Sorry, you need to enable JavaScript to visit this website.

Response – developing new homicide prevention initiatives

Developing and describing new homicide prevention initiatives.

First published
Homicide problem-solving guide
12 mins read

This section provides an overview of considerations you should make in designing your own intervention. It also provides some of the language that’s valuable in describing and disseminating your response to senior officers and partners.

We’ve also provided an example of a response plan that follows this approach.

Guiding principles for developing a homicide response

The language and approaches of public health have gradually entered discourse around serious violence. There is much that policing can learn from the public health response to acute health issues and epidemics. The following is a set of guiding principles for the development and deployment of epidemiological interventions during public health emergencies from the Centers for Disease Control and Prevention’s Field Epidemiology Manual (2019). It has been adapted here to apply to homicide but little has been changed.

Homicide interventions should be scientifically driven

Homicide interventions should be scientifically driven on the basis of:

  • established facts and data
  • current investigation findings
  • knowledge from previous investigations and studies

Although important socio-political forces (such as public fear or political outcry) might create pressures for rapid intervention, the interventions must be based on evidence.

However, adapting certain intervention components might be necessary to make them more acceptable and responsive to the needs of the affected community, potentially affected persons, elected officials and the media.

The type and number of interventions will vary

For any given homicide problem, the type(s) and number of interventions to be implemented will vary, depending on the nature of the homicide problem, including its cause and other factors.

The type(s) and number of interventions used might evolve as a function of incremental gains in information developed during the investigation.

Open two-way communication

Most homicide interventions demand – and even might be strengthened by – open, two-way communication between involved agencies and the public.

The environment where offences may occur

Crime including serious violence can be reduced by changing the environment where offending may occur and not only targeting potential perpetrators.

Situational crime prevention is one of the most important developments in modern crime prevention. By changing the focus of intervention to the place and away from the offender, this approach gives greater control and more potential to affect crime.

Situational crime prevention works by making crime less possible by:

  • reducing the opportunity to commit a crime effectively
  • making it less appealing by reducing the potential rewards and increasing the potential consequences

For most crimes, this works well. Opportunity and the appeal are important factors that influence perpetrator decision-making.

In thinking about the focus of your response, the history of problem solving provides a useful set of starting points from which to develop a solution that seeks to change the environment. These 25 techniques of situational crime prevention are described in ‘Becoming a Problem Solving Crime Analyst’ (Clarke and Eck, 2003).

Techniques of situational crime prevention

Increasing the effort the offender must make to carry out the crime

Technique Example activity
Harden targets Stab-proof vests
Control access to facilities Domestic violence prevention orders
Screen exits Knife arches
Deflect offenders Coordinate bookings of sports facilities by rival groups
Control tools/weapons Audit security protocols in knife retailers


Increasing the risks the offender must face in completing the crime

Technique Example activity
Extend guardianship Panic alarms for domestic abuse victims
Assist natural surveillance Incentivise community use of public parks
Reduce anonymity Geolocation tags for domestic abuse perpetrators
Use place managers Night-time economy security
Strengthen formal surveillance Police patrolling of violence hot spots


Reducing the rewards or benefits the offender expects to obtain from the crime

Technique Example activity
Conceal targets Urge caution and awareness after using ATMs
Remove targets Reduce outdoor activity following social media disputes
Identify property Device locator software
Disrupt markets Police crackdown on drug markets to punish and deter group violence
Deny benefits Remote locking devices and bank cards


Reducing or avoiding provocations that may tempt or incite offenders into criminal acts

Technique Example activity
Reduce frustrations and stress Cognitive behavioural therapy
Avoid disputes Taxi marshals in night-time economies
Reduce emotional arousal Cooling-off areas in night-time economies
Neutralise peer pressure Reduce weapon-carrying in groups
Discourage imitation Limit availability of weapons


Removing excuses that offenders may use to rationalise or justify their actions

Technique Example activity
Set rules Enforce consequences for violation of domestic abuse orders
Post instructions Clearly communicate consequences of group violence
Alert conscience Use community leaders to describe consequences of group violence
Assist compliance Ensure public services are fit for purpose and signpost their availability
Control drugs and alcohol Enforce responsible serving practices


A limitation of applying this approach to homicide however is that it’s an extreme event where the normal reasoning thought process may not apply.

You may therefore want to combine interventions that both decrease the attractiveness of an offence (which is affected in part by rational calculation) and increase capable guardianship or reduce opportunity.

Logic models and theories of change

Expected change

A logic model describes what change you would expect from an intervention.

Developing a logic model has many benefits for someone aiming to effect change in a problem or issue. It challenges the response team to think clearly about what their intervention is and is not. Many intervention teams have taken a different approach when asked to express their intervention as a logic model.

In general, this is a good thing as few interventions can maintain their integrity if the delivery team cannot identify what activity is and is not part of their response. In addition, a poorly defined intervention can almost never be evaluated, which limits the potential for the intervention to be rolled out beyond a pilot phase.

How the change should occur

A theory of change describes how the change should occur.

Because a theory of change is more focused on change, it has a greater emphasis on causal pathways and how, for example, inputs lead to activities, then outputs and outcomes.

Components of a response intervention

When developing an intervention you need to think carefully about:

  • resources and inputs – what needs to be in place before it’s implemented
  • the response activity
  • outputs – what the response should produce
  • possible unwanted or unintended effects

The Medical Research Council identifies four components of a logic model that should be considered in framing a response/intervention:

  • implementation – how and what an intervention delivers
  • mechanism – the way(s) in which the intervention creates change
  • outcomes – a measure of what the intervention ultimately aims to change
  • context – external factors that might affect implementation or the activation of the mechanism(s).

As well as providing a more detailed description of each component, we’ve included a worked example of what they might look like in practice.


Before beginning a response, you must consider what’s required to deliver the intervention effectively. These inputs can be a range of factors including:

  • shift changes
  • intelligence and data analysis
  • multi-agency collaboration of the purchasing of equipment

They’re the essential prerequisites for the intervention activity to take place and to work as intended.


Present at the same time but often outside the control of the intervention team is the context. In homicide, this might be:

  • a strong history of collaboration between social services and police in tackling county lines child exploitation
  • high levels of trust in the police
  • community dissatisfaction with group violence

Whatever the contextual factor, it’s important to understand this as it will have implications for all aspects of the response.


The response activity is the most important feature of the intervention to define. Sometimes this can be difficult to separate from routine police work and can require a collaborative effort to refine, particularly if the intervention involves multi-agency collaboration.

In homicide response, activities are highly varied but could include:

  • enhanced surveillance of domestic abuse perpetrators
  • hot spots policing
  • interception of weapons shipment



A benefit of producing a logic model is that it requires you to think about the mechanism (how the intervention will create change). In homicide prevention for example, you might:

  • use your knowledge of behaviour change, such as deterrence or diversion towards support services, to describe the mechanism
  • use the language of situational crime prevention to describe how an intervention reduces the opportunity to offend

Whatever mechanism you identify, you must discuss how the activity causes change. The mechanism does not have to be explained in great theoretical depth, but it should be plausible and based on evidence about effective behaviour change.

Outputs and outcomes

A mechanism does not have to affect the outcome directly and solely. In many interventions, the activity leads first to outputs. These are direct consequences of the activity.

In homicide for example, an activity such as knife sweeps in a local park would not directly reduce homicide but would reduce the accessibility of knives in the event of a conflict. This reduced accessibility may in turn affect the outcome – homicides with knives in public places.

If you cannot show that the activity produced an output, the causal link between activity and outcome cannot be made and you will not be confident that the intervention caused any change you see in this type of homicide.

It’s also important to note that, as with the other components of the response, you can have more than one output or outcome.

Unwanted or unintended effects

The final consideration for designing a response is to consider unintended and often undesirable consequences. These might be realised at an individual or community level.

For example, perceived excessive use of stop and search powers in an area can undermine an individual’s belief that police are trustworthy and competent (output). In turn this might make them less willing to report crime or to cooperate with police investigations (outcome).

Similarly, heavy deployment of focused deterrence – which uses a carrot and stick approach to support individuals to stop offending – in an area that has poor relationships with the police could:

  • create further division between the police and the community (output)
  • undermine the relationship between the community and service providers (outputs), potentially resulting in more violence in that community

Where possible, mitigation should be in place to reduce the likelihood of unintended outputs and outcomes, and these should be carefully monitored.

Response plan example

This response example outlines a combined deterrence and monitoring response to prevent post-separation domestic homicide.


Domestic homicide in the six months after separation.

Problem background

Police force A (population 1.2 million) had eight domestic homicides in 2021/22. This is a 10% increase on 2020/21.

In 2021/22, these were exclusively male-perpetrated offences against female partners.

In five of the eight cases, the perpetrator and the victim had separated in the previous six months and the perpetrator was no longer living with the victim.

In three of the five cases, there was a history of police-recorded domestic abuse and all three perpetrators had been subject to some form of domestic abuse notice or order.


Domestic homicide reviews have shown that in abusive relationships, the weeks after a separation are a particularly high-risk time for homicide (Saxton and others, 2022).

Script – how the homicide problem unfolds

Perpetrators have been banned from the home but know where a victim is likely to be (accessible victim).

Perpetrators living away from home may increase their alcohol consumption and increase rumination and grievance thinking. This can impair communication and reasoning and increase potential for them to return to the home (motivated offender).

Returning to the home increases the potential for a physical dispute (place).

Pinch points

Perpetrators may reason that they can commit an offence in the home undetected (absence of capable guardian – rational and susceptible to deterrence/attractiveness).

Perpetrators may not be reasoning about an offence or may be using faulty reasoning (potentially irrational – situational/opportunity intervention).


Provide perpetrators with a message or letter that there is a panic alarm in the home (also known as a duress alarm or domestic abuse alarm). It should be made clear that the alarm:

  • is linked directly to the police control room
  • constantly records audio in the home
  • can be used as evidence for prosecution

Provide the victim with the alarm.

Inputs – actors

  • Victim who has recently separated from an abusive partner.
  • An abusive partner who has recently been forced to leave the relationship/family home and is subject to a domestic violence order not to contact the victim.
  • Police domestic abuse coordinators/victim liaisons, force control room staff, response officers.
  • Third-party monitoring company engineers, control room staff.

Inputs – resources

  • Purchase of alarms and monitoring support.
  • Information sharing agreement between police and third party.


The intervention combines rationally-driven deterrence with increased situational prevention and formal surveillance. The perpetrator is made aware of the certainty of punishment through a victimless prosecution. In addition to the deterrent, the presence of a panic alarm in the home increases the capable guardianship by ensuring a rapid police response.

Unintended consequences

The installation of a device may anger the perpetrator, triggering a return to the home. A mitigation for this is to ensure that the perpetrator is made aware that the device was installed at the request of the police and not the victim.

Domestic abuse homicide intervention logic model

  • Court order.
  • Intervention support funding.
  • Trusting relationship between victim and services.
  • Police – security provider hotline and evidence feed.
  • Police – security provider information sharing agreement.
  • Installation and testing of alarm by security provider.
  • Identification of recent separation and high risk.
  • Visible RIPA sticker placed at home.
  • Delivery of deterrence message to partner.
  • Improved feelings of safety.
  • Partner does not attend home of victim.
  • Alarm activation prompts rapid attendance.
Potential mechanisms – intended
  • Abusive partner is deterred from attending victim's home.
  • Police emergency response attends home and intervenes before fatal incident.
Potential outcomes – intended
  • Reduction in domestic abuse and domestic homicide incidents.
Potential outcomes – unintended
  • Failure to explain intervention undermines police-victim relationship.
  • Letter triggers angry response from partner creating risk and distress for victim.


  • Victim understands and supports the order.
  • Home is suitable setting for alarm.

Delivering your intervention

Having developed your logic model and refined your theory of change, you’re ready to deliver the intervention. However, it’s important to return to the theory of change at regular intervals.

Police interventions, particularly when they involve external partners, are soft systems that operate alongside and must co-exist with other interventions and activities. Occasionally these systems clash practically or theoretically and it’s important to refer back to and possibly adapt the theory of change to accommodate these clashes.

For example, a focused deterrence intervention whereby police provide enhanced enforcement while local services provide enhanced support requires both sets of partners to collaborate slightly outside of their traditional roles. This may not be problematic at first, but once the novelty of the intervention wears off or the urgency of the problem wanes the intervention system may change. Do not be afraid to refer back to the theory of change model to:

  • reassert the programme goals
  • adapt the theory of change if a more sustainable model can be found.

Next steps

Once you have implemented your response, you can move on to the next stage of the SARA model – assessment.

Was this page useful?

Do not provide personal information such as your name or email address in the feedback form. Read our privacy policy for more information on how we use this data

What is the reason for your answer?
I couldn't find what I was looking for
The information wasn't relevant to me
The information is too complicated