An annex report about HMICFRS fieldwork to support the investigation of the super-complaint on the police response to stalking

Introduction

Background about the super-complaint

In November 2022, the Suzy Lamplugh Trust submitted a police super-complaint on behalf of the National Stalking Consortium about the police response to stalking. His Majesty’s Inspectorate of Constabulary and Fire & Rescue Services (HMICFRS), the Independent Office for Police Conduct (IOPC) and the College of Policing have jointly investigated this super-complaint and published a report.

As part of this super-complaint investigation, HMICFRS has undertaken fieldwork in six police forces in England. This annex report sets out the HMICFRS findings from this work. The words “we” and “our” refer to HMICFRS and the views and opinions of HMICFRS. They do not refer to HMICFRS, the College of Policing and the IOPC collectively, unless this is specified.

About us

HMICFRS independently assesses the effectiveness and efficiency of police forces and fire & rescue services to make communities safer.

In preparing our reports, we ask the questions that the public would ask, and publish the answers in accessible form. We use our expertise to interpret the evidence and make recommendations for improvement.

Previous HMICFRS work on stalking

We recognise that an effective police response to reports of stalking is of great importance. It is for this reason that we have previously conducted several inspections on the subject.

2017: joint report with HM Crown Prosecution Service Inspectorate

In July 2017, we and His Majesty’s Crown Prosecution Service Inspectorate published a joint inspection report: ‘Living in fear – the police and Crown Prosecution Service (CPS) response to harassment and stalking’.

That report contained 22 recommendations for the police and other organisations. In the foreword to the report we said that “stalking crimes are damaging and pervade all sections of society. They can devastate lives and in some cases they end in death. In a digital world, they are crimes that can take place all too easily and frequently.”

This statement is still true today, and it is deeply regrettable that since we published our report, more victims have died at the hands of stalkers. Many more victims have experienced life-changing physical injuries and/or psychological trauma.

Since this first report in 2017, we have undertaken much work, both on a one-off and continuing basis, to help the police to improve their response to stalking to make communities safer.

2018: inspection of Sussex Police

In 2018, the Police and Crime Commissioner for Sussex Police commissioned us to carry out a detailed inspection of the Sussex Police response to stalking and harassment.

In April 2019, we published the resulting report: ‘Stalking and harassment: An inspection of Sussex Police commissioned by the police and crime commissioner and an update on national recommendations in HMICFRS’s 2017 report’. Part A of the report dealt specifically with what we found. Part B looked at what police forces and national organisations had done to improve the police response, and we gave an update on the progress made against our recommendations.

We made four further national recommendations and found one area for improvement, to make sure that all forces continued to enhance their response to stalking and make victims safer.

2021: two thematic inspection reports into how the police work with women and girls

On 26 March 2021, the Home Secretary commissioned us to inspect how the police work with female victims, offenders and witnesses. The report was published in two sections. In July 2021, we published our interim report: ‘Inspection into how effectively the police engage with women and girls’. In September 2021, we published our final report: ‘Police response to violence against women and girls’. Both reports include many references to stalking and consideration of the police response to it. The interim report contained three recommendations. The final report expanded upon these recommendations and contained five further recommendations. These recommendations were not specifically about stalking. But because women and girls are disproportionately affected by stalking, we believe that more general improvements in the police response to crime against women and girls will have a positive effect on stalking victims.

Crime data integrity inspections

Between April 2016 and February 2020, we carried out our rolling programme of crime data integrity (CDI) inspections in all 43 territorial forces in England and Wales measuring the accuracy of recorded crime.

Police efficiency, effectiveness and legitimacy inspections

Since 2020, we have continued our CDI inspections but with an expanded focus on victims’ experiences. As part of these inspections, we sample some cases of stalking. This is in our police efficiency, effectiveness and legitimacy (PEEL) inspection programme, into which we have subsumed CDI inspections.

In our PEEL inspection programme between 2021 and 2022, we introduced an assessment concentrating on the experience of the service forces provide to victims of crime. This is called a victim service assessment (VSA). VSAs include an assessment of forces’ investigation standards.

In this annex report to the stalking super-complaint investigation report, we refer to relevant findings from all the above work.

Methodology

Between June 2023 and October 2023, we conducted our investigation fieldwork. We carried out fieldwork in six police forces in England. All the fieldwork was conducted online without physically visiting the forces concerned.

The fieldwork forces were Hampshire and Isle of Wight Constabulary, Humberside Police, Lancashire Constabulary, West Midlands Police, West Yorkshire Police and Wiltshire Police. We are grateful to the officers and staff in these forces for their help with the investigation.

Fieldwork consisted of three parts. We carried out a document review, in which we examined 254 documents. These included policies, procedures and other material related to stalking provided by the forces.

We also carried out a case file review where we examined 530 case files relating to stalking. The case files consisted of 384 cases that had been recorded as stalking; 72 breaches of order offences; 14 stalking cases that involved applications for Stalking Protection Orders (we intended to examine five of these cases from each force but there were insufficient numbers); and 60 malicious communications offences. We examined malicious communications offences because we have previously found in our other work on stalking (listed above in ‘Previous HMICFRS work on stalking’) that forces sometimes do not correctly identify these as stalking offences. The case file review was based on a random, non-statistical sample, so the results are not representative of all forces in England and Wales.

And we ran a total of 37 interviews and focus groups with police officers and staff, representatives from specialist stalking victim support organisations and staff employed by police and crime commissioners.

Terminology in this annex report

S2A and S4A offences

In 2012, new offences of stalking were inserted into the Protection from Harassment Act 1997 (the 1997 Act).

Section 2A of the 1997 Act creates an offence of pursuing a course of conduct that amounts to stalking. In this annex report, we have abbreviated this offence to ‘S2A’. S2A offences are called ‘summary only offences’, meaning that cases can only be heard in a magistrates’ court.

Section 4A of the 1997 Act creates a more serious offence of pursuing a course of conduct which amounts to stalking and either causes the victim to fear, on at least two occasions, that violence will be used against them or causes the victim serious alarm or distress which has a substantial adverse effect on their usual day-to-day activities. In this annex report, we have abbreviated this offence to ‘S4A’.

Course of conduct

We have used this term in accordance with the definition in section 7 (3) of the 1997 Act. That is:

“In the case of conduct in relation to a single person, conduct on at least two occasions in relation to that person, or

“In the case of conduct in relation to two or more persons, conduct on at least one occasion in relation to each of those persons.”

DA and non-DA stalking

We use two general terms to describe the type of stalking experienced by victims – ‘domestic abuse (DA) stalking’ and ‘non-DA stalking’.

We have used the statutory definition of domestic abuse: that is, offences which take place between two people who are “personally connected”.

Two people are personally connected to each other if any of the following applies:

• They are, or have been, married to each other.
• They are, or have been, civil partners of each other.
• They have agreed to marry one another (whether or not the agreement has been terminated).
• They have entered into a civil partnership agreement (whether or not the agreement has been terminated).
• They are, or have been, in an intimate personal relationship with each other.
• They each have, or there has been a time when they each have had, a parental relationship in relation to the same child.
• They are relatives.

Frontline and investigator

We have used the terms ‘frontline’ and ‘investigator’ to describe officers’ main functions and responsibilities.

By ‘frontline’ we mean officers who were primarily in neighbourhood policing teams, or those teams whose main role is to respond to emergency calls. These officers have received training on investigating priority and volume crimes through level one of the professionalising investigations programme (PIP). This is why frontline officers are sometimes called PIP 1-trained officers. In all the fieldwork forces, we found that frontline officers investigated some stalking offences. We discuss this further in the section ‘Forces often allocate stalking crimes to frontline officers for investigation’.

By ‘investigator’ we mean officers whose primary role is to investigate crime. These officers have typically received further training in conducting serious and complex investigations through level two of PIP. This is why these officers are sometimes called PIP 2-trained officers.

Dedicated stalking officers and staff

We have used the term ‘dedicated stalking officers and staff’ to describe those officers and staff whose main role is related to stalking. These officers and staff are those who are recognised as providing specialist capabilities and sometimes called ‘subject-matter experts’. We discuss these officers and staff under ‘Dedicated stalking officers and staff’.

Specialist stalking victim support workers

We have used the term ‘specialist victim support workers’ to describe those non-police workers who support stalking victims. In some areas, these workers are called independent stalking advocates’.

Special measures

In the context of this annex report, ‘special measures’ mean procedures by the police and other criminal justice agencies to allow victims to give their best evidence; for example, video recording of witness statements. These measures are prescribed by the Youth Justice and Criminal Evidence Act 1999.

Protective orders

In the context of this annex report, ‘protective orders’ mean restraining orders, Stalking Protection Orders (SPOs) and Non-molestation Orders (NMOs).

Initial risk identification

In police control rooms, forces use a model called Threat, harm, risk, investigation, vulnerability and engagement (THRIVE). Sometimes this model is incorrectly called a risk assessment process. But it is more appropriately called a risk identification and demand management tool. Nonetheless, THRIVE still allows officers and staff to identify risks that can be used in a later risk assessment, so it is important that it is completed.

Risk assessment

We use the term ‘risk assessment’ to mean a record of a structured professional judgment to help officers and staff decide the likelihood of risk of harm to victims. If the police do not conduct risk assessments in a structured way, including using a tool, it may lead to errors and omissions. In stalking cases, completing a risk assessment helps officers and staff decide on appropriate levels of intervention to:

• reduce the risk posed by suspects
• increase the protection available to victims

One of the purposes of a risk assessment is to allow the person assessing the risks to piece together the actions of the perpetrator so that they consider behaviour in its totality, rather than dealing with incidents in isolation. Similarly, the assessment should also consider the nature of earlier behaviour experienced by the victim.

Risk management

In this investigation, we have adopted the College of Policing’s definition of risk management. That is: “management of the responses adopted in cases where risk is identified, to minimise risk of further harm by the offender.” Risk management is sometimes also called safeguarding or safety planning.

Risk assessment and risk management are often viewed as the same thing, but they are different parts of a continuum of keeping victims safe.

Risk assessment is a way of gathering information, assessing what the risks are to the victim, and trying to quantify the likelihood of the risk occurring. Risk assessment processes should be reviewed regularly and updated as circumstances change.

Risk management is the process of identifying what safeguarding actions need to be taken to minimise or eradicate the risks to the victim. This can also be applied to those associated with the victim and to the wider community.

Risk management plans are vitally important to keeping victims safe, because they allow criminal justice practitioners to:

• understand the identified risks to the victim
• consider available interventions
• choose and take the most suitable actions to manage the identified risks and protect the victim

Term not used

Stalking offending that takes place between parties that are not personally connected is sometimes called ‘stranger stalking’. We have not used this term in this annex report. This is because it does not consider those offences that are committed by people who are known to, but not personally connected to, the victim, for example, work acquaintances.

Action taken in response to concerns we identified during fieldwork

Inadequate investigations

During our case file review phase, we found 27 cases across five forces where either:

• we were not satisfied that the victim had been properly safeguarded; and where the victim or others seemed to still be at risk; and/or
• in relatively recent cases, we considered there were still evidential opportunities that could be explored to bring about a successful conclusion to the case

We referred these cases back to the forces for review.

We were pleased that the relevant forces reviewed these cases urgently.

In one case, the force re-allocated an investigation to an investigator who sped up enquiries to arrest the suspect. Safeguarding measures were put in place, such as the installation of a monitored alarm and window locks. The suspect was arrested quickly, charged with S4A stalking, and remanded in custody. He pleaded guilty at his first court appearance and was sentenced to four years’ imprisonment.

Incorrect information in training and guidance

In one force, we reviewed a video training presentation that the force had required officers and staff to view. It contained information that was incorrect and was likely to lead to officers making bad decisions about investigations. The presentation said that S2A offences were no longer summary only offences as they were subject to a maximum term of imprisonment of 51 weeks. The presentation showed that these cases no longer had to be concluded within six months of the report being made.

We asked the force to remove this training presentation at once and take remedial action to correct this mistake.

We also found the same incorrect information in a different force, in its behavioural crime recording guidance.

At the time of our investigation, the incorrect information appeared to have originated in the ‘Home Office Crime Recording Rules for Front line Officers and Staff’ (the crime recording rules) document published online and in turn the legislation.gov.uk website. These said that in relation to S2A offences:

“(4) A person guilty of an offence under this section is liable on summary conviction to imprisonment for a term not exceeding 51 weeks, or a fine not exceeding level 5 on the standard scale, or both.

“(5) In relation to an offence committed before the commencement of section 281(5) of the Criminal Justice Act 2003, the reference in subsection (4) to 51 weeks is to be read as a reference to six months.”

However, section 281 of the Criminal Justice Act 2003, which enables the Secretary of State to increase the sentence from six months to 51 weeks has never been enacted for stalking.

When we alerted the Home Office to this error, corrections were made at once.

The incorrect information was also contained in the College of Policing e-learning product on stalking. The College of Policing took immediate action to remove it.

We do not know what the impact has been of the publication and dissemination of this incorrect information. But we think it unlikely that officers would have used it as the only basis for allowing investigations to continue beyond what should have been the six-month time limit. In the section on ‘Other issues of concern’ we discuss our findings about the failure by police to meet the time limit, leading to cases being dropped.

In our examination of the fieldwork forces’ policies, guidance and training, we also discovered several other errors, mainly related to the recording of stalking offences. In these cases, we asked the forces concerned to take remedial action. We have described some examples in our section on ‘Stalking policies’.

Home Office change to the guidance on flagging online crimes

At the time of our fieldwork, the published crime recording rules did not include the full guidance about the importance of adding flags to crime records. The police add flags to crime records so that important information about the crimes can be more easily identified; for example, flagging that an offence is DA related.

We asked the Home Office to correct the omission and we are pleased that they did this without delay. We describe this in more detail in our section on ‘Forces’ understanding of stalking’.

The structure of this annex report

In this annex report, we first set out our findings about the general approach that the six fieldwork forces had taken to stalking. This provides the context to the findings about the features of policing in the super-complaint in the relevant chapter.

Fieldwork forces’ approaches to the policing of stalking

Forces’ understanding of stalking

Forces often do not have a good enough understanding of stalking cases in their area

It is important that forces gather and analyse enough information about stalking victims and suspects. Good quality, comprehensive data that forces effectively analyse allows them to understand the nature of the stalking cases in their force area. This then allows forces to make good decisions about their response to stalking; for example, what resources they need to give the best service to victims.

In ‘Living in fear’, our 2017 report into the police and Crown Prosecution Service (CPS) response to harassment and stalking, we concluded that we were not convinced that any of the forces that we visited comprehensively understood the nature of stalking in their areas. There were some limited attempts to understand the profiles and needs of victims, but we saw no evidence of offender profiling.

In our fieldwork for this super-complaint investigation we found little had changed. Not all forces had a problem profile or some other form of product that helped them understand the nature of stalking, so that they could respond effectively.

Even when forces did have a way of doing this, it was not as comprehensive as it could have been. Examples included:

• no victim profile analysis
• no offender profile analysis
• one product which was limited to S4A offences only

None of the analytical products we reviewed contained information about stalking offenders being managed under multi-agency public protection arrangements (MAPPA).

We discuss in the section titled ‘Surveys and understanding victims’ perspectives’ how most of the fieldwork forces did not conduct stalking victim surveys. This means these forces could not properly understand how well they were responding to stalking offences.

Forces often do not have accurate information about stalking offences because the use of flags is not good enough

We discuss in our section on ‘The misidentification of stalking’ how forces are sometimes not identifying or accurately recording stalking offences.

But we also found that even when forces recorded stalking offences, they were not always recording other important information.

The Home Office requires all forces to place flags on some crimes so that they can gather more accurate information about crimes. For example, forces flag domestic abuse (DA) related offences. We asked fieldwork forces to provide us with lists of stalking cases that had been identified as DA related, by way of a DA flag, and those that had not. We found some DA cases that should have had a DA flag but did not. And we found some non-DA cases that did have a DA flag.

The Home Office also requires forces to use the online flag to identify crimes where “any element of the offence was committed online or through internet-based activities (e.g. through email, social media, websites, messaging platforms, gaming platforms or smart devices).”

The Home Office guidance states that the purpose of the online flag is to “help provide a national and local picture of how internet and digital communications technology are being used to commit crimes, and an understanding of the prominence of certain crimes that are happening online, compared to offline. The flag will help us prioritise spending and direct appropriate resources to tackle online crimes, e.g. in terms of technology, training, digital forensics capacity, and staff. It also gives an insight into the scale and nature of online crime, and can help enhance the development of policies to tackle them and protect victims.”

At the time of our fieldwork, the published ‘Home Office Crime Recording Rules for Front line Officers and Staff’ (the crime recording rules) did not include guidance about the importance of flagging crimes. Instead, the information was contained in separate, fuller guidance the Home Office sent to forces but did not publish. As we mention above in the section ‘Action taken in response to concerns we identified during fieldwork’, we asked the Home Office to correct the omission and they did this without delay. We cannot tell whether this had contributed to forces not always using flags correctly.

In our 2018 report into hate crime, ‘Understanding the difference: the initial police response to hate crime’, we recommended that all chief constables improve the use of the online flag. And in our 2019 stalking report, we concluded that Sussex Police, like most forces, was ill-placed to understand the increasingly common problem of online stalking because its use and knowledge of the online flag were very poor.

In our case file review for this super-complaint investigation, of the 384 stalking cases we examined, 123 involved some form of online offending. But forces only used the online flag in 36 cases.

In all the fieldwork forces, it was the responsibility of the officers who recorded offences and their supervisors to make sure they placed flags on offences. But none of the fieldwork forces had comprehensive systems to make sure officers did this.

In April 2023, one force, knowing that the online flag was underused, analysed the stalking cases it had recorded in the year ending 31 January 2023. Their analysis showed that officers had only flagged 7.3% of stalking offences as involving online offending, whereas in the previous year ending 31 January 2022, officers had flagged 8.7% of stalking offences as involving online offending. So the proportion of stalking offences that had been flagged had decreased.

The force undertook a dip sample of 50 stalking offences to try to understand what proportion of these should be flagged as having an online element. Their analysis concluded:

“A truer percentage of cyber-enabled stalking offences is closer to 16 percent if general internet/social media use is included, and even higher if texting/instant messaging is included.”

Despite this apparent underuse of the online flag, at the time of our fieldwork in July 2023, the force had not started any work to improve its use of this flag in stalking cases.

Forces cannot accurately predict future demand from stalking offences

A force management statement is a self-assessment that chief constables prepare and give to HMICFRS each year. It is the chief constable’s statement and explanation of, among other things, the demand the force expects to face in the foreseeable future.

We have described above how forces sometimes do not have sufficiently accurate and comprehensive information about stalking. This makes it harder to predict future demand.

In our fieldwork, we considered how the forces had tried to understand the future demand from stalking offences.

Four of the six forces had combined stalking offences with harassment offences in their force management statement. From this, we concluded that the forces did not sufficiently understand that the demand from the two types of offences may be different. For example, stalking offences require a consideration of the demand arising from the need to apply for Stalking Protection Orders.

Only two of the six force management statements included an estimation of the existing demand from online stalking. Two of the six forces had included a breakdown of what amount of the demand from stalking offences was DA based. None of the forces had estimated how demand from online or DA-based stalking was likely to change in the future.

Assessing performance

Audit and reassurance procedures

All six fieldwork forces had done work to understand the extent of the misidentification of stalking in their force area. But this work varied in its extent and focus. In two forces, a specific stalking audit had not taken place for two years. We think that audits should take place at regular intervals of no longer than two years so that forces have a good understanding of how they are responding to stalking offences.

Many of the stalking crime recording audit reports that forces shared with us showed that there were problems with the misidentification and recording of stalking offences.

In one force, an audit report from July 2023 concluded:

“Staff are not adhering to Crime Data Integrity guidelines in all cases. In some cases they are not identifying incidents as stalking or harassment. In other cases, they are criming [recording as a crime] incidents as stalking or harassment when the incident does not fit the definition.”

In another force, a stalking audit report from August 2023 found that there was:

“A lack of recognition of a course of conduct or behaviour which could indicate stalking or harassment; confusion about what stalking offence to record; and it was commonplace to find several reclassifications throughout the occurrence lifetime between different behavioural crimes.”

In one force audit we examined, we found a concerning example that showed how the force had failed to properly identify and record stalking offences:

“The victim reported harassment by her ex-partner and his parents. A stalking offence was not recorded as it was considered that there did not appear to be a course of conduct of stalking/harassment because the victim had been contacting her ex-partner and his family. An occurrence was created with the occurrence type ‘Domestic Incident (Non Crime)’ and finalised without a crime classification.

“However the victim had made several similar reports earlier in the month. Five Occurrences had been created, all of which had ‘Domestic Incident (Non Crime)’ as the Occurrence type. Four had been finalised without a crime classification and the other had the classification ‘Non Crime Domestic Incident.’ An examination of those reports identified that common assault and malicious communications offences should have been recorded and there did appear to be a course of conduct of stalking.

“The following month the victim was able to get a non-molestation order issued against her ex-partner and he breached it by calling her. Although that offence was being investigated, the occurrence had ‘Non-recordable miscellaneous’ as the occurrence type and hadn’t been given a crime classification, so the crime hadn’t been recorded.

“Although none of the four crimes against the victim had been recorded, a counter‑allegation by her ex-partner had been recorded as an assault with the victim of the unrecorded crimes as the suspect.”

Surveys and understanding victims’ perspectives

We believe it is important that forces undertake surveys with stalking victims. Consistent and high-quality force-level information on stalking allows chief constables to check their force’s performance in this area. It also allows them to assure themselves that their force’s performance is effective. And it lets them spot emerging trends or problems and act in response.

In our 2021 report ‘Police response to violence against women and girls’, we concluded that not conducting victim surveys meant forces missed opportunities to use the experiences of victims to improve practice.

In the year ending 31 March 2023 only two of the six fieldwork forces had surveyed stalking victims.

One of these two forces told us that its survey of 36 stalking victims revealed an overall satisfaction rate for victims of stalking of 88.9%. But some results from the individual survey questions reiterate our findings and concerns raised elsewhere in this annex report.

For example, in the same survey, only 59.3% of victims reported they had been kept informed about their case, only 33% of victims were completely satisfied with the updates they received, and only 37% of victims were completely satisfied with the police actions.

The other force that had conducted a recent survey of 375 stalking victims found a higher overall victim satisfaction rate for stalking victims (78%) when compared to the satisfaction rate for victims of all crime (69%). But in this force the survey did not differentiate between offences committed in a DA and a non-DA setting. The survey could have been improved by separating these responses so that the force could understand whether the service provided to these types of victims was different.

Another force had last conducted a survey in April 2022. It found that its overall stalking victim satisfaction rates in the preceding 11 months were generally in the 60‑70% range. But the force had not taken a more recent survey to understand whether this had improved or where other action was needed.

Stalking policies

It is important for forces to have clear policies on important policing subjects. This is because having such policies will:

• set standards for how the force wishes its officers and staff to deal with particular crime types
• provide information to officers and staff about specific local priorities and ways of working that may vary from the generic College of Policing authorised professional practice (APP)
• provide an easy reference point for officers and staff to refer to when seeking guidance
• set a standard forces can use to carry out audits

Force stalking policies are sometimes not clear or accurate enough

A detailed and comprehensive stalking policy is particularly important. This is because stalking investigations can be complicated and multi-faceted. And stalking investigations involve different procedures to other crime types. Also, some officers and staff may not regularly encounter stalking offences so they may not be familiar with how to deal with them.

Five of the six fieldwork forces did not have a specific stalking policy. Instead these forces had policies that combined stalking with harassment. One of these five forces had a policy that had high-level strategic statements but little information about operational procedures. It referred to the College of Policing’s advice for further detail. This meant that information about local procedures was not all kept in one single policy document.

We have also described how some other aspects of these policies can be improved in the section on ‘The misidentification of stalking’.

Training and guidance

Stalking training

In all six fieldwork forces, we were told that training for officers and staff would help resolve the problem of misidentification of stalking, and improve awareness.

The training provision was different in every force. But all the training included how to tell the difference between harassment and stalking.

Some force training concentrated specifically on why it is important to identify stalking. For example, research shows stalking behaviours are often present in the lead-up to domestic homicides.

In some forces, part of the training was provided by people who were not employed by the police; for example, specialist stalking victim support workers or psychologists. We believe this is a good thing because it helps officers and staff have a better understanding of stalking, and allows them to hear a different perspective.

In general, all the training we reviewed concentrated on how officers and staff can recognise stalking. But the training lacked content on other aspects of police practice related to stalking. For example, in some forces, the training did not include information about how to record stalking in accordance with the crime recording rules.

In one force, the training package for frontline officers did include information about how to record stalking, but it was incorrect. The training said that stalking behaviours in a relationship should be treated as coercive and controlling behaviour. This is contrary to the crime recording rules.

The training material that forces first gave us for our document review was for officers who had completed their initial training: that is, those who were no longer student officers. In our fieldwork we asked forces to show us the material for student officers.

One force told us that it did not give its student officers specific force-designed training about stalking. Instead, they completed the College of Policing e-learning package.

We found that the student officer training content varied from force to force. At the time of our fieldwork, none of the forces had recently reviewed their student training. We found it was missing important information; for example, how to properly record stalking offences.

Other information and guidance related to stalking

Some of the fieldwork forces circulated short-format information materials about stalking to officers and staff. These included a newsletter and video recordings.

We think these are good ways of informing officers and staff about stalking. We recognise that they may not always have the time to look at longer material.

Information about stalking on forces’ intranet sites

All the forces had stalking-related guidance on their intranets. At the time of our fieldwork, one force had only just set up its stalking intranet site.

The intranet sites generally contained useful information and links to other sources of information, such as the College of Policing stalking advice.

All the intranet sites varied, in terms of content. We were satisfied that these sites were accessible to officers and staff. And we were satisfied that the sites contained information that would help them deal with offences of stalking.

Responding to stalking offences

Dedicated stalking officers and staff

Five of the six fieldwork forces had at least one officer or staff member who worked exclusively on stalking offences. We spoke to these dedicated officers and staff. We found them well informed and well trained. We also saw good evidence of them working well with partner organisations.

The non-exhaustive list of tasks undertaken by these dedicated stalking officers and staff included:

• being a single and known point of expertise for the force (both internally and externally)
• being a single point of contact for direct liaison with specialist stalking victim support services
• screening cases suitable for Stalking Protection Orders and liaison with legal services
• formulating and providing training and guidance
• daily checking of other behavioural offences to make sure stalking offences had not been misidentified
• developing and administering multi-agency responses to stalking; for example, as part of stalking clinics

These dedicated officers and staff also checked stalking offence reports daily. This was to make sure the force had considered all necessary investigative and safeguarding actions. It also helped with generating remedial actions. And it ensured that the force had correctly recorded offences as either S2A or S4A.

The exact numbers of officers involved in these tasks varied between forces, but we considered that they added considerable value to the force response to stalking.

Resources to investigate stalking offences

Some of the forces had different arrangements for which teams should investigate stalking offences. We have discussed this in the section titled ‘Some forces have different ways of allocating stalking cases for investigation’.

Multi-agency stalking arrangements

Two of the six fieldwork forces had specific stalking multi-agency arrangements. These were administered with the help of the dedicated stalking officers and staff we described above. We have described these arrangements in more detail in the section on ‘The lack of provision of stalking intervention programmes’.

Digital forensic analysis

Five of the six fieldwork forces had digital forensic analysis policies that were likely to give sufficient priority to the analysis of devices in stalking cases.

In one force, we were provided with a prioritisation matrix for submissions to the digital forensic unit. Offences in the matrix were grouped depending on the apparent seriousness of the offence, which, along with an additional risk category score, informed decisions about case priority. We were surprised to see that stalking sat in the same priority group as burglary, robbery and some drugs offences. As stalking offences are more likely to have vulnerable victims and potentially dangerous suspects, it would be better to give these offences more priority.

Force crime registrars

Each force has a crime registrar who oversees compliance with the crime recording process. The registrar’s responsibilities include training staff in crime recording and checking compliance with the crime recording rules.

In most forces, we saw evidence that at the end of an investigation of an offence not classified as stalking, the force crime registrar or their staff may intervene and change the final classification to stalking. We also saw evidence of disagreements between force crime registrars and officers or supervisors about these decisions.

Changes to the classifications of offences are an important administrative exercise to make sure forces keep accurate records of crime. But it shows that recording and investigating officers – and their supervisors – lack understanding about and ‘misidentify’ stalking. Also, if a victim has not been referred to a specialist stalking victim support service, changing the classification at the end of an investigation will not trigger a referral.

Initial risk identification

Force control room staff often complete a risk identification process in stalking cases

We found that control room staff completed the threat, harm, risk, investigation, vulnerability and engagement (THRIVE) tool in 398 of the 470 stalking and breach of protective order cases that we examined.

Officers and staff often do not identify all the risks to stalking victims

It is possible that control room staff may not identify some risks in the initial THRIVE process. It is also possible that the risks to victims change after control room staff undertake the THRIVE process. So later risk assessments provide valuable opportunities for officers to conduct more comprehensive assessments, including using information provided by victims, and from background research.

In our case file review we found that there was some risk of serious harm or homicide to the victim in 297 of the 470 stalking and breach of protective order cases we assessed. But there was only evidence that the police had identified all of these risks in 66 of the 297 cases. This is especially concerning. This shows that officers and staff are missing opportunities to identify the risks to stalking victims. We have described in the section on ‘The misidentification of stalking’ how one effect of this is that officers sometimes record S2A stalking offences when they should have recorded the more serious S4A offence.

Risk assessment

In DA stalking cases, all of the fieldwork forces used the Domestic Abuse, Stalking, Harassment and Honour-Based Violence (DASH) risk assessment model or Domestic Abuse Risk Assessment (DARA) tool.

In our 2017 report into the police and CPS response to harassment and stalking, we said we were concerned that there was no risk assessment process for non-DA stalking victims. So we recommended that the National Police Chiefs’ Council (NPCC) stalking and harassment lead should ensure that the risks to victims of stalking are properly assessed by commissioning the development of an evidence-based approach to risk assessment in stalking offences. We also recommended that until the completion of the above review, forces should use a DASH model (or equivalent) for all stalking offences as an interim measure.

Officers do not complete a record of structured risk assessments in every stalking case

Overall, in our case file review, we found evidence that the police had only completed a record of structured risk assessments in 190 of the 384 stalking cases we examined. We did not assess whether frontline officers had completed the risk assessments at the time, or whether they were completed subsequently.

But we think this lack of evidence of risk assessments is unacceptable. This is because stalking suspects present a considerable risk to victims.

In the DA stalking cases we examined, there was evidence that the police had completed a DASH or a DARA in only 143 of the 197 relevant cases. In the non-DA stalking cases we reviewed, the police had only conducted a structured risk assessment in 28 of the 187 relevant cases.

In forces that had conducted their own audits of stalking cases, we also found evidence that the police had not completed risk assessments in stalking cases.

For example, in July 2023, one force conducted an audit which said:

“As a force overall we are failing to submit a DASH risk assessment when required with only 41% of occurrences out of 102 having had a DASH risk assessment attached. There are a few cases where the victim is not engaging or has been supported by another force however, the majority have no rationale for the lack of DASH.”

At the time of our investigation, the West Midlands Police and Crime Commissioner had given West Midlands Police funding to train 50 members of staff in a risk screening and initial assessment process called the screening assessment for stalking and harassment (SASH). The SASH process is sometimes used by specialist stalking victim support services. This is positive, but at the time of our fieldwork this was a new initiative. So the force had not decided how best to implement the SASH process.

Some forces have inefficient systems for completing risk assessments

We found that even when some forces had policies stating which risk assessments to complete in stalking cases, there were system problems that made them inefficient.

In one force, the way the IT system was set up meant that officers had to submit a completed DASH form for one offence, and a blank or dummy DASH form for any linked offences.

In another, the force policy stated that officers had to submit two different types of risk assessment forms. These largely served the same purpose.

The ‘S-DASH’ question set

In non-DA stalking cases, five of the six fieldwork forces had policies requiring completion of an S-DASH. The S-DASH is 11 questions intended to identify the risk to victims of stalking. These 11 questions are taken from the DASH.

In our report ‘Living in fear’, we stated that the S-DASH questions are insufficient on their own to properly assess the risks to the victim. The interviewer should also ask about the effect of the behaviour on the victim. On 4 June 2019, the NPCC stalking and harassment lead sent a letter to all chief constables stating:

“Completion of these questions is not regarded as a risk assessment in isolation and does not show the full effect of the stalking behaviour and the consequent needs of the victim.”

The letter goes on to advise that the NPCC lead was working on a tool to risk assess non-DA stalking cases. It says that, in the absence of a suitable risk assessment tool for stalking, forces should use a DASH for non-DA stalking cases until this tool is complete.

Only one of the six fieldwork forces had adopted our recommendation (and the advice of the NPCC lead) discussed above about risk assessments in non-DA cases.

The stalking screening tool

In response to the recommendation, the NPCC stalking and harassment lead started work on a stalking risk tool called the stalking screening tool (SST). But we think it is regrettable that, seven years after our initial report, there is still no widely used risk assessment for non-DA stalking victims. Only a small number of forces have so far piloted the SST.

Three of the six fieldwork forces were pilot sites for the introduction of the SST.

But none of these three forces had done a good job of implementing the pilot and making sure the SST was completed in stalking cases. We found a lack of direction and leadership about the implementation of the SST in these three pilot forces. In the stalking cases we examined, the SST had rarely been completed. Of the 183 stalking cases we examined in the three pilot forces, officers and staff had only completed the SST nine times.

In one SST pilot force, an entry in the force management statement summarised:

“Priorities are not connected within stalking offences. 70% of stalking is linked to DA where there is a DASH assessment completed. The remaining 30% has no current risk assessment due to the limited capacity to implement one.”

Risk management

In ‘Living in fear’, we reported that the police often did not complete a risk management plan for stalking victims. We recommended that the NPCC stalking and harassment lead should make sure that the risks to victims of stalking are properly managed. We also recommended that work on an evidence-based approach to risk assessment in stalking offences should also consider whether to include a risk management plan with any risk assessment tool.

In the report we noted that some forces’ DASH forms did not incorporate a risk management planning section. Because of this, it was more likely that officers and staff would fail to complete a risk management plan.

Some forces have replaced the DASH form with the DARA. The DARA does include a risk management section. But police forces have not yet widely adopted the DARA.

In our 2019 stalking report we found a similar problem in Sussex Police.

After we made our recommendation, the NPCC began to develop the SST. We have discussed this above. We are pleased that the current version of the SST refers to the need for a risk management plan. But as we have said above, the SST has not been widely used.

Officers often do not complete risk management plans

In our case file review, we only found evidence of a risk management plan in 80 of the 470 stalking and breach of protective order cases we examined. This includes only 25 of the 194 S4A offences we examined. This is especially worrying and means that we could not be sure that all stalking victims were safeguarded.

Police forces in England and Wales use the DASH more widely than the DARA in DA cases. There is also no commonly used risk assessment tool for officers in non‑DA cases. We think both of these factors have significantly contributed to officers not completing risk management plans. This is because, as we describe above in ‘Risk management’, the DASH sometimes does not include a risk management plan. And in non-DA cases, the SST has not been introduced in all forces.

In the fieldwork forces that had conducted audits of stalking cases, we also found evidence that the police had not completed risk management plans in stalking cases.

In one fieldwork force, an August 2023 report about stalking concluded:

“There was a lack of comprehension, both as to how to deal with the inherent risk factors involved in stalking cases and in more general terms, and how to formulate an action plan mitigating that risk and identifying protective factors.”

The same report said:

“There is no one place to record what safeguarding has been carried out and of course multiple options to use.”

It is also important to note that this may not be a problem just in stalking cases. For example in our 2018 hate crime inspection, we found that the police often do not complete risk management plans for other types of crimes.

Offender management

Stalking perpetrators can be dangerous and can have complex needs. They need careful management to address their offending behaviour and reduce the risk of harm to the public. Multi-agency public protection arrangements (MAPPA) mean the police, probation and prison services can work with other agencies to assess and manage violent and sexual offenders to protect the public from harm. MAPPA are well established and exist in all police force areas.

In ‘Living in fear’, we concluded that a substantial number of potentially dangerous stalking perpetrators were probably not being managed under recognised offender management processes, such as MAPPA.

In March 2023, His Majesty’s Prison and Probation Service published revised MAPPA guidance which contained instructions regarding stalking offenders. The guidance states that:

“Those with convictions for stalking or who display stalking behaviours must also be considered for referral to Category 3 if they do not fall into Category 2 management.”

At the time of our fieldwork, none of the College of Policing’s stalking advice contained any reference to MAPPA. But the advice to investigators (PDF document) includes a link to the College’s principles for the identification, assessment and management of serial or potentially dangerous domestic abuse and stalking perpetrators (PDF document) which includes information on MAPPA. The advice to investigators also says that investigators should “explore support available from local risk management teams, e.g. integrated offender management, management of sexual and violent offender units, and other multi‑agency risk management panels/processes”.

Only one of the six fieldwork forces’ stalking policies made any reference to MAPPA arrangements. The one force policy that did mention MAPPA only offered general information about the need for officers to consider individuals “jointly in a multi-agency meeting with regards to some form of offender management being put in place”.

In our case file review, we did not look at any activity after the police had decided the outcome of an investigation. This means we could not tell whether the police made any MAPPA referrals and what the outcome of these referrals had been.

Victim care

Officers often do not recognise the enhanced status of stalking victims under the Victims’ Code

In our fieldwork we considered how well forces complied with the rights of victims that are contained in the Code of Practice for Victims of Crime in England and Wales (the Victims’ Code).

The Victims’ Code makes it clear that some victims are entitled to an ‘enhanced service’. One type of such victims are those who have been persistently targeted. The Victims’ Code states:

“You are eligible for Enhanced Rights under this Code as a persistently targeted victim if you have been targeted repeatedly as a direct victim of crime over a period of time, particularly if you have been deliberately targeted or if you are a victim of a campaign of harassment or stalking.”

In our case file review, we found evidence that officers recognised the victim was entitled to enhanced rights in only 97 of the 384 stalking cases we examined.

The College of Policing ‘Advice to police responders to ensure an effective response to reports of stalking or harassment (PDF document)’ says:

“Victims who are persistently targeted over a period of time, particularly if they are a victim of a sustained campaign of harassment or stalking, will be entitled to an enhanced service under the Victims’ Code of Practice.”

We examined force stalking policies and training documents to see whether forces had included content about the need to give stalking victims enhanced rights.

Only one of the six forces’ stalking policies included the fact that stalking victims are entitled to an enhanced service.

None of the training in any of the fieldwork forces included this. One force had produced draft training material that stated officers should “consider” giving stalking victims enhanced rights.

We think that one reason why officers often do not recognise the enhanced status of stalking victims is because this is often not included in stalking policies and training.

We did not examine all the other non-stalking related policies and training material in forces. So it is possible that the Victims’ Code and enhanced rights for victims were included in other documents.

Officers often do not complete victims needs assessments

The Victims’ Code requires the police to conduct a victim needs assessment (VNA) for all victims.

In our December 2023 report ‘Meeting the needs of victims in the criminal justice system’ we summarised that the VNA should form the bedrock of the police response to victims. If the police do not consider a victim’s needs at the start of an investigation, it is more likely that other criminal justice bodies will not consider these needs in the later stages of the investigation and prosecution.

The same report found that the police often did not complete VNAs and recommended that:

“By 31 December 2024, the College of Policing should work with the National Police Chiefs’ Council and chief constables to develop minimum standards for the completion of victim needs assessments. These should include standards for timeliness of completion and clarity on the information to be recorded.”

We reported similar findings and made a similar recommendation in ‘The poor relation’, our report on crimes against older people. We said: “There is little evidence that the police are routinely assessing victims’ needs”.

In this super-complaint investigation, we found that the police had completed a VNA in only 82 of the 384 stalking cases we reviewed.

Five of the six fieldwork forces did not have a comprehensive and structured VNA. In the absence of this, officers recorded information in different places on crime records. We found that this led to errors and omissions. For example, officers had recorded that stalking victims were not entitled to an enhanced service when they were entitled to this.

Some forces do not understand how well they look after victims

In our December 2023 report ‘Meeting the needs of victims in the criminal justice system’, we found a lack of good quality data about compliance with the Victims’ Code. And because of this, we said we could not assess whether the police had provided victims with the right information at the right time.

In our super-complaint investigation fieldwork we found that forces had done little audit work to find out whether victims had been treated according to the Victims’ Code. One force told us that it did not have enough resources to do such audits.

We have also described above in ‘Surveys and understanding victims’ perspectives’ how most of the fieldwork forces did not conduct surveys of stalking victims’ experiences.

Other aspects of victim care are sometimes not good enough

In our case file review, we examined whether there was evidence that officers had overtly minimised or trivialised the behaviours reported to them. We found some evidence of this in 46 of the 470 stalking and breach of protective order cases we reviewed. Examples include officers not treating the cases as a serious S4A stalking offence or treating the reported behaviour in isolation.

We also found that in some forces the police sent victims generic letters that did not mention the victims’ entitlements to enhanced rights. We also reported similar findings in our ‘Meeting the needs of victims in the criminal justice system’‘Meeting the needs of victims in the criminal justice system’ inspection report. In that report, we said that communication with victims was not consistent, and the service provided often fell short of the minimum standard.

Some forces had preset their crime recording system so that officers should update victims every 28 days. We found that some officers did not change this default setting depending on the specific needs or wishes of victims. Stalking victims are entitled to an enhanced service and so may sometimes need more frequent updates. But officers had not accounted for this.

Forces can do better at keeping victims safe online

We found little evidence that officers had the IT hardware and software they needed to safeguard victims of online stalking. It is likely that many stalking offences feature online behaviour. So this should be an important part of the service forces provide to victims.

When we asked fieldwork forces about their approach to online safeguarding, most of the things they told us about related to the recovery of evidence. For example, some forces told us about ‘digi-dogs’. These are dogs that have been trained to find hardware such as USB sticks.

Some forces told us about their digital media investigators (DMIs). These are officers who have been trained in retrieving evidence from, and providing evidence about, digital devices.

Forces told us that the DMIs can advise investigating officers about online safety for victims. But we found no evidence that the DMIs routinely did this, or that the investigating officers requested the advice.

Examples of forces trying to keep victims safe online

Some forces provided us with details of how they have tried to deal with the problem of keeping victims safe online.

At the time of our fieldwork, West Midlands Police had recently bought a handheld wireless intrusion detection system. This is a device the police can use to sweep a victim’s address or other property for surveillance devices.

West Midlands Police also told us it had a policy that officers should refer victims of online crime to The Cyber Helpline. This is a national charity that provides free, expert help for victims of cybercrime, digital fraud and online harm.

Lancashire Constabulary has a Digital Media Investigation Unit (DMIU). The force told us that the DMIU trained the constabulary safeguarding teams in digital safeguarding and investigation. It also told us that the DMIU reviews crime records to identify opportunities to help investigating officers. Unit staff can also visit victims to give digital safeguarding advice. In our fieldwork we did not assess in detail how this unit operated.

Humberside Police told us that all appropriate officers were required to complete the College of Policing’s Operation Modify remote learning. This training helps officers to get the digital skills they need to investigate effectively.

While we think this is encouraging, it is clear that most forces do not do enough to keep victims safe online.

Overall victim care was often not good enough

In our case file review, we considered whether all the above aspects – when taken together – meant that the overall standard of victim care was good. In 196 of the 470 stalking and breach of protective order cases we reviewed, we decided it was not. Case study 1 is just one example which illustrates where victim care was not good enough.

The features of policing complained about

The super-complaint identifies four important areas of concern around police responses to stalking, as well as highlighting a number of other issues. In summary, these are:

• the misidentification of stalking
• flawed investigations and No Further Action (NFA) decisions
• the failure to offer or refusal to apply for an interim or full Stalking Protection Order (SPO)
• the lack of appropriate response to breaches of protective orders

The misidentification of stalking

Guidance and policies are sometimes confusing

Five forces had policies that covered both stalking and harassment. This was reflected in the titles of the policies, ‘stalking and harassment’, and this form of words appeared throughout those policies. One force policy was called a ‘stalking or harassment policy’.

We were told that the use of the word “or” was deliberate. It was to emphasise the difference between the two offences. This follows the College of Policing guidance documents which changed in 2019.

We understand why most force policies cover both offences in one single document. This is mainly because the offence of harassment and the offence of stalking are both contained within the Protection from Harassment Act 1997. But the conflation of harassment with stalking in policies reinforces the mis-held belief that these offences are mainly the same but with some differences.

In one force policy, the first line read “stalking and harassment is an offence which…”. We think this wording is likely to confuse the reader.

We know these differences in wording appear minor. But it is important for forces to avoid confusing officers and staff. And it is important to avoid reinforcing a belief that the offences are essentially the same and require the same response. They are not.

A common description of stalking is used across guidance throughout the criminal justice system. Stalking is described as “a pattern of unwanted, fixated and obsessive behaviour which is intrusive”. This description is abbreviated as the mnemonic FOUR – fixated, obsessive, unwanted, repeated. Only two of the six policies from the fieldwork forces included reference to FOUR.

In the force policies we reviewed, we found a variety of statements which were intended to help officers and staff differentiate between stalking and harassment.

For example, one force policy said:

“Stalking is different to harassment as it relates to fixation and obsession of a person rather than nuisance behaviour.”

Another force policy said:

“Stalking will often focus on a person, whereas harassment will often focus on disputes.”

This phrase also featured in the force’s training provision which we describe in the section on ‘Training and guidance’.

The origin of this statement seems to be\ the College of Policing ‘Advice to police responders to ensure an effective response to reports of stalking or harassment (PDF document)’ which says:

“Stalking will often focus on a person, whereas harassment will often focus on disputes.”

The College of Policing has told us that it intends to amend this wording. We think changing this will help officers and staff avoid making false assumptions when a victim tells them about a dispute.

The ‘Home Office Crime Recording Rules for Front line Officers and Staff’ (the crime recording rules) give guidance to forces on the recording of stalking (and other) offences. Four of the six fieldwork forces’ stalking policies did not have any information about this or links to relevant documents.

One force published a new force stalking (and harassment) policy in August 2023. But the section titled ‘Crime Recording Considerations’ included an incorrect instruction to officers and staff. It said that behavioural crimes should be recorded in addition to the most serious other notifiable offence reported at the same time. This recording rule had changed in May 2023.

Other factors affecting the misidentification of stalking

Generally, the first opportunity to identify stalking correctly is in police control rooms. Most stalking offences are reported by the victim, or on their behalf, in this way.

So we examined how police control room staff in the fieldwork forces dealt with stalking allegations. Control room staff first record incidents on their force command and control system. There are several different types of command and control room systems in operation across the 43 forces in England and Wales.

Control room staff are expected to record incidents in accordance with the national standard for incident recording (NSIR). The main aim of the NSIR is to ensure that police record all incidents, whether crime or non-crime, in a consistent and accurate manner.

We were very surprised to find that only one fieldwork force had a specific ‘type code’ (also known as an ‘opening code’) to identify stalking. In the other five forces, stalking had to be recorded under the different codes for domestic abuse, harassment or something else.

While we do not know for sure, we suspect the same situation exists across many police forces in England and Wales.

In one force, its ‘overview of stalking’ document from March 2023 showed the potential impact of not having effective ways to easily identify stalking incidents in control rooms. The document explained that a keyword search for the word “stalking” in incidents recorded during the year ending 31 December 2022, showed that this word had been recorded in 4,000 incidents. These ranged from domestic abuse and harassment through to six other categories of incident, including one of “other”. But in the same year, the force only recorded 2,400 offences of stalking. This creates an obvious and very large discrepancy.

The force examined a randomly selected sample of these incidents. Its review showed that it had recorded 64% of the reports in the sample as offences other than stalking. In 12% of reports, no offence was recorded at all.

The force’s report concluded that it was highly likely that this meant one of two things: the reports were either being re-categorised or closed as further information came to light during the investigation; or the force was under-recording stalking incidents.

The report recommended introducing a force qualifier for stalking. At the time of our fieldwork, the force had not introduced one but was considering doing so.

Tools to help identify stalking

We have described in more detail in the section on ‘Risk assessment’ how the National Police Chiefs’ Council (NPCC) stalking and harassment lead started work to develop and pilot the stalking screening tool (SST) to help officers identify stalking.

Three of the six fieldwork forces had been pilot sites for the introduction of the SST.

But none of these three forces had widely introduced the pilot. And where they had introduced it, they had not used it to its potential.

One force told us that when it introduced the pilot, it was not a success. Many officers in the first pilot site had not been trained. As a result, during the initial pilot period, officers and staff only completed an SST in 10-25% of stalking cases. The force launched the pilot again in a different pilot site but said this had not led to a substantial change in the level of completion of the tool.

Another pilot force told us the number of officers it had trained in the SST was low. And when we examined a proposed training product for new student officers in that force, we saw it did not even mention the SST. An audit by the same force in February 2023 showed that the SST had only been used in 20% of stalking offences in the initial pilot period.

The SST is a good way to identify stalking offences, so it was disappointing that the pilot forces had not used the SST in all relevant cases.

In our view, these findings suggest an insufficiently purposeful approach by leaders responsible for piloting the SST.

Forces sometimes do not record offences accurately

We found that 70 of the 530 cases we reviewed had not been recorded correctly in accordance with the crime recording rules. In 58 of the 70 cases, this was because stalking had not been correctly recorded.

In our police efficiency, effectiveness and legitimacy (PEEL) inspection programme between 2021 and 2022, 22 forces were subject to an examination of their crime data integrity (CDI). The inspection work found 477 reported offences that should have been recorded as stalking.

However, of these 477 offences, only 296 were correctly classified and recorded as stalking. Of the remaining 181 offences, 67 had been incorrectly recorded as other offences such as harassment. A total of 114 offences had not been recorded at all.

We concluded that, when stalking offences are reported to the police, about:

• 3 in 5 offences of stalking are classified and recorded correctly
• 3 in 20 offences of stalking are incorrectly recorded as other crimes, such as harassment
• 1 in 4 offences of stalking are not recorded at all.

Forces do not always recognise that some breaches of protective orders are also offences of stalking

We examined whether, and how well, the six fieldwork forces identified that some breaches of protective orders were also offences of stalking. We have described above which breaches of orders we examined in our fieldwork in the section ‘Terminology in this annex report’.

It is important that forces recognise that some breaches of protective orders by offenders can be evidence of a stalking offence. Otherwise, breaches of protective orders could wrongly be dealt with in isolation, and not as part of a wider pattern of offending. For those breach of protective order cases that end in prosecution, this may mean that the Crown Prosecution Service (CPS) and the courts do not have the full facts of the case and therefore might not have the appropriate sentencing powers or protective orders available to them.

This is reiterated in the crime recording rules, which state:

“Recordable breaches are state-based offences, so need to be recorded together with the most serious victim-based offence reported at the same time.”

The same guidance has the following information specifically in relation to stalking offences:

“NB. Any notifiable breach offences should be recorded in addition to any course of conduct offence.”

In our case file review, we found that none of the fieldwork forces were good at recognising that some breaches of protective orders were also offences of stalking.

The police had not recognised almost half (35 of the 72) of the breaches of protective orders cases as a further offence of stalking when they should have.

In one force, an audit of breach of protective order offences recorded (and conducted) in 2022 revealed significant problems in how the force was recording breaches of protective orders. In three separate audits which took place across separate months, on average 46% of cases were not recorded correctly – again, almost half. Offences that the police had not correctly recorded included stalking and other behavioural offences.

At the time of our fieldwork, the force did not seem to have acted on these findings. It had not carried out any more audits related to breaches of protective orders.

Previous HMICFRS recommendations about breaches of protective orders relevant to stalking

In our 2019 stalking report, we raised a cause of concern that police forces were dealing with breaches of protective orders in isolation and were not recognising or properly addressing the wider patterns of victimisation. We were concerned that, as a result, forces might not have been adequately assessing the risks to some victims, and might not have been appropriately investigating and prosecuting cases.

We made three recommendations:

• Within six months, chief constables should ensure that forces record stalking or harassment offences, if appropriate, when victims report breaches of protective orders.
• Within six months, the NPCC lead and the CPS lead should consider whether they can do more to inform police officers and lawyers of the importance of treating breaches of protective orders as evidence of a wider pattern of offending, and when and in what circumstances officers and lawyers should treat this as further evidence of stalking or harassment.
• Within six months, chief constables should ensure that officers are aware of the importance of treating breaches of protective orders, where appropriate, as part of a wider pattern of offending, and ensure that force policy and guidance help officers to do this.

In August 2019, the NPCC lead for stalking and harassment responded to the second recommendation by writing to chief constables asking forces to “now consider your own local procedures to ensure that breaches are being dealt with as per the recommendations made”.

We have reported above on our findings about the first recommendation. We also examined how the police service has responded to the second and third recommendations.

Force and College guidance about breaches of protective orders relevant to stalking

We think officers and staff need to know that breaches of protective orders could be a sign of increased risk to the victim. This is because, if an offender is willing to breach an order imposed by a court, it may show that they have a disregard for the law, and may mean they will reoffend against the victim.

Despite the importance of linking breaches of protective orders to the increased risks to stalking victims, and our earlier recommendation, this was only included in two of the six policies of the fieldwork forces and only adequately covered in two forces’ training provision.

At the time of our fieldwork for this super-complaint investigation, the College of Policing’s ‘Advice to police responders to ensure an effective response to reports of stalking or harassment’ did not contain any information about breaches of protective orders. Nor did the College of Policing stalking or harassment e-learning packages.

However, the College’s advice for investigators (PDF document) and advice for supervisors (PDF document) do contain guidance that when breaches of protective orders occur, the police should consider any further stalking or harassment offences.

The College of Policing will update its guidance for police on stalking. This will include amendments to emphasise the importance of police responders treating breaches of protective orders as potential further instances of stalking and also as significant escalations of offending.

Officers sometimes incorrectly record offences as malicious communications offences instead of stalking

In our 2019 stalking report we raised a cause of concern that forces might not be properly assessing the risks to victims and the likelihood of repeat offending, because they were recording some stalking cases as malicious communications offences rather than stalking offences. So victims may not be getting the support they need, and cases may not be treated as seriously as they should.

We recommended that the NPCC lead for stalking and harassment write to chief constables to ask them to review offences recorded as malicious communications. We wanted forces to check if they should also be recording them as stalking or harassment.

In August 2019, the NPCC lead for stalking and harassment responded to the recommendation by writing to chief constables. He asked forces to “consider undertaking a proportionate local audit to establish what the recording practice is locally within your Force area and put in place any further guidance where required”.

We know that some stalking offenders use phone networks to call or send messages to victims. Stalking offenders also use the internet. In our case file review, we found that 123 of the 384 stalking cases reviewed involved these forms of digital communications.

So it is important that officers and staff are alert to the fact that malicious communications could be stalking. It is also important that this fact is reinforced by force policies, training and guidance.

At the time of our fieldwork, the College of Policing stalking e-learning and the relevant authorised professional practice (APP) did not mention the importance of considering whether malicious communications offences are instead offences of stalking. The College of Policing is committed to developing its APP on stalking or harassment with consideration of the learning from this super-complaint investigation. It will ensure that content throughout the APP supports officers and staff to identify stalking when it occurs online, including through repeated and targeted messaging, and highlights the risk that these stalking behaviours may be misidentified as malicious communications.

In our fieldwork, only one of the six force stalking policies reinforced the importance of not confusing the recording of stalking with malicious communications. Five of the six forces included at least some reference to malicious communications offences in the stalking training, but most of the training could have been clearer about the impact of not recording this offending correctly.

We found that half of the six fieldwork forces had incorrectly recorded some cases as malicious communications instead of stalking. Of the 60 cases of malicious communications we examined:

• five should have been recorded as stalking
• an additional six should have been recorded as harassment
• one should have been recorded as coercive and controlling behaviour

We are concerned that, despite the recommendation we made in 2019, police forces still inaccurately record stalking offences as malicious communications.

Officers do not always identify the right type of stalking offence

It is important that when officers recognise stalking offences, they correctly record them as either an S2A or an S4A offence. This is because, as we say above in ‘Terminology in this annex report’, S4A offences are more serious and so the sentences that courts can impose are more serious and appropriate to the type of offending. And as we describe at ‘Failure to meet six-month time limit is leading to cases being dropped’, officers may incorrectly think that statutory time limits are relevant to the case.

Recording S2A offences when the evidence points to S4A offences may show that the recording or investigating officers have not conducted an effective investigation. It might also show they have not recognised the psychological impact or other effects on a victim. If, subsequently, the CPS does not identify that a case has been wrongly recorded as an S2A offence, this could mean that a suspect is prosecuted for a less serious offence. This, in turn, may have an impact on the later sentencing if the suspect pleads, or is found, guilty.

In our fieldwork, we found that all six forces had recorded some stalking offences as S2A when they should have been recorded as S4A offences.

Of the 190 S2A cases we examined in the case file review, 16 should have been recorded as S4A offences. Conversely, only one of the 194 S4A cases reviewed should have been recorded as S2A.

Of the 190 S2A cases that we examined, officers completed structured risk assessments in only 95. This is a worryingly low number and we have discussed this in the section on ‘Risk assessment’.

In 83 of these 95 cases that had a risk assessment, the person completing it had given it a risk grading. Ten were graded high risk, 38 were medium, and 35 were low. This shows that S2A cases often include evidence that the victims are at high risk. This may also show that, in these cases, the stalking offending is serious and so the offences have been incorrectly recorded. It is also possible that the risk grading has been influenced by the victims’ earlier experiences so this information, while important, is not necessarily evidence of poor crime recording decisions.

One force’s stalking analysis from November 2021 concluded that there was a large variation in risk of harm to victims within S2A offences. The analysis found that a high‑risk Domestic Abuse, Stalking, Harassment and Honour-Based Violence (DASH) assessment included multiple threats to kill as part of the course of conduct in a small sample of S2A stalking offences. The analysis concluded that this and other evidence were a sign that the offences should have been considered as S4A offences instead.

We have discussed in the section ‘Stalking investigations are often not effective enough’ that some officers do not recognise the psychological impact of stalking on victims. This may also mean that some officers do not make good decisions about what type of stalking offence to record and deal with.

Potential implications of misidentification by other criminal justice organisations

The super-complaint only relates to the impact of the misidentification of stalking by the police. For this reason, we have not reviewed the involvement of the CPS, the Probation Service or the courts in the criminal justice response to stalking.

Nevertheless, it is important to acknowledge the vital part all criminal justice agencies play in tackling stalking. It was for this reason that in 2017 we carried out the joint (with His Majesty’s Crown Prosecution Service Inspectorate) inspection of harassment and stalking.

If, for any reason, prosecutors do not identify stalking, and/or authorise different charges such as harassment, it could lead to unintended consequences and inappropriate outcomes.

We found one case in our case file review which we think highlights this risk. However, we do not know the full circumstances behind the CPS and court’s decision‑making rationale, and recognise that the charge could have been appropriate. We also acknowledge that in April 2023 the CPS published revised guidance helping prosecutors to identify the correct offence when course of conduct offending occurs. The offences we examined in our case file review pre-dated the publication of this guidance.

Flawed investigations and No Further Action decisions

Some forces have different ways of allocating stalking cases for investigation

It is important that the police make good decisions about what type, and what amount, of resources they allocate to investigate crimes. This is because officers with the right skills and experience should be matched to the needs of the investigation. It also means that the force can properly meet the needs of the victims and witnesses.

In 2019 we published our report on our joint inspection (with His Majesty’s Crown Prosecution Service Inspectorate) of the police and CPS response to crimes against older people. In this report, we raised a cause of concern that the systems of crime allocation used by police forces did not always consider the needs of victims and the complexities of cases.

We recommended that, by January 2020, the NPCC lead for crime investigation should work with other interested parties to review allocation processes. We also wanted the NPCC lead to recommend systems that more easily help forces to allocate an appropriate investigative response.

In response to this recommendation, the NPCC lead for investigations has told us that they have worked with the College of Policing, and the College has updated the investigation APP. This APP now makes specific reference to offences like stalking where consideration of risk of harm to victims, repeat victimisation, and recognition of any historical unreported previous incidents, should inform the allocation decision.

In our fieldwork we examined how the six forces allocated stalking offences for investigation. All the forces had policies that set out how this should happen, and which teams should investigate stalking offences. In our case file review, we considered how well each force had followed its policy.

We found that each force had different arrangements for how it allocated stalking offences.

One force had introduced a new crime allocation policy in June 2023. But the policy did not mention what team should investigate S2A stalking offences, despite these offences making up the bulk of recorded stalking offences in that force. The policy did say that S4A offences committed in a DA context should be allocated to a vulnerable adult team that had been set up to deal with DA offences. But the policy did not mention how officers and staff should allocate non-DA offences.

In another force, the policy was to allocate all non-DA stalking offences to frontline officers and to allocate DA stalking offences to domestic abuse safeguarding teams. But even in this force, one area had deviated from the policy. The local commander had decided that all stalking offences would be investigated by the domestic abuse safeguarding teams. Staff in the force told us this was because the local commander recognised the risks to non-DA stalking victims.

Two fieldwork forces had set up teams of officers – predominantly frontline or PIP 1 officers – to investigate the majority of offences with viable lines of enquiry. The force identified these investigative lines with a screening process. The intention was that these teams had experienced and better trained supervisors to help guide the investigators. It was envisaged that these teams would investigate most stalking offences.

Forces sometimes do not allocate stalking offences for investigation

In our case file review, we found some stalking cases that were closed without being given to an investigating officer and without any investigation taking place.

In one case the officer making the decision not to allocate the case noted: “I cannot justify tasking NPT (neighbourhood policing team) to deploy and speak to the suspect at this time because it would not be proportionate in current climate of demand & available resources as the risk is not there.”

Forces often allocate stalking crimes to frontline officers for investigation

Cases involving stalking can be serious and complex. These merit the attention of teams that include personnel with the training and skills to investigate serious and complex offences. These are sometimes known as ‘PIP 2 investigators’. Only one of the six fieldwork forces had a policy that all stalking offences should only be investigated by such teams. Despite this policy, in the force concerned, of the 67 stalking offences we examined, 18 had been allocated to a PIP 2 team. The other 49 offences had all been allocated to frontline officers, sometimes referred to as PIP 1 investigators.

In another force, the stalking policy said that: “the starting point for a complex stalking case should be a PIP 2 investigator”.

But despite this, in our case file review of the force concerned, we found 26 of the 32 S4A stalking cases (so the more serious offences, and in all probability the most complex) had been dealt with by frontline officers.

Most forces use risk of harm to the victim as the basis for crime allocation

Most of the fieldwork forces made decisions about which team investigated stalking offences based on the perceived risk of harm to the victim. These risk judgments came from what the forces called ‘risk assessments’. We discuss risk assessments in more detail in the section on ‘Risk assessment’.

In most fieldwork forces, for DA stalking cases, the policies set out that medium and high-risk cases should be investigated by specialist DA teams. The force policies explained that low-risk DA stalking cases should be allocated to frontline officers. The risk assessment tools used to support these allocation decisions were either the DASH or the Domestic Abuse Risk Assessment (DARA).

Non-DA stalking cases were allocated to either frontline officers and/or CID in most forces. Forces based the allocation decisions on what they described as an assessment of risk. It used an S-DASH question set. This is problematic, as the S‑DASH is not a risk assessment tool, as we discussed above in the ‘Risk assessment’ section. So this might mean bad decisions are made about which officer is best suited to investigate an offence.

More generally, using risk assessments in isolation to allocate crimes for investigation is also problematic. This is because the risk assessment only deals with the risks to the victim, and does not assess the complexity of the investigation.

For example, a victim at low risk of harm may report a very complex crime. The investigation could require specialist skills such as conducting a video recorded interview. But when allocating by risk, an investigation like this might be handed to a frontline officer. However, a frontline officer may not have the experience, skills or time to investigate the offence properly.

Conversely, an offence reported by a victim who has been assessed as at high risk of harm may be simple to investigate. But this offence, when allocated by risk, would be assigned to a specialist team.

In ‘Living in fear’, our 2017 report into the police and CPS response to harassment and stalking, we stated that the main factors that affect allocation of investigations are crime type, risk, complexity and availability of resource. When allocation is not based on all these factors, then victims may be let down.

Forces sometimes allocate stalking offences to numerous different investigating officers

We also found some evidence of forces passing cases consecutively to numerous investigating officers. While sometimes this is inevitable, it makes building a relationship with the victim more difficult, and could lead to the victim losing confidence in the police.

We say more about this in the section titled ‘Victim care’.

Stalking investigations are often not effective enough

In our super-complaint fieldwork, we considered how well forces had investigated 470 stalking and breach of protective order cases. We did not include an assessment of the investigation standards of malicious communications offences. This was because, in many cases, these investigations were not related to stalking.

Of the 470 stalking and breach of protective order cases reviewed, we concluded that in 164 the investigation was not good enough. This is concerning, but it is not a new finding. In our 2017 report ‘Living in fear’ and our 2019 stalking report, we drew similar conclusions.

In 122 of the 470 stalking and breach of protective order investigations – more than a quarter of them – we found the investigating officers had not pursued all reasonable lines of enquiry before closing the case. Examples included cases in which officers had failed to:

• take victim and/or witness statements
• conduct house-to-house enquiries to trace potential witnesses
• seize CCTV evidence

In most forces, supervisors are responsible for making sure that investigations have been completed to a satisfactory standard. We discuss in the section ‘The supervision of stalking investigations is not good enough’ the important role that supervisors play in the investigation of crimes.

We also found that officers did not always recognise the psychological impact of the stalking behaviour on the victim. We only found evidence that the investigating officer had recorded that this had been recognised in 129 of the 470 stalking and breach of protective order investigations we reviewed. This is concerning.

In our victim service assessments as part of our 2021 and 2022 PEEL inspection programme, we examined 133 stalking cases from 42 forces. We judged that the investigation had not been effective in 33 of these cases.

It is also important to note that the lack of effective investigation is not exclusive to stalking.

On 9 June 2023 we published ‘State of Policing: The Annual Assessment of Policing in England and Wales 2022’. In that report, we stated that forces were often not getting the basics of policing, such as investigating crime, right. The report said that in our PEEL inspections, responding to the public and investigating crime were still the worst-performing areas.

On far too many occasions in our fieldwork, we found examples – such as case study 6 – where the investigation was poor, and the victim had been let down by the police.

But we did find some examples of officers completing effective investigations. One example is included in case study 7.

In the following sections, we provide further detail about other ways in which forces could have improved some stalking investigations.

The police often do not arrest stalking suspects

In most stalking investigations, we would expect officers to check suspects’ accounts of their involvement with the victims. One way in which this can be done is to arrest the suspect and interview them at a police station.

The police can also decide that rather than arresting the suspect, their account can be gathered by inviting them to voluntarily attend a police station to be interviewed under caution. This is called voluntary attendance. It would not generally be appropriate to use voluntary attendance in stalking investigations because, without an arrest, it is not possible to place suspects on bail, with or without conditions.

Arresting a stalking suspect is not necessarily a prerequisite for an effective investigation. But we would expect that investigating officers would show in their investigation plans why this was not necessary, especially as part of their plans to safeguard victims.

Of the 470 stalking and breach of protective order cases we examined, the suspect was arrested in only 161 cases. In the cases where no arrest was made, we found few cases where there was any recorded rationale for the lack of arrest.

If a victim decides not to support a prosecution, police and prosecutors should consider whether it is possible to bring a prosecution without that support. This is called an evidence-led prosecution When the police do not make an arrest, and they do not seek an offender’s account by voluntary attendance, it is unlikely that the police could pursue an evidence-led prosecution.

In July 2023, one fieldwork force had audited 102 stalking offences. The force’s analysis concluded that:

“In relation to arresting and interviewing suspects only 23% of suspects are being questioned by police. 77% of suspects are not being arrested and therefore not subject of bail conditions which is not likely to give the victim the confidence in the police in the future.”

The police often do not secure evidence from digital devices

In one fieldwork force, we were told that in the preceding year, the digital forensic unit had only received 20 requests for digital examination of devices in stalking cases.

In our case file review, we found there was a lack of submissions for digital forensic analysis and evidence retrieval. Of the 384 stalking cases, there were 123 applicable cases. But we found that the investigating officer had only sent the digital devices for analysis in eight of these cases.

There are sometimes avoidable delays to investigations

The negative effect on victims of delayed investigations is significant. Delays may lead to increased victim anxiety and repeat victimisation, and/or may lead to the victim becoming disillusioned and deciding not to support the investigation.

In our case file review, we found some investigations were subject to delays. But in 22% (103) of the 470 stalking and breach of protective order cases we reviewed, we considered that the delays were avoidable. This was for a variety of reasons, including officer leave and sickness. Some of these delays were caused by waiting for digital devices to be examined and evidence retrieved. We have discussed this further in the section titled ‘Digital forensic analysis’.

As part of our PEEL 2021/22 victim service assessments, we judged that there had been a delay to the investigation in 24 of 133 relevant stalking cases, and in only 7 of those 24 cases was the delay justifiable.

The police are still not using the power of search often enough

It is important that investigating officers use, where appropriate, all the available powers of search when investigating stalking offences. These powers allow the police to secure as much evidence as possible and to build better cases. This makes a successful outcome for the victim more likely.

As we detail in the section ‘Terminology in this annex report’, S2A offences are summary only offences and can only be heard in a magistrates’ court. The law does not ordinarily allow officers to search premises for evidence when they investigate such offences. However, the 2012 stalking legislation introduced a provision to allow officers investigating stalking offences under S2A to apply to a magistrate for a warrant to enter and search premises for evidence.

In our previous stalking inspections in 2017 and 2019, we saw very little evidence of officers using this power. Because of this, in our 2019 stalking report we raised a cause of concern. We recommended that by 31 July 2019 the NPCC stalking and harassment lead should remind chief constables to make sure that officers know about the powers of entry and search for stalking, and use them appropriately.

On 26 August 2019, the NPCC stalking and harassment lead acted on this recommendation and wrote to chief constables.

But despite this, in our case file review, we again found little evidence of the use of the power of search to gather evidence in stalking cases. In the 384 stalking cases we examined, we found evidence of searches of premises in 19 cases.

In July 2023, one fieldwork force audited 102 stalking offences. The force’s analysis concluded that: “There were no searches carried out in any of the cases in the audit with no rationale as to why a search was not necessary.”

The quality of investigations is better when investigators deal with cases

We have discussed in the section titled ‘Some forces have different ways of allocating stalking cases for investigation’ how the fieldwork forces had different procedures for allocating stalking cases for investigation.

In our case file review, we examined whether cases had been investigated by frontline officers or by investigators.

We found that, in general, stalking cases were handled less well when investigated by frontline officers.

Of the 67 S2A offence investigations that we concluded were not good enough, 49 had been dealt with by frontline officers.

Of the 73 S4A offence investigations that we concluded were not good enough, 54 had been dealt with by frontline officers.

Again, this is not a new finding. In our report about the police and CPS response to crimes against older people, we found that offences investigated by frontline officers had been dealt with less well overall than those handled by investigators. We concluded that one reason for this could have been that frontline officers had not received enough training to investigate more complex offences.

In our super-complaint investigation, one fieldwork force’s force management statement concluded that:

“Due to the lack of PIP 2 investigators, PIP 1 investigators are taking on DA cases which is not desirable for the level of complexity and quality required for these cases.”

Problems with investigative capacity are not unique to stalking. In our August 2022 report‘The police response to burglary, robbery and other acquisitive crime – Finding time for crime’ ‘The police response to burglary, robbery and other acquisitive crime – Finding time for crime’, we said that forces lacked the investigative capacity and capability to effectively deal with serious acquisitive crime.

In ‘State of Policing: The Annual Assessment of Policing in England and Wales 2022’‘State of Policing: The Annual Assessment of Policing in England and Wales 2022’, we said that forces were struggling to cope with a chronic shortage of experienced detectives.

With this background, we do not think it is a realistic expectation that PIP 2 investigators should deal with all cases of stalking.

The supervision of stalking investigations is not good enough

We have previously reported that, generally, too many police investigations have inadequate supervision. In our 2022 report ‘Finding time for crime’, we observed that in a third of all cases we examined in our PEEL inspections, there was not enough evidence of good-quality and intrusive supervision of investigation plans. Supervisory updates were sometimes out of date, and lacked detail and scrutiny. We recommended that forces improve crime scene management and levels of supervision to make sure that they investigate effectively.

While poor supervisory standards are never acceptable, the ramifications for victims of this in stalking cases can be far more serious than in other offences.

In our 2017 report ‘Living in fear’, we said that an examination of prosecution case files indicated that the supervision of stalking investigations was not as effective as it should be.

In our PEEL 2021/22 victim service assessments, we only found evidence of effective supervision in 80 of 120 applicable stalking cases. In those assessments, effective supervision means directing and advising the investigator, and overseeing actions.

In our case file review, we found that supervision of investigative actions was not good enough in 191 of the 470 stalking and breach of protective order investigations.

On too many occasions, we found examples of poor supervision or a lack of it. One such example is set out in case study 8.

We were, however, pleased to find some good examples of active and effective supervision.

Some stalking investigation outcomes are inappropriate

In the investigation summary report, the data shows how few stalking investigations result in offenders being charged.

In our fieldwork we found some evidence that the recorded outcomes for some investigations were inappropriate. We have included some examples of these inappropriate outcomes elsewhere in this annex report, such as case study 6.

In March 2023, one fieldwork force conducted an analysis of 50 stalking reports where an investigation had been closed, and a suspect had not been charged. The analysis showed that in 16% of the cases, the case was concluded by an officer giving the suspect ‘words of advice’. In July 2023, another analysis exercise in the same force found similar. The force realised this practice was unacceptable, and it changed its procedures to prevent this.

Another fieldwork force told us that since 1 September 2021 it had given conditional cautions to 140 stalking offenders. But the force told us that of these offenders, 34 had not complied with the conditions of the cautions. In these circumstances, the CPS ‘Director’s Guidance on Charging’ says that cases should be referred to the CPS for a prosecution decision. But the force only did this in 11 of the 34 cases.

Sometimes the police do not refer outcome decisions to the CPS

As stated above, there are specific arrangements prescribed by the Director of Public Prosecutions about charging decisions. The CPS ‘Director’s Guidance on Charging’ makes it clear that only the CPS can make decisions about charging in stalking cases.

It is important that the police send cases to the CPS for charging decisions. This is because it allows the CPS to consider the right charge, taking into account the seriousness of the offending and the needs of the victim. In our case file review, we found the police had charged some cases without reference to the CPS.

We found in our previous stalking inspections in 2017 and 2019 that the police sometimes incorrectly charge offences that should be referred to the CPS. This can cause problems for the CPS as it may need to change the charge if it considers other charges are more appropriate. It may also mean that reasonable lines of enquiry have not been fully investigated; for example, from an S2A to an S4A stalking offence or vice versa. It is also confusing for victims and may lead to delays in the criminal justice process.

The police sometimes do not record the outcomes of crimes correctly

When the police conclude an investigation, they assign a crime outcome type to the crime record. The crime recording rules specify what crime outcome types the police should use and in what circumstances.

We think it is important that the police use the correct crime outcome types as set out in the crime recording rules. One reason is because it makes it easier for forces to understand what improvements are needed to make sure they are not preventing victims from accessing justice.

In July 2023 we published our PEEL spotlight report ‘Police performance: Getting a grip’. In this we concluded that forces must get better at reviewing their use of crime outcome types. Then they can assure themselves that they are correctly applying outcome types. And if they are not, forces can work out why and address any problems.

In our PEEL 2021/22 victim service assessments, we found that the police had applied an incorrect outcome in 21 of the 114 relevant stalking cases we reviewed.

In our case file review, we concluded that the police had assigned the wrong crime outcome type in 49 of the 530 cases we examined. This shows that forces still need to improve the accuracy of how they record crime outcome types.

We also discuss crime outcome types and give some other examples of incorrect recording in the section ‘Failure to meet six-month time limit is leading to cases being dropped’.

The CPS sometimes have a different view to the police about stalking cases

The CPS guidance for prosecutors on stalking or harassment states:

“It is also important that breaches of protective orders are carefully considered, as new offences may also have been committed in addition to the breach of the order. Prosecutors must review all new offences to assess whether a prosecution should follow for any new offence as well as the breach of the order.”

In our case file review, we found one case where the CPS did not authorise a stalking charge where there had been multiple breaches of protective orders. However, as we say above in the section ‘Potential implications of misidentification by other criminal justice organisations’, we did not speak to CPS prosecutors to understand the decision-making rationale. Therefore we cannot say whether a stalking charge would have been appropriate or inappropriate.

The CPS guidance for prosecutors on stalking or harassment also states:

“If there is sufficient evidence for the elements of stalking involving fear of violence or serious distress or substantial adverse effect on the victim’s usual day-to-day activities, the offence contrary to section 4A Protection from Harassment Act 1997 may be the best offence to charge to reflect the seriousness and extent of the offending, give the court adequate powers to sentence and enable the case to be presented in a clear and simple way. The non-exhaustive list in section 2A (3) Protection from Harassment Act 1997 provides examples which can be stalking. These may overlap with CCB [controlling or coercive behaviour] or harassment. However, behaviours which show a fixation, are obsessive, unwanted, and repeated might most clearly be presented to a court as stalking rather than harassment or CCB.”

We found examples where the police had requested a charging decision from the CPS for harassment where they should have requested a decision for a stalking charge. We also found one case where the CPS authorised a harassment charge where the police thought a stalking charge might have been appropriate. However, we do not know the full circumstances behind the CPS decision-making rationale. Therefore we cannot say whether a stalking charge would have been appropriate or inappropriate.

The failure to offer or refusal to apply for an interim or full Stalking Protection Order

Fieldwork forces had made very few successful applications for interim or full Stalking Protection Orders in the year 2022/23

In our fieldwork, we asked forces to tell us how many times they had successfully applied for Stalking Protection Orders (SPOs) in the year ending 31 December 2023. We found that five fieldwork forces had collectively only made 14 successful applications for SPOs during that period. One force had not successfully applied for any SPOs in that period.

We think this collective total is unacceptably low. It also reflects the very low SPO applications granted rate across all forces in England and Wales. The low number of successful applications for SPOs means that some victims may not have got the protection they needed.

In addition to the successful applications, the forces had made a small number of applications for SPOs which the courts had not granted. We have described some of the reasons for these unsuccessful applications below.

We were surprised and disappointed to find that not all the fieldwork forces had good systems for gathering information about SPOs. We expected to find good systems for monitoring how many SPOs they had applied for, how many had been successful and how many had not been successful (with the reasons).

In some cases, the fieldwork force stalking leads did not have the information we asked for readily to hand. In one case, the force had to ask its own legal department for this information. If force stalking leads do not have a good understanding of how the force is performing in relation to SPOs, it is difficult for the force to identify missed opportunities to protect victims.

We discuss the systems and processes that the fieldwork forces used for considering the necessity for, and applying for, SPOs in the sections below.

The police are failing to routinely consider whether a SPO is needed

In our case file review, we examined whether there was evidence to show that the police had considered applying for a SPO. Of the 384 stalking cases reviewed, we could only find evidence that the police had considered a SPO in 16 cases.

In our case file review, we found many examples where the police should have considered a SPO. One example is case study 13.

Linked to our findings on inadequate supervision, we found that supervisors were routinely failing to check that officers had considered applying for SPOs in stalking cases. We also found that senior officers in the fieldwork forces had not made sure SPOs were applied for in cases where they were needed.

There appear to be wider issues with the police use of protective orders. In our August 2024 report on progress to introduce a national operating model for rape and other serious sexual offence investigations in early adopter forces, we found that the police could use protection orders more effectively to safeguard victims.

The police often do not tell victims about SPOs

Of the 384 stalking cases we examined in our case file review, we only found evidence that the police had told the victim about SPOs in six cases.

It is possible that officers had told the victim about SPOs, but not recorded it. Based on other findings set out in this section, we think it is much more likely that officers themselves had not considered SPOs in stalking cases. And so they had not told the victims about them.

Not all investigating officers have received training on SPOs

All the fieldwork forces told us that they had provided some awareness training about SPOs to frontline and investigating officers. Sometimes this was as part of wider stalking training. From the records forces gave us, we could see that forces had not trained all relevant officers on SPOs. But some forces could not provide us with accurate records of the numbers of officers trained.

In one force, the SPO training material referred to SPOs incorrectly as “stalking prevention orders”. In our fieldwork interviews, some officers we spoke to made the same mistake. These things show that the police have not made sure SPOs are well understood in operational practice.

Some forces have inadequate systems for identifying opportunities to apply for SPOs

In all the fieldwork forces, the primary responsibility for considering whether a case was suitable for a SPO application lay with the investigating officers. In those forces, many investigating officers had received no specific guidance or training on SPOs.

In some fieldwork forces, the responsibilities of dedicated stalking officers and staff included identifying cases where a SPO may be appropriate. In those forces, the dedicated stalking officers and staff placed information on the crime reports to alert the investigating officers to this possibility. But the dedicated stalking officers and staff did not follow up to make sure investigating officers had considered a SPO. Some dedicated stalking officers and staff told us this was because they did not have the time to do this check, because of other work commitments.

One fieldwork force had introduced guidance which said that officers should only consider applying for a SPO when the suspect lived inside the force area. This is not the only requirement contained in the Home Office ‘Stalking Protection Orders: Statutory guidance for the police’. The guidance says that the police can also apply for SPOs “in respect of a person (respondent) who resides in the chief officer’s police area, or who the chief officer believes is in that area or is intending to come to it”. This means the force may have left some victims without appropriate protection.

Systems for making SPO applications are ineffective and inefficient

If an investigating officer has identified that a case may be suitable for a SPO, an application process follows. The Home Office and the College of Policing provide guidance about how to do this:

• Home Office ‘Stalking Protection Orders: Statutory guidance for the police’
• College of Policing Stalking or harassment APP

All the fieldwork forces had a process whereby investigating officers had to prepare SPO applications. This was in addition to investigating the case with a view to criminal proceedings. We think that forces have introduced this way of working because it is easier for an investigating officer, with existing knowledge of the case, to also compile a SPO application.

But we think that it is possible that investigating officers may prioritise the criminal investigation over the SPO application. This is especially so as officers and staff we spoke to told us that they thought the SPO application process was complicated and time-consuming.

We also think that competing demands on investigating officers’ time, for example from other cases, may lead to either delays in SPO applications or the applications not being made at all.

Also, at the time of our inspection, some investigating officers had not received any training on SPOs. So it is possible that some officers did not have the knowledge or confidence to make SPO applications.

The Home Office statutory guidance on SPOs states that a police superintendent’s authority is needed to make an application at court. One fieldwork force had a process whereby officers sent potential SPO applications to the force legal department for checking, before seeking authority to go ahead from a superintendent. We were told that this was so that the legal department could advise the superintendent on the likelihood of the application succeeding at court. The force saw this as a pragmatic solution to the problems it had previously met securing SPOs.

Forces use legal departments to make SPO applications

All the fieldwork forces we visited used legal departments to make SPO applications. This was mainly because their legal departments also took responsibility for applying for other protective orders, such as Domestic Violence Protection Orders (DVPOs), Sexual Harm Prevention Orders (SHPOs) and Sexual Offences Prevention Orders.

In one fieldwork force, a senior officer told us that they had become frustrated with their legal department. They viewed it as unwilling to make applications unless it was sure that they would be successful. One example they gave us was that the force lawyers would sometimes adopt a ‘wait and see’ approach instead of applying for a SPO. The senior officer told us this happened particularly when a criminal case was ongoing, and it was possible that a court would impose a restraining order on the offender.

In our case file review, we also found one case in each of two forces where a force legal department wanted to wait for the outcome of criminal proceedings before considering a SPO application.

Our findings about other types of orders were similar. In our August 2024 report on progress to introduce a national operating model for rape and other serious sexual offence investigations in early adopter forces, we looked at how forces applied for SHPOs. We found that force legal teams are in charge of applying for SHPOs. But some investigators said legal teams do not ask for interim SHPOs often enough. Instead, they said lawyers often delay applying for SHPOs until court cases are over. This could put some victims at risk.

In the same inspection, we found that some forces contract legal services to external providers. Investigators told us this makes it more difficult to challenge things that are not working.

At the time of our fieldwork, some forces had recognised the need to improve their processes for SPO applications, and particularly the liaison between officers and legal departments. These forces were considering introducing small ‘ancillary orders teams’. These teams would provide a dedicated resource that could help investigators with the application process for all protective orders, including SPOs.

One fieldwork force told us that it had recognised that it did not have enough resources in its legal department to cope with the volume of applications for protective orders. So the force was intending to recruit dedicated ‘vulnerability lawyers’ to specialise in such applications.

Some force lawyers in some fieldwork forces told us that if the police increased the number of SPO applications, they would struggle to cope with this extra demand. They told us that without extra resources, this would lead to delays in applications.

In some cases, officers advise victims to take out Non-molestation Orders instead of pursuing SPOs

In our fieldwork interviews, victim support workers told us that officers sometimes advise victims to seek Non-molestation Orders (NMOs). In certain circumstances, victims may have to pay for these. We also found evidence of this in our case file review.

In one fieldwork force, we found evidence that more than one officer had recommended that victims seek help with NMOs from a national organisation that operated for profit. In another force, we found evidence that officers had advised victims to contact a domestic abuse service different from the one commissioned by the police and crime commissioner (PCC). This was an inefficient use of the service provided by the PCC.

We believe the practice of recommending that victims seek NMOs in stalking cases when a SPO application is a possibility is unethical and unacceptable. This is because:

• NMOs are civil orders, and, unlike SPOs, a breach of the order is not a criminal offence. An NMO is therefore less powerful than a SPO.
• Some victims will not qualify for legal aid to help with an NMO application and so will have to pay for the application themselves.
• It inappropriately places the burden of responsibility on victims to protect themselves when the police should do this.
• It sends the wrong message to suspects, who may think the police are not prepared to protect victims.

Even when we found no evidence that the police had recommended that the victim apply for an NMO, in some cases we could see that the victim had decided to do this themselves. We do not know why victims decided to do this. But it could be because the victims thought that the police had failed to protect them.

Problems with the SPO application process prevent the police from securing more SPOs

The problems that we have described above are mainly the result of the way that the police have introduced SPOs into police practice. But there are other problems that are preventing more victims being protected by SPOs.

In our fieldwork interviews, officers told us that the SPO application process is complicated and time-consuming. Officers told us that in some respects, the application process had turned into a quasi-criminal trial. They also said the need to meet the criminal burden of proof in applications for full SPOs was a barrier to making SPO applications.

Officers also told us that disclosure of evidence can be a problem in applications for interim SPOs. The intention of interim SPOs is that they are applied for in the early stages of a criminal investigation, soon after the police receive a stalking report. In such cases, the police may not have interviewed the suspect, and they may not want to disclose to the suspect all the information they have about the case. But the SPO application process is based on the police disclosing information to the court (and the suspect) on the reasons for (and evidence about) why a SPO is appropriate. Officers told us that, from their experience, courts were unlikely to consider whether a SPO was appropriate unless police gave them all the relevant information.

The Home Office SPO statutory guidance makes no provision for the hearing of evidence in private by the court and without disclosure to the suspect.

One force told us that its local magistrates had previously been unwilling to approve applications for SPOs when the suspect was on police-imposed conditional bail. This was because the judges’ views were that in these circumstances SPOs were not necessary. They believed the bail conditions already protected the victims. Because of the court’s view, the force did not consider applying for SPOs in such cases.

Officers told us that while the statutory guidance does not bar applications for SPOs after a suspect’s trial, as the CPS would usually apply for a restraining order, there would be little point in the police also applying for a SPO.

Some officers told us that, unlike Domestic Violence Protection Orders, SPO applications needed the support of the victim to be successful. This is contrary to the SPO statutory guidance which states that the court may consider hearsay evidence. In this context, hearsay evidence means evidence not given in-person by the victim about what the victim had seen, heard or experienced. Officers told us that in their experience this was unlikely to be acceptable to the court. For various reasons, stalking victims sometimes withdraw their support for investigations. Officers told us that when this happens, the police are unlikely to consider applying for SPOs.

The SPO statutory guidance about hearsay evidence is another example of how the SPO application process can be complex and time-consuming. The guidance states that if the police wish to rely on hearsay evidence, and this is contested by the respondent, the police must serve the evidence on the respondent at least 21 days before the hearing.

It is then possible that the court will decide that it will not accept the hearsay evidence and it wants to hear the evidence in person. The statutory guidance makes provision for victims and witnesses to give their evidence by way of special measures (see ‘Terminology in this annex report’) and the court will consider applications by the police for this to happen. This is yet another example of the parallels between the SPO application and criminal trials processes.

The lack of police response to breaches of protective orders

We have discussed in the section on ‘The misidentification of stalking’ what we found about how well forces identify whether breaches of protective orders are also stalking offences. We also discussed how well forces identified the elevated risks to victims when protective orders are breached. Aside from the importance of dealing effectively with breaches of protective orders for these reasons, good police responses demonstrate support for victims and can keep victims safe.

Some forces do not know enough about their response to protective orders

In our fieldwork, we asked forces to give us information about their response to breaches of protective orders because breaches of these orders can also be offences of stalking. Some forces did not routinely monitor their performance on this.

Two fieldwork forces told us that they had acted to improve their understanding of the problem. In June 2023, the vulnerability board in one force had decided to include information about the response times to reports of breaches of civil orders in the force‑wide data analysis. In July 2023, another fieldwork force had audited its response to protective orders because of an action in its vulnerability improvement plan.

A different fieldwork force only monitored its performance about breaches of Domestic Violence Protection Notices (DVPNs) and Orders (DVPOs), but did not include other breaches, such as of restraining orders. We found this surprising, given that breaches of restraining orders are more serious than breaches of DVPNs and DVPOs. Courts can impose restraining orders after criminal trials. If breached, restraining orders are subject to more serious sentences than DVPNs and DVPOs.

Some forces do not have clear policies on which teams should investigate breaches of protective orders

It is important that forces have clear policies about which teams should deal with breaches of protective orders. One reason for this is because some breaches can also be offences of stalking.

The force crime allocation policies we examined in our fieldwork often did not include material about breach of protective order offences. We found that forces’ general presumptions were that response officers would deal with breaches of protective orders. This was because forces thought these offences were simple and straightforward and could be dealt with quickly. In some cases that will be true. But in other cases, as we have shown in the section on ‘The misidentification of stalking’, breaches of protective orders may be part of wider stalking behaviour. They can be part of a complex pattern of offending, and the victim may be at more serious risk of harm. In these cases, we consider it would be more appropriate for investigators to deal with the breach of order offences.

We have discussed in our section titled ‘Flawed investigations and No Further Action decisions’ how forces need better ways to allocate crimes for investigation.

Forces do not always give officers information about dealing comprehensively with breaches of protective orders

In our fieldwork, we found that force stalking policies and training often did not include information on the importance of officers dealing comprehensively with breaches of protective orders. In one force, neither the stalking training provided to control room officers and staff, nor that provided to student officers included this important information.

In the fieldwork forces, we did not examine all the policies and training that might have contained this information, for example more general DA training. So it is possible that officers and staff received this information in other ways.

Officers do not always arrest offenders when they breach a protective order

In our case file review, we considered how often forces arrested a suspect after a report of a protective order breach. Of the 72 breach of order cases we looked at, we found that the offender was only arrested in 39 cases. That is just over a half of cases.

We think this proportion of arrests is worryingly low. This is because it suggests that the police may not be treating breaches of protective orders as seriously as they should. And the police may not be protecting as many victims as they should. It also sends the wrong message to suspects: that the police may not always take firm action in breach of order cases.

There are sometimes delays when the police deal with breaches of protective orders

One fieldwork force had recognised that its response to breaches of protective orders needed to improve. It changed its policy about how quickly to send officers to calls for service. This meant that officers should respond to all such calls within the response time of eight hours. Previously the response time had been more flexible.

In our case file review, we considered whether the police response time was acceptable when a breach of a protective order was reported. In 14 of the 72 relevant cases we examined, we concluded that the police could have avoided delays.

Court action on breaches of protective orders might influence police action

In this investigation, we did not consider the wider criminal justice system response to breaches of protective orders. But we did hear in one force that the approach of courts in response to breaches of orders may influence officers’ decisions about how to deal with these cases.

One senior officer told us that officers were sometimes frustrated by the courts’ apparent lack of comprehensive action on breaches of protective orders. The force was concerned that the courts sometimes dealt with the breaches leniently, for example with small fines for the offender.

The senior officer told us that if officers perceived that the courts were not dealing with breach offences robustly, it might mean that officers were less likely to arrest and charge offenders.

The force was so concerned about the perceived lack of comprehensive action by the courts that it collected information about the sentences. It did this so that it could use the information and bring it to the attention of the courts.

Other issues of concern

Other issues of concern set out in the super-complaint include:

• lack of referrals to specialist services
• officers providing dangerous or unhelpful advice
• failure to meet the six-month time frame leading to cases being dropped
• the lack of provision of stalking intervention programmes

Lack of referrals by the police to specialist stalking victim support services

The Code of Practice for Victims of Crime in England and Wales (Victims’ Code) outlines that victims have the right to be referred to services that support victims. And, if eligible, they have the right to be offered a referral to specialist support services and to be told about other support available. We discuss the Victims’ Code in more detail in the section titled ‘Victim care’. Victim support services provide general support to victims of crime, such as reassurance and practical crime prevention advice, but may not have sufficient training or expertise to understand fully the negative effect of stalking behaviours on victims.

Specialist stalking victim support workers, where available, should have received specific stalking training. So in some cases they may be better placed to identify the risks to stalking victims, especially of any escalation of behaviour by the perpetrator. And they may be better placed to help victims to make effective safety plans.

In our report ‘Living in fear’, we found no evidence that the victim was referred to specialist stalking victim support services in 55 of the 112 cases we reviewed.

Our 2019 stalking report included an update on ‘Living in fear’. In that report, we found that the number of referrals to a specialist stalking victim support service was low compared to the number of recorded stalking offences in the year before the inspection. We recorded that an area for improvement for the force was to review the process by which it referred victims to specialist stalking victim support services.

The police are still not referring stalking victims to specialist stalking victim support services when they should

In our case file review, we looked to see if forces had referred each stalking victim to a specialist stalking victim support service. Of the 384 stalking cases we reviewed, we found evidence that the police had done this in only 63 cases. Two force areas did not have any specialist stalking victim support services for non-DA stalking victims.

It is possible that the police may have referred stalking victims to specialist stalking victim support services in other ways; for example by the general victim support services that exist in each area, or in DA cases by advocacy services and through multi-agency information sharing arrangements such as a multi-agency safeguarding hub.

It is also possible that victims may have contacted specialist stalking victim support services themselves.

The police do not always tell victims about specialist stalking victim support services

We also found evidence that officers do not always tell stalking victims about specialist stalking victim support services. Officers recorded that they had given the victim information about these services in only 100 of the 384 stalking cases we reviewed. It is possible an officer had informed the victim about specialist stalking victim support but had not recorded this on the crime record. We discuss victim care more generally in the section titled ‘Victim care’.

A February 2023 audit of 40 stalking offences in one fieldwork force showed that the force could only be sure an officer had provided the victim with details of victim support services in 21 cases.

Police force areas have different arrangements for specialist stalking victim support and some have no services for non-DA stalking victims

Since 2014, PCCs (or their mayoral equivalents) have been responsible for making sure there are enough support services for victims.

The right support for victims of crime is of paramount importance to ensure their continued safety and co-operation. In 2016 the Victims Commissioner’s report ‘What works in supporting victims of crime’ highlighted the importance of a trained and professional single point of contact for victims. The report said those providing support services should have enough knowledge of the criminal justice system. They should also provide emotional support and timely and accurate information for victims.

In our fieldwork forces we found different arrangements for the provision of specialist stalking victim support services. In all six forces there were services that supported DA stalking victims. In one force area, the DA stalking provision did not cover the whole of the force area. This was because of different arrangements in different local authority areas within the force boundaries.

Two of the six forces did not have any local provision of support for non-DA stalking victims. In one of those forces, officers told us the force policy was to refer victims to the national stalking support service Paladin. But the force could not tell us how many times this had happened. In the other force, staff members from the PCC’s office told us there was no support for non-DA stalking victims because Ministry of Justice funding had been “ring-fenced” for DA victims only.

Two of the six forces had a single provider for both DA and non-DA victims. In two forces, while there was specialist stalking support provision for both DA and non-DA stalking victims, different organisations provided the services.

In one force the specialist stalking victim support services provider told us that due to the volume of stalking victims, it only allocated resources to high-risk victims.

Some victims of stalking may not even be referred to general victim support services

In forces where specialist stalking victim support services do not exist for non-DA stalking victims, the local general victim support services should support these victims. For this to work effectively, the police need good systems for making referrals.

There are two types of general victim support referral mechanisms in forces:

• Opt-out – the force automatically passes victims’ details to general victim support services unless victims specifically state that they do not want this to happen.
• Opt-in – the force only refers victims if they specifically consent to the passing of their information to the general victim support services.

In 178 of the 384 stalking cases we examined as part of our case file review, we could not find evidence that the police had referred the victim to general victim support services.

We consider this to be unacceptable. And – regrettably – it is consistent with our findings from previous inspections.

In our 2018 hate crime report ‘Understanding the difference: the initial police response to hate crime’, we highlighted that some processes were more effective at referring victims to support services than others. We also said that some forces sometimes did not refer victims when they should have.

In that report, we concluded that there was a significant difference in the referral rate according to which system was used. Victims living in an area with opt-out systems were more likely to receive support services. This is because there is less opportunity for officers to make mistakes or omissions.

In our 2019 report on crimes against older people, ‘The poor relation: The police and Crown Prosecution Service’s response to crimes against older people’, we again concluded that the police were not appropriately referring some victims to support services.

In our fieldwork for this investigation, we found that errors and omissions in referrals included officers not ticking the right box on crime reports. Or they might tick a box on a linked offence (such as criminal damage) but not on the associated stalking offence. If an officer does not tick a box for an associated stalking offence, the victim support service would not be aware that the victim had also been the victim of a stalking offence. So they may not adapt their service accordingly.

In 2023, the Home Office introduced new crime recording rules. The resulting changes mean that the police do not have to record other, less serious offences taking place at the same time as the stalking offence. We think that this should have a positive effect on this aspect of the problem of general victim support referrals. This is because the reporting officers do not have to record the need for general victim support services on multiple offences.

Force policies and training often do not give officers and staff enough information about specialist stalking victim support services

In our fieldwork, we examined force policies and training documents to see what information forces gave officers and staff about specialist stalking victim support services.

Only one of the six forces’ policies included specific reference to the importance of referring victims to specialist stalking victim support services, and included information about the relevant organisation. Much of the force stalking training material that we examined did not have this information.

Some forces listed details of the national stalking support organisations but did not list details of their local specialist stalking victim support services.

One force had an internet information web page called ‘Support for victims of stalking and harassment’. However, this did not include any reference to the local stalking support service. But it did list the national specialist stalking victim support services.

Force systems to make sure officers refer victims to specialist stalking support services are poor

We found that (unlike the often-automatic referrals for general victim support) forces’ referral processes for specialist stalking victim support often relied on officers emailing or calling the support service. This relies on officers (and supervisors) being aware of the processes involved. We have detailed above how this information is not always readily available.

None of the fieldwork forces had good governance processes to make sure officers made referrals appropriately in stalking cases. Similarly, none of the fieldwork forces had information readily available to show how often they had made specialist referrals. So forces could not tell whether the referral system was working well, or whether there were problems.

Where forces had dedicated stalking officers and staff, these officers could sometimes find cases where investigating officers had missed opportunities for referrals. In these cases, we saw evidence that the dedicated stalking officers and staff had reminded investigating officers to do this. But the dedicated stalking officers and staff told us they did not have the capacity to check if the investigating officer had then made a referral.

Supervisors have a responsibility to make sure that investigating officers refer victims to specialist stalking victim support services. But we have described in the section titled ‘Flawed investigations and No Further Action decisions’ how supervision is often not good enough in stalking cases.

Specialist stalking victim support services sometimes operate in different ways

In our fieldwork we did not speak to specialist stalking victim support workers in all the forces we visited. We sometimes spoke to staff employed by the PCCs instead. This was because forces told us they thought this was more appropriate as PCCs commission the specialist stalking victim support services.

In one force area, the specialist stalking victim support workers for non-DA stalking victims told us they had received specialist stalking training, but that the specialist stalking victim support workers for DA cases had not. In the same force area, support workers for non-DA stalking victims completed different risk assessments from support workers for DA stalking victims.

In another force area, we found that the specialist stalking victim support workers completed an S-DASH form for victims instead of completing a risk assessment. As we describe in the section on ‘Risk assessment’, the S-DASH is not a risk assessment. In some other force areas, the support workers completed an enhanced risk assessment called a screening assessment for stalking and harassment (SASH).

Partnership working is often good, but there are sometimes exceptions

In our fieldwork we found that specialist stalking victim support workers and the police dedicated stalking officers and staff had good working relationships. They worked well together to support victims and keep them safe.

But some specialist stalking victim support workers told us they sometimes found it difficult to work with investigating officers. This was sometimes because the investigating officers were not available. In these circumstances, the support workers told us that the dedicated stalking officers and staff were valuable contacts. They supplied information and/or passed messages to the investigating officers.

In one force, the specialist stalking victim support workers told us that they had found some investigating officers to be obstructive and unwilling to share information. The support workers told us they thought this was because the force had not told its officers about support workers and their role.

In the same force, the support workers told us that as the force had no dedicated stalking officers or staff, they had no point of contact to tell them about a case and identify the investigating officer. The support workers had to call 101 and ask the control room for information about the investigating officer. The support workers then had to try and contact the investigating officer, which they sometimes found difficult.

Information sharing arrangements in stalking cases can be improved

Specialist stalking victim support workers told us that when the police referred cases to them, the police most often supplied them with a copy of the crime report and the relevant risk assessment, if the investigating officer had completed one.

The support workers told us that if they were given more details about the victim and the suspect, for example details of other crime reports, it would help them provide a better service to the victims.

Because the police often do not share all the information they hold with specialist stalking victim support services, this could mean that support workers ask victims the same questions that officers have already asked. This could be re-traumatising for victims. The Victims’ Code states that organisations should share information with other agencies, so victims do not have to repeat themselves. This guidance is repeated in the College of Policing guidance about the Victims’ Code.

We found evidence that the same lack of information sharing happens in reverse. Specialist stalking victim support workers told us that they would not routinely share with the police information that they had gathered from the victim. This could include completed risk assessment forms.

In one force, we even found different information sharing arrangements between the police and specialist stalking victim support services within different areas of the force.

We found poor information sharing both ways in most forces. We also identified some innovative practice in respect of information sharing.

Officers sometimes give victims potentially dangerous or unhelpful advice

In our case file review, we examined whether there was evidence that officers had advised victims on how to reduce the risk of harm to themselves.

Of the 470 stalking and breach of protective order cases we reviewed, we found that officers had recorded that they had advised on this in only 202 cases. And we considered that in 31 of these 202 cases, the advice officers gave victims was potentially dangerous or unhelpful.

So, in summary, we found that officers often did not record whether they had given victims advice about how to reduce the risk of harm. And when officers had recorded this, there was some evidence officers had given bad advice.

It is possible officers had given victims appropriate advice about how to reduce the risk of harm to themselves, but they had not recorded it.

The bad advice officers gave was mainly about how victims could protect themselves from online offending. Some officers advised victims to block the suspect on social media sites. We are concerned that this could lead to the suspect escalating their behaviour by visiting the victim.

In ‘Living in fear’, we found that in some of the harassment and/or stalking cases we reviewed, this bad advice extended the risks to victims’ families. This was because the suspect contacted them instead of the victim.

Some specialist stalking victim support workers also told us that officers had given some victims they had worked with inappropriate advice. One example they gave us was of a victim who the police had told to change her mobile phone number.

In ‘Living in fear’, we noted similar things about crime prevention advice. We made a recommendation to the College of Policing that when it was compiling new authorised professional practice, it should include improved guidance to officers on crime prevention advice for victims, particularly about online offending.

In our 2019 stalking report, we also made a recommendation to Sussex Police about inappropriate crime prevention advice.

In November 2020, the College of Policing published its revised advice to police responders. This states that:

“Officers should be mindful not to advise:

• Victim to come off social networking sites or block the suspect or any associates.

• Victim to change their mobile phone number.

• Any action that may further isolate the victim.

• Any action that might make the victim feel they are to blame.”

In our fieldwork, we also examined the force stalking policies and the training forces gave to officers and staff. We found that some of these did not cover the importance of officers and staff giving good advice to victims, particularly about online offending.

It is disappointing that despite the changes the College of Policing made to its advice to police responders, some forces have not changed their own policies and training to include this important guidance for officers and staff, to help keep victims safe.

Failure to meet six-month time limit is leading to cases being dropped

As a result of section 127 of the Magistrates’ Courts Act 1980, the police or CPS must lay information about summary only cases before a court within six months of the offence being committed. This act is often known as the statute of limitations. S2A is a summary only offence and so this act applies to S2A offences.

We outline in our section ‘Flawed investigations and No Further Action decisions’ that the ‘Home Office Crime Recording Rules for Front line Officers and Staff’ specify what crime outcome types should be used and in what circumstances. When crimes are closed because the six-month time limit has elapsed since the last behaviour occurred, the crime outcome type ‘suspect identified but the time limit for prosecution has expired’ (crime outcome type 17) should be applied to the crime record.

In our case file review, 8 of the 190 S2A stalking cases we examined had been assigned crime outcome type 17. But in each of these eight cases, we found evidence of inappropriate and avoidable delays.

It is possible that the problem of forces failing to meet the six-month time limit is worse than the crime outcome type figures from our fieldwork forces suggest. This is because we found that forces sometimes assign the wrong crime outcome types to offences, as we describe in the section titled ‘Flawed investigations and No Further Action decisions’.

In one fieldwork force, we found two S2A stalking cases where the police had given each offender a conditional caution. In each case, the offender did not comply with the conditions of the caution, and the police closed the case with no further action because the statutory time limit had expired. But the police finalised one of the cases as outcome 15 and one as outcome 16.

In another force, we found that in the year ending 31 March 2023, the police had closed and recorded three S4A cases as outcome type 17. This was clearly an error as S4A offences are not subject to the statutory time limit.

Some of the specialist stalking victim support workers we spoke to also told us that they knew of investigations that the police had closed because the statutory time limit had expired.

We know from our inspection into how well the police and other agencies use digital forensics in their investigations that some forces’ digital forensic systems are slow, ineffective and less professionally managed than the other police forensic disciplines.

In our case file review of S2A stalking cases, we did not find any evidence of cases being closed because the force took longer than six months to investigate because of delays with digital forensic examination. But on the basis of the previous inspection work, we think that digital forensic delays may cause investigations to take longer than six months, resulting in closure of the investigation. In these cases, the crime would be given the outcome type 17.

It is also possible that the police could investigate cases as S4A cases and could run over the six-month time limit. The police could then decide the case had been incorrectly recorded and was in fact an S2A offence. Or they could present the case to the CPS, which might conclude that it only amounted to an S2A offence.

In both scenarios, the above cases would become time-barred. It is important to note that we did not find any evidence of these scenarios in our fieldwork.

In summary, we did find some evidence of cases that the police had closed because the six-month time limit had expired. But in view of the very high proportion of cases that the police close for other reasons, we think the police close most stalking cases before the end of the six-month period.

The lack of provision of stalking intervention programmes

The College of Policing guidance for the identification, assessment and management of serial or potentially dangerous domestic abuse and stalking perpetrators (PDF document) details that forces should provide appropriate perpetrator interventions.

In our 2017 report into the police and CPS response to harassment and stalking, we concluded that there was a lack of suitable programmes for stalking offenders that would help reduce the likelihood of reoffending and protect members of the public.

In that report we recommended that chief constables should work with criminal justice partners to identify what programmes are available to manage offenders convicted of stalking offences in their respective force areas and, in the absence of such programmes, review whether they should be established.

In our 2019 review of this recommendation, we expressed our disappointment that forces had been both slow to begin and slow to complete this work. The recommendation we made in 2017 was intended to increase the focus of police forces and partners on considering and addressing the motivations of perpetrators.

In the fieldwork for this super-complaint investigation, we found that in four of the six fieldwork forces there were no available stalking intervention programmes.

The two forces which ran stalking intervention programmes operated different schemes.

Staff employed by PCCs told us that there was not enough government funding for stalking perpetrator intervention programmes. Sometimes, bidding processes for what funds do exist are run by different government departments, namely the Home Office and the Ministry of Justice. The government departments sometimes stipulate that the funds can only be used for certain types of offenders, for example, DA offenders.

In our fieldwork, we found that two of the six fieldwork forces had joined with PCCs and other partners in making successful bids for funding. One force had made an unsuccessful bid. Three forces had not made any bids for funding.

About the data

Data in this report is from a range of sources, including:

• our annual police efficiency, effectiveness and legitimacy (PEEL) data collection
• our victim service assessment audits
• our inspection fieldwork

When we collected data directly from police forces, we took reasonable steps to agree the design of the data collection with forces. We gave forces an opportunity to quality assure and validate the data they gave us, to make sure it was accurate.

British Transport Police was outside the scope of investigation. Any aggregated totals for England and Wales exclude British Transport Police data, so will differ from those published by the Home Office.

Methodology

Victim service assessment

Our victim service assessments (VSAs) track a victim’s journey from reporting a crime to the police, through to outcome stage. All forces will be subjected to a VSA within our PEEL inspection programme. Some forces will be selected for additional testing on crime recording. The selection process ensures we assess every force on its crime recording practices at least every three years.

We have published details of the technical methodology for the victim service assessment on our website.

We analysed VSA data from our 2021–2022 cycle of PEEL for all 43 forces in England and Wales.

Case file review

We carried out a case file review in six police forces in England, where we examined 530 cases relating to stalking (including 2A and S4A offences), breaches of protective orders, and malicious communications offences.

We did not undertake a full assessment of the way the malicious communications cases (60) had been dealt with; these cases were only reviewed to establish whether stalking behaviours were present and therefore whether these cases should have been recorded and investigated as stalking offences.

The case file reviews were based on random non-statistical samples, so the results are not representative of all forces in England and Wales.

Back to publication

Report on the Suzy Lamplugh Trust’s super-complaint: The police response to stalking