Enhanced video response (EVR)

I am currently a chief superintendent in Dorset Police, having worked in various ranks and roles over my 20-year career to date. As part of the Applied Criminology and Police Leadership Master’s Programme with the University of Cambridge, I explored the use of a virtual response offering to victims of crime in Dorset. I built on previous research conducted on rapid video response (RVR) within Kent Police.  

This trial supports the chief constable’s drive to improve how the force responds to the public and investigates crime. This is in line with the latest His Majesty’s Inspectorate of Constabulary and Fire & Rescue Services (HMICFRS) PEEL report in 2022. 

Introduction and literature review 

For victims and witnesses of crime, the journey begins in force control rooms. These are the gatekeepers that decide the level of service a victim will receive and how quickly (Ekblom and Heal, 1982; Waddington, 1993; Waddington, 1999). However, the available options for handling a call remain limited. They have not evolved beyond sending an officer to the scene, dealing with the report via phone or taking no further action.  

Many people are still promised a visit from a police officer when it is not required, causing an oversupply of services (Walley and Jennison-Phillips, 2020). The time it takes to answer the initial call for service is vital and directly correlates to victim satisfaction levels, and potentially to the outcomes of the case (Clark, Arial and Harinam, 2022).  

Dorset Police is not immune to these pressures and risks. The most recent Dorset Police HMICFRS inspection highlighted grave concerns about how long the force was taking to meet calls for service and the standard of investigations once those calls had been completed (HMICFRS PEEL, April 2022). Dorset Police has taken these recommendations seriously and expressed a desire to make changes.  

While overall crime continues to fall across England and Wales (Ariel and Bland, 2019), the public’s demand for police services continues to rise. Prior research indicates that 94% of surveyed victims expect to see a police officer when they call for help (Ekblom and Heal, 1982). However, responding to incidents that do not require a physical police response prevents officers from attending to victims who require in-person assistance (Sumrall and others, 1981).  

Using data from 15 UK police forces, Walley and Adams (2019) investigated the demands placed on policing and how forces responded to these demands between October 2018 and February 2019. All UK forces in the sample groups acknowledged that they were unable to meet the demands placed on their organisations, and that ‘true victims’ were left waiting.  

Qualitative research conducted in 2015 shed light on the effects of police budget cuts and the resulting increase in demand and pressure on police officers. This research highlighted that the demand placed on officers had increased in tandem with the perception of a decrease in officer numbers, with this imbalance between resources and public demand putting the officers at greater risk. The number of officers leaving the police force had increased, as had sickness rates and low morale. The study highlighted the sense of helplessness the officers felt: they joined policing to provide a beneficial service to the public and keep them safe, but felt they were unable to do so due to the constant stream of demands placed on them (Elliott-Davies and others, 2016). 

There has been a shift towards exploring technological advances, particularly virtual responses, to meet the needs of victims of crime while addressing backlogs (HMICFRS, 2023; National Police Chiefs’ Council, 2020). This shift is presumably due to the changes required throughout the pandemic, which is supported by research conducted in Kent Police.  

Kent Police – Rapid Virtual Response  

Rothwell and others (2022) evaluated the effects of a virtual deployment to low- or medium-risk domestic abuse (DA) victims in Kent, known as rapid virtual response (RVR). This experiment was the first of its kind in England and Wales. A randomised controlled trial (RCT) was conducted, giving callers the option of seeing a police officer on their device remotely. The goals and objectives were to improve victim services, reduce victim wait times, increase efficiency and determine the arrest rates of the two sample groups.  

Findings highlighted that video call meetings with non-emergency DA victims reduced the demands placed on the organisation, reduced wait times for victims (RVR was 656 times faster at responding to victims of DA), and increased their satisfaction levels as measured in follow-up surveys.  

My research looked to build on this revolutionary research by replicating some of its design. I explored the types of victims to which the virtual approach was applied, how this approach affected their support of the criminal justice system and the standards of the investigations once attended. 

Methodology 

My study on enhanced video response (EVR) was a conceptual replication of the RVR RCT. However, the EVR study aimed to enhance the existing body of knowledge by broadening the scope of individuals eligible for a virtual response beyond DA calls. Specifically, this study sought to include all call and crime types that meet the ‘Grade 3 calls’ criteria. Examples of Grade 3 calls in Dorset would include a historic serious assault, a domestic incident where the offender is no longer in the presence of the victim or a harassment/stalking offence.

Only Grade 3 calls were included, as the accumulation of unmet demand is primarily concentrated within these call types. These contacts involve offences of substantial severity that inflict significant harm on victims. However, on assessment at the initial point of call, it was deemed that an immediate policing response (on blue lights) or attendance within 60 minutes was not required.  

After the call taker conducted the risk assessment and determined it to be at a Grade 3 level, the system instructed the call taker to complete a call script in collaboration with the caller. If the individual making the call preferred a virtual response option, the call handler administered the eligibility questions using a pre-programmed set of drop-down questions.  

An RCT design was applied. Callers (victims and witnesses) to the police were randomly allocated to either a treatment group, which received the virtual response intervention, or a control group, which received the standard in-person deployment.  

A dedicated Force Communications Centre supervisor was assigned throughout the trial to oversee the process, with primary responsibility for continuously monitoring incoming Grade 3 calls. On meeting the specified eligibility criteria, the supervisor added the call log reference to a customised spreadsheet exclusively accessible to the supervisor and the EVR sergeant. From here the log was randomised based on the last digit of the call reference number. For those that fell into the EVR category, the call log was passed to the awaiting EVR sergeant.   

After reviewing the log, the EVR sergeant randomly assigned the treatment call to a virtual officer who was available at that time. Once a virtual officer had been assigned a call, their status was updated to ‘dispatched’ on the call log. After establishing contact with the victim and initiating the primary investigation, the virtual officer included an ‘at-scene’ tag. After concluding the virtual communication, individuals indicated their departure from the call log. At this juncture, the remaining portion of the primary investigation was finalised. This meant completing the investigation plan and considering evidential factors prior to transmitting the investigation record to an independent evidential review officer.  

EVR is a further deployment option through which a police officer visits the scene of an incident virtually – through a smartphone, computer or iPad – rather than in person. Individuals eligible for a virtual response were sent a text message or email link to a platform called GoodSAM, which was already procured by the force and is used by most police forces nationally. Through that procurement process, data security and management of police information (MoPI) compliance was reviewed and embedded along with the ability to use cloud storage for the footage obtained. The system is very simple for victims to use. Once the victim has clicked the link, the officer appears on screen without the need to download any software or apps onto their devices.

To maintain business-as-usual staffing levels, no police officers from patrol teams were used during the trial setup to avoid any potential impact on the validity of the control group. Dorset is divided into two local policing areas (LPAs): the county policing area and the Bournemouth, Poole and Christchurch policing area. This RCT was conducted for three months in the county LPA using 10 restricted police officers and two sergeants. 

Findings 

A total of 4,950 Grade 3 logs were created in the county LPA during the trial period, of which 3,891 (78.6%) were excluded. The most common reasons included a concern for welfare which required officers to attend in person to mitigate any risk, the victim declining EVR (219) or the report originating from a third party not directly involved in the incident.  

Of the remaining 1,059 cases eligible for a virtual response, 659 (62.2%) were randomly allocated to the treatment group (EVR) and 400 (37.8%) to the control group through the allocation process described above.  

In the control group, 97 cases (24.3%) were physically attended. The study highlighted that a call log remains on the ‘open summary queue’ for several days, even weeks in many cases, due to no further harm reported or because the force has been unable to source a police officer to attend. Moreover, 167 (41.8%) cases were sent to an individual officer’s inbox for them to manage deployment whenever available at a future date. 136 (34%) were resolved over the telephone by a police officer allocated to attend, with the victim never seeing a police officer.   

When the EVR team were on duty, they commenced an investigation within 1.28 hours on average. However, for the control group this was 203.10 hours. The wider findings on overall timelines are detailed further below, however these findings are statistically significant.   

After the trial, all 1,059 cases were assessed to identify the effect of the intervention on policing outcomes. These findings are listed in Table 1 and discussed below.