing on assaults between hospital accident and emergency departments and local law enforcement. There were also knife awareness campaigns on TV, the internet and posters, including “It Doesn’t Have To Happen”. It is important to note that regions outside of TKAP likely addressed youth violence and knife crime locally, and efforts to tackle these issues were already in progress, as seen with the Metropolitan Police’s Operation Blunt targeting knife crime in London prior to TKAP. Data was collected from 2007 to 2010 from young people aged 13 to 24 in the UK nationwide. During the first year, reductions in serious violence were recorded. In the final two years the TKAP it was found that there were improvements in areas where TKAP was not present and in some cases, improvements were greater in non-TKAP areas. It was reported that the differences between TKAP areas and non-TKAP areas which made comparisons difficult to establish an appropriate comparison group. Furthermore, other national efforts to reduce violence were present in non-TKAP areas in the same time period as TKAP areas, which could also account for the results. Due to these confounding variables, it is uncertain at what level TKAP was effective in reducing knife carrying and knife crime.

Palasinski et al. (2021) conducted four studies that investigated the motivations that drive knife-carrying and the efficacy of anti-knife carrying messaging via mixed methods. Study I used a structural equation model which identified aggressive masculinity as a knife-carrying factor. In Study II, anti-knife carrying slogans were developed which focused on the need for physical defence, control of status and trust in authorities. The findings revealed that injury slogans were rated as the most effective. Study III analysed real-world posters with anti-knife crime messaging. It was found that the most persuasive type of poster contained fresh injury imagery as it was deemed to be “believable” and evoke the most emotion. Study IV consisted of the researchers using computer graphics to create a male avatar which had many forms of disfigurement. It was found that graphic eye injury was the most persuasive, the researchers deemed this to be in accordance with the emphasis of eyes in psychology and their importance in cognitive, emotional and social functions (Palasinski, et al., 2021). These findings used empirical data to identify motivations behind knife carrying they also provide insights into the effectiveness of which anti knife carrying messages are the most persuasive.

The final study by Cogan et al. (2021) investigated young people’s attitudes towards knife seizure imagery. It was found that low-crime and high-crime areas had a negative reaction to the images shown, found them to be counterproductive, and believed that these images encouraged knife carrying. High-crime area participants reported that knife carrying was normalised for their protection and self-defence where as low-crime area participants deemed the images to sensationalise knife carrying. Concerns were raised by high-crime area participants detailing that the imagery stigmatised the issue and needed the appropriate context with the imagery. They also questioned the motives behind using such images. Challenging negative stereotypes and raising awareness about knife crime and its complexities are crucial. Hobson et al. (2022) study looked into the effectiveness of fear-based anti-knife crime campaigns using knife-related imagery. This study found that exposure to the imagery increased awareness of mortality but did not affect participants’ willingness to carry a knife or what participants perceived as beneficial aspects of knife-carrying. The images used were graphic, this had raised concerns about potential negative effects on mental health. Hobson et al. (2022) found that the hypothesis; high knife-related self-esteem/cultural worldviews would increase knife-carrying intentions, was in fact, not supported. As the consequences of carrying a knife did not moderate the effect of exposure to the graphic knife crime imagery, the authors suggested further research is imperative to explore the mental health consequences of fear-based anti-knife crime campaigns, as well as, exploring the relationship with knife-related self-esteem/cultural worldviews further.

The primary aims of this systematic review were to collate and review the evidence surrounding knife crime interventions and their efficacy of reducing knife crime, highlight the limited studies available in this field and, to discuss issues in designing and implementing impact studies for knife crime interventions, especially in the contexts of violence reduction units (VRUs)/VRPs.

The evidence discussed in this review suggests that SVC, and specifically knife crime, are complex issues to challenge, manage and support in a community. However, there have recently been greater efforts focusing on a whole systems approach, which recognises that everyone and every organisation has a contributing role to the bigger aspiration of reducing the risk of SVC and knife crime. The public health perspective seeks to address the root causes for SVC and knife crime, promoting the input at multiple levels to prevent and reduce the risk of SVC in communities (WHO, 2022).

This review has included discussion of evidence from interventions and prevention programmes that aim to educate about the implications of knife crime as well as changing attitudes towards using and carrying knives. There is theoretical evidence to support a more whole systems approach, especially underpinned by a public health perspective WHO (2021). Therefore, this review included discussion of prevention and intervention programmes being implemented at school and community levels. Despite some initial positive findings, several gaps and limitations were identified in the existing literature. Methodological issues, such as small sample sizes, lack of control groups and self-reported data, were prevalent in many studies. Additionally, there was a lack of long-term follow-up to evaluate the sustained impact of interventions. Standardised measures and comparisons between different types of interventions were also lacking, making it challenging to determine the most effective approaches. To enhance the feasibility of implementing standardised and robust data collection in organisations such as VRPs/VRUs, sound data sharing agreements with partners, providers, schools, local authorities and well-planned evaluations with the involvement of researchers from the conception of the evaluation should be implemented to begin addressing the limitations.

The body of evidence assessed has applications that are potentially useful for VRPs/VRUs and organisations associated with VRUs. One of the VRUs role in collaborating with local police, charities and local stakeholders includes assisting and implementing interventions to reduce violence, therefore, being appropriately informed about the current evidence base for knife crime is vital. With the evidence assessed in this review, VRUs can make informed decisions surrounding creating knife crime interventions. Overall, the findings indicate that addressing risk factors whereby schools, community, law enforcement and health-care settings can be used is essential for addressing knife carrying and knife crime. While simulation workshops and educational interventions in the reviewed studies demonstrated favourable outcomes, it is essential to acknowledge that a significant portion of these positive results relied on self-reported data regarding participants’ intentions or attitudes towards knife-related crimes. There is a lack of evidence at this stage to support that changes in attitudes lead to tangible changes in actual behaviour. Therefore, evaluations mapping the long-term outcomes are required to generate the evidence to underpin this.

In light of our systematic review findings, it is evident that interventions targeting attitudes and motivations for knife carrying at the community level play a crucial role in addressing knife crime. Our analysis, including the CIRV by Williams et al. (2014), demonstrates the significance of community-based strategies, highlighting the potential impact they can have. Furthermore, it is recognised that the importance of health-care settings in providing mental health support, risk reduction resources and diversionary routes away from knife carrying, such as employment opportunities is beneficial. While the review did not identify a direct hospital-based intervention, the evaluation of “Operation Blade” by Bleetman et al. (1997) emphasises the role of health care in addressing knife-related injuries and prevention. These findings highlight the multifaceted nature of addressing knife crime through diverse intervention strategies.

We recognise that the Browne et al. (2022) systematic review identifies knife crime offender characteristics, the research concluded that stop and search, knife amnesties, media campaigns and curfews failed to demonstrate significant impact on knife crime reduction versus interventions which aimed to provide individualised support such as employment, housing and education to offenders and resulted in a decrease in weapon carrying. Despite these promising findings only six studies surrounding knife crime intervention was assessed.

Other widely known and well-funded resources such as the YEF toolkit provide many categories of violence and their respective evidence bases, however, the knife crime evidence base in the toolkit is one of the least populated, again highlighting the need to close the large gap in evidence. According to the toolkit, focused deterrence possesses the “high” rating for the impact on violent crime. As part of the toolkit, the YEF also produce technical reports that encompass the available evidence for the specified category of violence. Under the focused deterrence technical report (Gaffney et al., 2021), the main body of evidence that is cited by Braga et al. (2019). The Braga et al. (2019) systematic review did not reveal itself in our review as it did not match the inclusion criteria as there was a notable emphasis on gangs.

We also recognise that excluding papers specifically focused on DV as part of our selection criteria represents a noteworthy limitation in our systematic review. Browne et al.’s (2020) observation that women involved in knife crime often have associations with domestic settings while men are more commonly linked to community settings underscores the importance of considering DV-related knife crime. While our review primarily focuses on community-based knife offending, this exclusion may result in overlooking a distinct subset of knife offenders and a valuable dimension of the issue of knife crime interventions.

The Gaffney et al. (2021) technical report acknowledges the “lack of systematic reviews on this intervention” and highlights the need for a richer evidence base of knife crime interventions to be developed to further support the positive outcomes of evaluated interventions. The technical report acknowledges the difficulties of creating deterrence knife crime interventions due to the need of both police cooperation and cooperation of communities where there maybe issues of trust with authorities. The technical report also suggests that obtaining a positive relationship with the police can be challenging as deterrence intervention requires second chances being given to known offenders, and this may go against the beliefs and values of police work. Therefore, it is suggested to build trust with local communities by increasing community buy-in with service providers/coordinators and the police to increase the chances for successful implementation. There are also other practicalities of implementing a robust methodological approach to community research that are challenging such as the short-term nature of funding cycles and issues establishing good data monitoring and data collection. These may contribute considerably to the lack of impact evidence of knife crime interventions available.

To begin to address these challenges, future studies could aim to recruit larger samples with more diversity to improve the generalisability of their findings. Furthermore, from the process of critical appraisal, the studies mentioned in this review do not provide enough information regarding the completion of outcome data, details of whether delivery of interventions were completed as planned and whether cofounders are addressed appropriately to name a few. Future evaluation studies could look to critical appraisal tools to ensure the evaluations encompasses the elements needed for a high-quality study. Also, the implementation of control groups could be used to improve the effectiveness of evaluating interventions. This can be achieved by using the step-wedge design that allows all participants in a study to eventually receive the intervention being studied but in a staggered or stepped manner. In this design, the intervention is first introduced to a group of participants, then, at prearranged intervals, additional groups of participants are sequentially introduced to the intervention until all participants have received the intervention or prevention (Hemming et al., 2015). This design allows for the comparison of outcomes between groups that have received the intervention at different times, while ensuring that all participants receive the intervention. Also, this design mitigates potential confounding variables from comparing different groups at a single time point. By controlling for these temporal effects, the observed changes can be correlated with outcomes to the intervention. The step-wedge design is often used in situations where it may not be ethical or practical to withhold the intervention from some participants, such as in studies involving high-risk and vulnerable populations or interventions with known benefits, such as knife crime interventions (Hemming et al., 2015; Grayling et al., 2017) However, when implementing the step-wedge design, it requires making sure that the data collection is coordinated in accordance with the required time intervals for delivering the intervention to the appropriate groups. Additionally, the step-wedge design may take longer than other study designs as it entails implementing interventions in stages, which may be an issue for intervention providers as there are usually short-term deadlines organisations such as VRUs face.

Future studies could incorporate objective measures, such as hospital admissions data or police records, to supplement self-reported data, include longer follow-up periods to better evaluate the long-term effectiveness of interventions and focus specifically on high-risk populations to better understand the factors that contribute to knife crime and develop targeted interventions. Furthermore, educational measures could be used to represent short-term outcomes for intervention and prevention programmes with a follow-up post-intervention/prevention to capture the longitudinal outcomes needed to evidence long-term viability of these programmes. Overall, using international research and standardised measures and international research can also enhance the generalisability and comparability of findings. We note that there is difficulty in carrying out robust evaluations due to issues such as inadequate data collection and data sharing. Therefore, it is recommended that practitioners, stakeholders and partners who develop and implement evaluations that a pre-arranged data sharing agreement should be adhered to from both the providers/organisation and researchers to mitigate any risk of data issues that could negatively affect the robustness of the data and intervention. There is also a scarcity of research that assesses the differences between interventions. This can lead to difficulties in making decisions regarding the type of intervention that would be suitable for initiatives and programmes ran by VRPs/VRUs to invest in. Furthermore, few studies investigated the perceptions and experiences of young people who are/have been involved in knife carrying and knife crime combined with empirical data, this is important to capture when planning and implementing evaluations. Another common issue is that there are a lack of longitudinal studies assessing the long-term outcomes of anti-knife crime interventions as most interventions focus on the immediate or short-term follow-up attitudes and perceptions towards knife crime. Therefore, it is unclear whether any changes observed in the short-term are sustained in the long-run.

In this systematic review the complications that arise when addressing SVC and knife crime interventions have been identified. Previous literature suggests imploring a whole systems approach which is underpinned by the public health perspective to tackle knife crime. This systematic review also analyses several prevention and intervention programmes which emphasis the necessity for health-care settings, schools, local law enforcement and communities to collaborate for successful embedment of anti-knife carrying and anti-knife crime. Despite promising initial findings, limitations and gaps in the evidence base were identified which included a lack of longitudinal studies and methodological issues such as not implementing a process evaluation. The importance of well-planned data sharing agreements and collection to enhance robustness of evaluations as well as implementing a stepped-wedge design to enable interventions to have an adequate control group are discussed. The implications of this review regarding VRUs and their partners is noteworthy as we hope to shed light on the evidence gap, its challenges and offer guidance for further development of efficacious knife crime interventions.

This work was undertaken as part of the work commissioned by the West Midlands Violence Reduction Partnership.

Table A1