Policing a pandemic: changes in police response to intimate partner violence (IPV) during the first lockdown in England

Author:
Lucy TraffordUniversity of Oxford, UK

Search for other papers by Lucy Trafford in
Current site
Google Scholar
Close
Open access
Get eTOC alerts
Rights and permissions Cite this article

This article investigates the police response to intimate partner violence (IPV) during the first lockdown in England (23 March–30 June 2020), a time when more people were confined to their homes and rates of IPV increased globally (). Having gained access to data for domestic abuse-flagged incidents from a south-eastern police force, this research compares quantitative data for 6808 incidents during the first lockdown in England with 6408 incidents during the equivalent 2019 timeframe. Quantitative analysis was conducted by comparing descriptive statistics and chi-squared testing.

This study finds that age distribution changed for victims and suspects, with IPV decreasing among younger age groups and increasing within older age categories. Shifts occurred in the categorisation of IPV crimes with an increase in crimes that can be committed remotely, most notably stalking and malicious communications. Additionally, the risk level differed for IPV incidents, with a reduction in incidents recorded as medium- and high-risk. Whereas, standard-risk incidents rose substantially, causing a change in the distribution of risk levels across reported incidents. This shift was reflected in fewer arrests, except among higher-risk incidents which maintained higher arrest rates. Official outcomes similarly decreased, with fewer court disposals and more simple cautions.

Abstract

This article investigates the police response to intimate partner violence (IPV) during the first lockdown in England (23 March–30 June 2020), a time when more people were confined to their homes and rates of IPV increased globally (Williamson et al, 2020). Having gained access to data for domestic abuse-flagged incidents from a south-eastern police force, this research compares quantitative data for 6808 incidents during the first lockdown in England with 6408 incidents during the equivalent 2019 timeframe. Quantitative analysis was conducted by comparing descriptive statistics and chi-squared testing.

This study finds that age distribution changed for victims and suspects, with IPV decreasing among younger age groups and increasing within older age categories. Shifts occurred in the categorisation of IPV crimes with an increase in crimes that can be committed remotely, most notably stalking and malicious communications. Additionally, the risk level differed for IPV incidents, with a reduction in incidents recorded as medium- and high-risk. Whereas, standard-risk incidents rose substantially, causing a change in the distribution of risk levels across reported incidents. This shift was reflected in fewer arrests, except among higher-risk incidents which maintained higher arrest rates. Official outcomes similarly decreased, with fewer court disposals and more simple cautions.

Key messages

  • During the first lockdown in England there was a significant change in victim and suspect demographics, with IPV increasing among older victims and suspects and decreasing for younger victims and suspects, possibly reflecting an increase in IPV between cohabiting couples.

  • Crime types differed, with increases in non-physical IPV which could be committed remotely, indicating that non-cohabiting suspects found alternative methods of abusing their victims during lockdown.

  • Police responses to IPV altered during lockdown, with reductions in arrests (except among higher-risk incidents) and decreases in court disposals, possibly reflecting changes in IPV risk levels recorded.

Introduction

Impacts on IPV of epidemic proportions

At the beginning of 2020 when the spread of COVID-19 became a worldwide pandemic, governments enforced lockdowns globally to maintain public safety, causing more people to live behind locked doors than ever before. Yet, staying at home is far more dangerous for some than others, and always has been. Prior to the pandemic, the rate of female homicide reached a six-year high, with DA accounting for 46.0 per cent of female homicides in England and Wales (ONS, 2020a). Smith (2020) equated this to two per week. These tragic outcomes are more pronounced among women, with 7.0 per cent of male homicides linked to DA (ONS, 2020a). Hence, ‘home’ was declared the most dangerous space for women and children in 2018 (UNODC, 2018). Additionally, although intimate partner violence (IPV) rates have steadily increased since 2013 (Walklate et al, 2021), IPV has been found to increase during economic and humanitarian disasters (Peterman et al, 2020).

Restrictions confining individuals to their homes with few exemptions provided perpetrators with increased opportunities to control and abuse their victims (Walklate et al, 2021). Tactics perpetrators frequently use to enforce and maintain control appear to have been exacerbated by lockdown: isolation, surveillance, restrictions on activities and movement (Williamson et al, 2020). Fear of COVID-19 enabled perpetrators to threaten victims with infection and to limit victims’ access to medical treatment, personal protective equipment and information during the pandemic (Moreira and da Costa, 2020). For victims, coping mechanisms, respite and informal support networks through work, friends, leisure and daily routines were limited extensively by restrictions (Bates et al, 2021).

Furthermore, victims experiencing high levels of coercion could have access to technology controlled by perpetrators and conversations overheard, making them feel unsafe to report abuse (Women’s Aid, 2020). Victims for whom reporting could act as a precursor to leaving the relationship may have been dissuaded due to lockdown increasing risks of separation, because of transport limitations, fear of catching COVID-19, financial difficulties and struggles in finding alternative accommodation through official or support networks (Williamson et al, 2020). Fear of infection and overwhelming services could also have caused victims to normalise the violence (Virkki, 2007) more frequently during lockdown and therefore not seek assistance, despite the government and police urging victims to report IPV (Home Secretary, 2020). Hence, DA was deemed the ‘shadow pandemic’ (UN Women, 2020).

Living with IPV can cause victims and their children severe emotional, psychological and physical harm with reduced life quality which continues long after the relationship ends (Neale, 2018). Often, IPV is inter-connected with child abuse, with children purposely or inadvertently harmed by abuse (Hester and Radford, 2006). Hence, the social, health and employment costs of IPV are estimated at around £66 billion annually in England and Wales alone (Oliver et al, 2019). IPV is experienced irrespective of ability, age, class, ethnicity, sexuality and in public and private spaces (Westmarland, 2015). Although, as females tend to constitute the majority of victims, experiencing more severe injuries and long-term impacts, IPV is widely recognised as a gendered phenomenon (Brooks-Hay and Lombard, 2018). In England and Wales 28.9 per cent of women (around 4.8 million) and 13.3 per cent of men (around 2.2 million) have reported experiencing DA since they were 16 (ONS, 2019a).

During the first lockdown in England (23 March–30 June 2020), the rates of IPV rose globally (Williamson et al, 2020) and nationally (ONS, 2020b). DA services, including charities supporting victims and perpetrators (Grierson, 2020) experienced unprecedented rises in calls for help (WHO, 2020). Looking beyond calls for help, Refuge experienced a 50.0 per cent reduction in refuge vacancies between May and June 2020 (Justice Inspectorate, 2021a), while 14 women and two children were murdered in relation to DA between 23 March and 27 April 2020, double Smith’s (2020) predicted annual rate. Yet, Bates et al (2021) found that while DA homicides increased during each lockdown this was not statistically significant and remained in line with annual rises, suggesting that lockdown had complex and nuanced impacts on victims’ safety. As some faced heightened risks following government restrictions Bradbury-Jones and Isham (2020) referred to lockdown as ‘a pandemic paradox’.

Police: a focal contact point

Prior to the pandemic, IPV placed a significant strain on scarce financial, emotional and temporal police resources (Neale, 2018). IPV often represents the largest category of calls to the police (Friday et al, 2006), with the highest repeat victimisation of any crime (Walby and Allen, 2004). In England and Wales during 2019, the police were called to 1,316,800 IPV incidents (ONS, 2020c), with physical violence, threats and coercion being the most frequent IPV crimes reported to the police and sexual assault the least reported (ONS, 2020c). As well as being one of the most frequent crimes reported to the police, IPV is one of the most challenging crimes for police to resolve. This is because of the complexity of incidents, high levels of intimacy between parties and shared care of dependents which challenges the dichotomous nature of criminal proceedings and makes certain criminal justice system (CJS) responses inappropriate (Stark, 2007).

Victims and perpetrators generally encounter the police more frequently than any other element of the CJS. This is because few incidents progress to court and fewer end in prosecution, with attrition at every stage (Owens et al, 2014). Charges are often dropped due to lack of evidence or as victims do not wish to pursue criminal sanctions (Hester and Westmarland, 2005). The disparity between cases reported and those progressing to prosecution has been greatest during the pandemic with courts shut and ever-increasing backlogs (Criminal Justice Joint Inspection, 2021). This, coupled with social and health services being reduced, and appointments moved online (Walklate et al, 2021), has made frontline police officers a crucial contact point for IPV victims during lockdown.

Policing IPV in a pandemic

Policing IPV during the pandemic was especially difficult. The London Metropolitan Police Service received 41,158 DA calls in the first lockdown, a 12.0 per cent increase on the equivalent 2019 timeframe (ONS, 2020b). However, this increase in police calls was not replicated nationwide, with Walklate et al (2021) noting variation across forces and Hohl and Johnson (2021) identifying reductions across six forces. Police also experienced an increase in the complexity of IPV cases, with complications created by parties cohabiting, difficulties in finding alternative accommodation, fears of infection and a rise in individuals struggling to gain help for mental health issues (Hohl and Johnson, 2021). As frontline police respond in-person, police forces needed to balance the safety of victims, perpetrators and their own officers in response to COVID-19. Indeed, officers faced threats to their personal safety and mental health as key workers, with perpetrators frequently threatening officers with infection (Justice Inspectorate, 2021a).

Table 1:

Descriptives for victim and suspect demographics

Table 1: Descriptives are presented for victim and supect demographics between lockdown (2020) and lockdown equivalent (2019). Frequency did not demonstrate a large change for relationship type (partners, ex-partners and other sexual relationships), with the greatest change being a 1% increase for partners. Cohabiting showed a greater change, with yes answers increasing by 4.3% and no decreasing by 3.8% during lockdown. Victim gender showed no change for male victims and a 0.1% increase for female victims. Suspect gender followed a similar pattern with a 0.6% increase for female and male suspects. Vulnerability for victims remained relatively stable, with a 0.7% decrease for yes and increase for no during lockdown. There was no change in suspect vulnerability. For all categories discussed, missing data showed little change, the greatest being a 1.2% decrease for suspect gender.
Demographics 2019 Lockdown equivalent 2020 Lockdown
Frequency Percentage Frequency Percentage
Relationship type Partners 2890 45.1 3138 46.1
Ex-partners 3501 54.6 3656 53.7
Other sexual relationship 17 0.3 14 0.2
Cohabitating Yes 948 14.8 1302 19.1
No 2998 46.8 2929 43.0
Missing 2462 38.4 2577 37.9
Victim gender Female 4074 63.3 4317 63.4
Male 2325 36.3 2473 36.3
Missing 9 0.1 18 0.3
Victim vulnerability Yes 625 9.8 619 9.1
No 5783 90.2 6189 90.9
Suspect gender Female 1057 16.5 1164 17.1
Male 2958 46.2 3183 46.8
Missing 2393 37.3 2461 36.1
Suspect vulnerability Yes 241 3.8 256 3.8
No 6167 96.2 6552 96.2
Total 6408 100.0 6808 100.0

In initially responding to IPV incidents, frontline police officers determine associated risks, the sole or main perpetrator, crimes committed, whether the incident exemplifies IPV and the most appropriate response. In policing the pandemic, interviewed officers indicated that the scope for individual decision-making and discretion expanded (Hohl and Johnson, 2021). These assessments influence the decisions reached by other police officers and members of the CJS by determining what constitutes crime scene evidence, which questions should be asked of involved parties and whether cases should be passed on to the specialist DA unit for further investigation (Myhill and Johnson, 2016). Thus, frontline officers have the greatest discretion in IPV incidents, following an unusual organisational structure whereby discretion diminishes with police hierarchy (Stanko, 2007). Although the Crown Prosecution Service (CPS) makes the final legal decision of whether to prosecute IPV cases, the police decide which incidents to refer to the CPS (CPS, 2013), acting as the unofficial gatekeepers of the courts (Hartman and Belknap, 2003). Thus, it is police decision-making that dictates whether IPV cases proceed through the CJS. The importance of understanding how additional pressures from the pandemic and rises in reporting have influenced police decision-making and responses during lockdown has never been as pressing.

Table 2:

Descriptives for crime characteristic

Table 2: Descriptive statistics are presented for crime characteristics between lockdown and lockdown equivalent. Within crime type there is no percentage change for misappropriation of property and other crimes, although there is an increase in their frequency. Crimes requiring contact with the victim (physical, sexual abuse, restrictions of freedom) decrease while those that can be committed without face-to-face interactions (malicious communications, stalking, fraud, criminal damage and injunction breach) increased.
Crime characteristic descriptives 2019 Lockdown Equivalent 2020 Lockdown
Frequency Percentage Frequency Percentage
Crime type Physical abuse 2584 40.3 2648 38.9
Non-crime 1921 30.0 2051 30.1
Harassment 554 8.6 422 6.2
Malicious communications 335 5.2 450 6.6
Sexual abuse 284 4.4 272 4.0
Criminal damage 193 3.0 241 3.5
Threats 159 2.5 142 2.1
Controlling/coercive behaviour 126 2.0 113 1.7
Misappropriation of property 111 1.7 119 1.7
Stalking 73 1.1 265 3.9
Restrictions of freedom 17 0.3 10 0.1
Fraud 16 0.2 25 0.4
Injunction breach 15 0.2 30 0.4
Other 3 0.0 5 0.0
Missing 17 0.3 15 0.2
Suspect substance use Alcohol 1205 18.8 1101 16.2
No substance used 947 14.8 1083 15.9
Alcohol and drugs 160 2.5 163 2.4
Drugs 68 1.1 102 1.5
Missing 4028 62.9 4359 64.0
Total 6408 100.0 6808 100.0

Project aims

This project aimed to understand if the first lockdown in England triggered an increase in IPV incidents within a south-eastern police district, in comparison to the equivalent 2019 timeframe, and if so, the magnitude of that increase. It analysed whether changes occurred in the nature of IPV reported (by reviewing incident characteristics), those reporting IPV (by analysing suspect and victim demographics) and whether police responses to IPV incidents altered (by comparing the arrest rates, allocation of DA risk levels, frequency of official outcomes and incident disposal types). It was then considered whether differences in police responses were influenced by changes identified in the frequency of reports, types of IPV reported or those reporting IPV.

Methodology

This research emerged from a wider project which explores how police understand and respond to IPV. Ethical approval was sought and granted by the University of Oxford and a south-eastern police force, with a data sharing agreement produced. This project looks at 14,338 incidents between married, dating, sexual and ex-partners (intimate partners), flagged as DA by the police, between 23 March and 30 June during 2019 and 2020. This data was recorded by frontline officers on the district Record Management System (RMS) primarily for police monitoring and future analysis, making data variables and recording techniques nonideal for research analysis (Bowstead, 2019). However, police data provides rich contextual incident information, not accessible through surveys (Joshi and Sorenson, 2010), and a large sample size which promotes confidence in the statistical significance and robustness of these findings.

This data includes victim/suspect demographics (gender, vulnerability, relationship status, cohabitation), incident characteristics (crime classification, suspect substance use) and police responses (DA risk level, arrest, official action taken, disposal type). As this medium-sized police force naturally covers a variety of urban and rural areas, with mixed age, religion and class demographics, this dataset reflects a varied sample of IPV incidents within England.

To understand whether differences occurred in victim/suspect demographics, incident characteristics and police responses, two datasets were created. The first sample consisted of incidents reported to and recorded by the police during the first English lockdown, which was introduced on the 23 March 2020 and ran until the 30 June 2020. This dataset will subsequently be referred to as ‘lockdown’. The second dataset included all incidents reported to and recorded by the police during the same period (23 March–30 June) in 2019 and will be referred to as ‘lockdown equivalent’. This equivalent dataset acted as a control group, enabling assessment of how the victim/suspect demographics, incident characteristics and police responses changed during the first lockdown.

The two datasets were cleaned by removing duplicates (identified by the police and through comparing identical timestamps for incidents recorded), as well as incidents that did not fall within this project’s definition of IPV (victims/suspects aged <161), and records that lacked critical data (crime classification, DA risk level). This reduced the lockdown dataset from 7418 to 6808 (8.2%) incidents and the equivalent dataset from 6920 to 6408 (7.4%). As the decrease in incidents is similar across both datasets and there was not a large size reduction following data cleaning, this suggests that there is a good level of data for comparisons and reliability is maintained. While age was recorded by the police for each incident, age was categorised for this study. Due to the large number of crimes recorded by the police, offences were grouped into representative categories (for example, ‘physical abuse’).

Descriptive statistics were conducted to identify whether differences in victim/suspect demographics, incident characteristics or police responses occurred between the ‘lockdown’ and ‘equivalent’ timeframes. Chi-squared tests of independence were run for variables demonstrating the greatest change, with missing data removed, to determine which changes were statistically significant. For rigorous testing, a 99 per cent confidence interval was used. To ensure that none of the expected frequencies were <5, categories within crime and disposal type were merged. Crime types were categorised into direct, remote, either (for crimes that could be perpetrated directly and remotely: for example, threats, injunction breach), other (crimes that were not necessarily IPV-related: for example, fraud, dangerous driving) and no crime. Disposal type was split into court orders (cautions/summons), out of court disposals (cautions/community orders), prosecution problems and no crime.

Findings

The number of IPV incidents recorded by this south-eastern police force rose from 6408 to 6808 (6.2%) during lockdown. This reflected annual rises in IPV and mirrored increased reports nationally and globally (Williamson et al, 2020; ONS, 2020b), yet contrasted with reductions identified across six forces during lockdown (Hohl and Johnson, 2021). Findings are split into a discussion of victim/suspect demographics, incident characteristics and police responses. All changes are represented as percentage point differences from the equivalent to lockdown dataset.

Victim and suspect demographics

Relationship and cohabitation

Victims and suspects cohabiting increased during lockdown by 4.3 percentage points (pp) (from 14.8% to 19.1%), which proved significantly correlated with lockdown (Table 3). This increase in IPV between cohabiting partners was expected due to parties having greater exposure to each other and additional pressures caused by the pandemic, including childcare and financial difficulties, as well as the impact of lockdown restrictions on mental wellbeing (Arenas-Arroyo et al, 2021). However, relationship type (partners and ex-partners) proved not to be statistically significant, possibly suggesting that more ex-partners cohabited. No substantial changes occurred for victim gender, suspect gender and victim and suspect vulnerability (mental health, disability, youth, elderly).

Table 3:

Chi squared results for test of independence between lockdown and victim/suspect demographics, incident characteristics and police response

Table 3: Chi squared results are presented for tests between lockdown and victim/suspect demographics (cohabitation, victim age and suspect age). All proved significant with similar pearson values, although the highest occured for suspect age (43.64) and the lowest for victim age (32.27). For lockdown and incident characteristics, crime type is significant (pearson value 84.73), as is suspect substance use (12.21). Chi squared for police response (D A risk level, arrest, official outcome and disposals) are all significant, with the highest pearson value occurring for D A risk level (176.59). The pearson values are much lower for the other variables (arrest 14.96, disposal 5.59 and official outcome 0.62).
Significant changes Significance Pearson’s Chi 2 Likelihood Number Degrees of freedom Cramer’s V
Victim/suspect demographics Cohabitation p<0.001 39.55 39.88 9344 1 0.065
Victim age p<0.001 32.27 32.26 14935 6 0.046
Suspect age p<0.001 43.64 43.66 9552 6 0.068
Incident characteristics Suspect substance Use p<0.001 12.21 12.17 14949 1 0.029
Crime type p<0.001 84.73 86.45 14911 5 0.075
Police response DA risk level p<0.001 176.59 176.56 12620 2 0.118
Arrest p<0.001 14.96 14.92 14949 1 0.032
Official outcome p<0.001 0.62 0.62 14191 1 0.007
Disposal p<0.001 5.59 5.57 14150 3 0.020
Table 4:

Descriptive statistics for police responses

Table 4: Descriptive statistics are provided for police responses during lockdown and lockdown equivalent. The percentage and frequency decreases for high and medium D A risk level, while it increases for standard. There is a decrease in those arrested and experiencing official action and an increase for those not arrested and not experiencing official action. For disposal type, all official outcomes decreased, except for simple caution (summonsed, charged, conditional caution). Evidential problems decreased substantially, other prosecution problems and not in public interest decreased, as did other agency outcomes. Community resolution percentage did not change, while no crime decreased and unknown outcomes significantly increased.
Police response 2019 Lockdown equivalent 2020 Lockdown
Frequency Percentage Frequency Percentage
Domestic abuse risk level High 1137 17.7 867 12.7
Medium 1607 25.1 1217 17.9
Standard 2802 43.7 3582 52.6
Missing 862 13.5 1142 16.8
Arrest Yes 1813 28.3 1681 24.7
No 4595 71.7 5127 75.3
Official action taken Yes 481 7.5 383 5.6
No 5809 90.7 5831 85.6
Unknown 118 1.8 594 8.7
Disposal type Evidential problem 3642 56.8 2060 30.3
No crime 1960 30.6 2071 30.4
Other prosecution problem 1830 28.6 1658 24.4
Summonsed – alternate offence 228 3.6 173 2.5
Charged 136 2.1 132 1.9
Unknown 118 1.8 594 8.7
Prosecution not in public interest 60 0.9 39 0.6
Conditional caution 42 0.7 4 0.1
Summonsed 41 0.6 12 0.2
Caution – alternate offence 26 0.4 22 0.3
Other agency outcome 9 0.1 3 0.0
Community resolution 6 0.1 5 0.1
Simple caution 1 0.0 33 0.5
Youth conditional caution 1 0.0 0 0.0
Charged – alternate offence 0 0.0 2 0.0
Total 6408 100.0 6808 100.0

Victim and suspect age

Figure 1: Figure showing trends in age categories for victims during lockdown and lockdown equivalent. The largest category is 21-30, with the older age categories progressively decreasing. Victims within the 16-20 age category are relatively lower, although slightly higher than 51-60. During lockdown the number of victims decreased for 16-20 and 21-30 but increased for all other categories, except 71+.
Figure 1:

Victim age categories in lockdown and lockdown equivalent

Citation: Journal of Gender-Based Violence 6, 3; 10.1332/239868021X16528069833875

The most obvious changes occurred for victim and suspect age. The frequency of victims in the 16–20 age category decreased by 2.3pp (9.2% to 6.9%), while the number of suspects decreased by 1.6pp (5.0% to 3.4%) during lockdown. Although the age group 21–30 remained the largest contributor to total IPV incidents for both parties, the total number of suspects fell by 1.6pp (21.5% to 19.9%), while victims decreased by 1.9pp (34.6% to 32.7%). This decrease in victims and suspects aged 16–20 and 21–30 is unusual, as IPV is most common among younger couples. Indeed, the highest risk of IPV occurs for women in adolescence and young adulthood (Joshi and Sorenson, 2010).

Interestingly all age groups above 30 experienced an increase for victims and suspects. The second largest age group, those aged 31–40, demonstrated a 1.4pp increase in victims reporting IPV (30.2% to 31.6%). While the number of suspects in this age group rose by 3.2pp (18.4% to 21.6%) during lockdown. This is a substantial increase, particularly among suspects, and this trend continued across all older age categories. This suggests a possible link between cohabitation and increased IPV, as individuals aged above 30 are more likely to cohabit (ONS, 2019b), especially during lockdown when an increase occurred in young adults moving back to live with their parents (Graham, 2021). For both datasets, the largest age category not cohabiting was 21–30-year-olds (37.0% and 38.2%), while the largest age category cohabiting was 31–40-year-olds (31.3% and 31.7%). Additionally, older age groups of working age, usually separated during work hours, are more likely to experience situational pressures from working at home together, additional childcare requirements and financial pressures linked to redundancies and furlough (Walklate et al, 2021). Chi-squared testing identified a statistically significant correlation between lockdown and victim/suspects age (X2(6, N = 14935) = 32.27, p = <0.000 and X2(6, N = 9552) = 43.64, p = 0.000 respectively).

Figure 2: Figure showing trends in age categories for suspects during lockdown and lockdown equivalent. The largest category is 21-30, with the older age categories progressively decreasing. Suspects within the 16-20 age category are relatively lower, although slightly higher than 51-60. During lockdown the number of suspects decreased for 16-20 and 21-30 but increased for all other categories.
Figure 2:

Suspect age categories in lockdown and lockdown equivalent

Citation: Journal of Gender-Based Violence 6, 3; 10.1332/239868021X16528069833875

Despite the increase in victims and suspects cohabiting, almost half of the incidents (43.0%) continued to be perpetrated by those not cohabiting, possibly because the largest age category of victims and suspects continued to be 21–30-year-olds. This age group may have been less affected by lockdown, possibly as partners still came into contact during lockdown or as victims experienced IPV remotely from non-cohabiting or ex-partners. Alternatively, this age group frequently rely on technology and may have been more likely to perpetrate crimes remotely, such as malicious communications and stalking (Duerksen and Woodin, 2019), which ties in with changes in crime classification.

Incident characteristics

Increased remote violence

The most significant changes during lockdown occurred for stalking and malicious communications, crimes that together account for 10.5 per cent of total IPV cases during lockdown. These crimes are ‘remote’ as they can be perpetrated through a medium rather than physical contact (language, technology and presence) (Duerksen and Woodin, 2019). Stalking increased the most, by 2.8pp (1.1% to 3.9%), and malicious communications rose by 1.4pp (5.2% to 6.6%). There was also a small increase in ancillary crimes linked to IPV such as criminal damage which rose by 0.5pp (3.0% to 3.5%). Hohl and Johnson (2021) similarly identified a rise in stalking and the victim support charity Paladin noted a 50.0 per cent rise in referrals and increased police response (Bracewell et al, 2020). Differences between lockdown and the equivalent dataset may partially be explained by the Stalking Protection Act 2019 which came into force in January 2020. This encouraged changes in crime recording, with a greater focus on distinguishing between stalking and harassment. Although subsequent increases were expected in stalking, this cannot account for the 2.8pp rise.

This unexpected shift in crime classification may be explained by the high percentage of non-cohabiting and ex-partners, which continued to constitute 43.0 per cent and 53.7 per cent of all IPV suspects respectively. Physical violence was most likely harder for non-cohabiting and ex-partners during lockdown because of restrictions confining them and their victims at home. While working from home, furlough and redundancy increased time for surveillance and contacting victims. Hence, this rise in remote crimes may be due to non-cohabiting and ex-partners having to change their methods of controlling and abusing victims. Cohabiting abusers were less likely to perpetrate remote violence, with greater access to their victims during lockdown. Indeed, non-cohabiting suspects committed 93.5 per cent of remote crimes with changes in crime classification proving statistically significant for non-cohabiting suspects during lockdown (X2(4, N = 4217) =314.03, p=<0.000).

Considering that stalking in-person became harder during lockdown due to restrictions, it is likely that the nature of stalking changed, with more perpetrators utilising technology which can be an easier and faster method of contacting victims. Indeed, reliance on technology increased globally, including among IPV victims (Bracewell et al, 2020). The use and development of new apps during lockdown, including Zoom and House Party, also enabled stalkers to conveniently track victims and escalate stalking. Alternatively, as stalkers are ‘not law-abiding, compliant citizens’ (Bracewell et al, 2020), they may have been less likely to follow lockdown restrictions, while lockdown would have made it easier to locate and monitor victims in-person.

Reductions in direct violence

Comparatively direct violence (physical abuse and sexual abuse) decreased, with physical abuse decreasing by 1.4pp and sexual abuse by 0.4pp. Research from other countries similarly found no increase in physical violence (Arenas-Arroyo et al, 2021). This shift in forms of violence seems unusual, as physical crimes would be expected to increase further with restrictions limiting movement and greater proximity between those cohabiting. Although, physical violence perpetrated by cohabiting partners, which was hidden at the aggregate level, did increase substantially (9.5pp).

Chi-squared tests proved that changes in crime classification were significantly correlated with lockdown (Table 3). These results indicate that the types of IPV crimes perpetrated were altered by the imposition of lockdown restrictions, reflecting Women’s Aid’s (2020) findings that 93.5 per cent of female victims experienced changes in abuse.

Incidents recorded as ‘no crime’ increased as expected with the rise in IPV cases, by 0.1pp (30.0% to 30.1%), suggesting that the police perception of what constitutes IPV did not change between lockdown and the equivalent dataset.

Substance use

Suspect substance use was significantly correlated with lockdown (X2(1, N = 14949) = 12.21, p = <0.000). Interestingly, the greatest decrease occurred for alcohol use which fell by 2.6pp (18.8% to 16.2%), despite Sallie et al (2020) finding that a third of adults consumed alcohol more frequently during the pandemic. The reduction in IPV incidents involving suspect alcohol consumption might be explained by the 16.0pp increase in IPV incidents reported during work hours (42.0% to 58.0%), with partners constantly at home, as alcohol is less likely to be consumed during working hours compared to evening and weekend hotspots (Joshi and Sorenson, 2010). Suspect drug use, which occurred in 1.5 per cent of reported IPV incidents, increased by 0.4pp (1.1% to 1.5%), possibly because drug use increased at home during lockdown because of boredom and restrictions (Zaami et al, 2020), or as more drug users committed IPV.

Police responses

DA risk levels

DA risk levels are assigned by police at the scene or immediately following, with details recorded on the RMS. These increase in severity through standard, medium and high. During the first lockdown, incidents classed as standard-risk increased by 8.9pp (43.7% to 52.6%). This may be because low-level IPV was perpetrated by a greater variety of individuals, as officers mentioned in interviews with Hohl and Johnson (2021). The rise in IPV among older age groups could support this notion of reports increasing among those who would not usually report or experience IPV. This could be affected by situational factors, such as lockdown restrictions, financial pressures following redundancies and furlough and greater exposure to partners, children and pets (Arenas-Arroyo et al, 2021), mirroring Johnson’s (2010) suggestion that situational factors can increase violence among partners who are less inclined to commit IPV. Increased reporting of standard-risk IPV may also be explained by victims who have experienced low-level abuse prior to the pandemic recognising their partner’s behaviour as abusive, through more frequent and abusive behaviour or greater exposure to their partner (Hohl and Johnson, 2021). This is supported by Hattery (2009) who explains that IPV warning signs identified by Browne (1989) (monitoring, exclusivity and possessiveness) can initially be interpreted by female victims as expressing romantic ideals, perpetuated by media representations of romance.

Changes in distribution of DA risk levels

During lockdown, the percentage of incidents recorded as standard-, medium- and high-risk changed significantly. In 2019 43.7 per cent of total incidents were recorded as standard-risk, 25.1 per cent as medium-risk and 17.7 per cent as high-risk. This demonstrates an inverse relationship, with the proportion of incidents increasing as police risk level decreased. Comparatively, during lockdown, standard-risk incidents constituted 52.6 per cent of incidents, medium-risk 17.9 per cent and high-risk 12.7 per cent. This shows a substantial shift in the distribution of risk levels allocated to incidents. Instead of an incremental reduction, evident from 2019, where there was a steady decrease in difference between standard- and medium-risk (18.6%) and medium- and high-risk (7.4%), during lockdown there was a larger difference between standard- and medium-risk (34.7%) and a smaller difference between medium- and high-risk (5.2%). Indeed, the number of incidents recorded as high-risk reduced by 5.0pp (17.7% to 12.7%), while medium-risk dropped by 7.2pp (25.1% to 17.9%). Thus, far fewer medium- and high-risk incidents were recorded by the police during lockdown. Chi-squared tests demonstrated a statistically significant link between DA risk level and lockdown (X2(2, N = 12620) = 176.59, p = <0.000).

Discussions with the force illuminated a policy change in lockdown, to prioritise support for the most serious cases by limiting incidents recorded as medium- and high-risk. As physical abuse constituted a lower percentage of IPV incidents during lockdown, these changes in risk allocation may have been influenced by fewer injuries caused through direct physical violence because physical abuse incidents, especially those resulting in injuries, are often recorded as higher risk (Myhill, 2019; Petersson and Strand, 2020). The reduction in suspects’ alcohol consumption may partially account for this decrease in incidents regarded as more serious, as alcohol consumption has been linked to increases in risks of IPV perpetration, physical aggression and victim injuries (Trujillo and Ross, 2008; Dichter et al, 2011).

Arrest

Reflecting this reduction in medium- and high-risk IPV incidents, the frequency of arrests decreased by 3.6pp (28.3% to 24.7%). Table 3 demonstrates that arrest rate was significantly correlated with lockdown. Changes in frequencies and composition of DA risk levels recorded by the police could partially explain changes in arrest rates, as perpetrators whose abuse is rated as higher-risk are more likely to be arrested (Petersson and Strand, 2020). Indeed, chi-squared tests found that arrest frequency was significantly related to DA risk level for equivalent and lockdown datasets (X2(2, N = 5546) = 1458.58, p = <0.000 and X2(2, N = 5666) = 1997.97, p = <0.000 respectively).

The reasons for fewer arrests are likely to mirror explanations for reductions in medium- and high-risk incidents recorded, as arrest is similarly associated with substance use, physical violence and infliction of injuries (Trujillo and Ross, 2008). Dichter et al (2011) also identified arrest as less likely for incidents reported during work hours. As more incidents were reported during work hours (16.0pp), with partners together throughout the day during lockdown, this may have contributed to fewer arrests.

Interestingly, results showed significant increases in the percentage of medium- and high-risk incidents ending in arrest (7.2pp and 7.6pp respectively), while arrest decreased for standard-risk incidents by 1.5pp. This suggests that police perceived arrest as more pressing for high- and medium-risk incidents, confirmed by police force policy changes to prioritise arrest in the most serious cases during lockdown.2 Arrest may have been viewed as necessary in more medium- and high-risk incidents due to the increased dangers posed by health, financial and lockdown pressures, with greater opportunities to control and abuse victims (Williamson et al, 2020) and subsequently higher risks of recidivism. These implications for IPV perpetration may have caused victims to experience and exhibit greater levels of fear during lockdown, which could partially explain the increased arrest rate for medium- and high-risk victims, as victim fear has been positively linked to arrest (Dichter et al, 2011). This reinforces how changes in incident characteristics can affect police decision-making.

Official action and disposal type

Disposal types recorded by the police include: whether the suspect was charged, summoned to court, cautioned, given a community resolution, where the victim did not support prosecution, prosecution was not in the public interest, ‘no crime’ was committed or a disposal has not yet been recorded on the RMS system (unknown).

It should be noted that unknown outcomes rose by 6.9pp (1.8% to 8.7%), possibly explained by differences in the recording and sampling timeframe between the two datasets. The 2019 equivalent data was sampled a year later than the 2020 lockdown data which provided ‘equivalent’ incidents with an additional year to progress through the CJS. Furthermore, there remained a substantial backlog of cases following court closures when these police datasets were sampled in February 2021 (Dearden, 2021). This may explain why, during lockdown, fewer IPV incidents resulted in official action, while unknown disposal types (incidents that have not yet had a result recorded on the RMS) increased.

Outcomes involving the courts dropped during the first lockdown, with a reduction in incidents where suspects were charged or summoned by 1.7pp (6.3% to 4.6%). Considering that courts were closed for a substantial period during the first lockdown, which caused a huge backlog of cases and drops in referrals and disposals for magistrates and crown courts (Criminal Justice Joint Inspection, 2021), it is unsurprising that the number of court outcomes reduced. However, the greatest decrease occurred for summoning, rather than charging. This is unusual as information for summoning must usually be laid before a magistrate within six months of the offence and would be expected to be recorded on the RMS during the seven months between police recording and sampling (Magistrates’ Courts Act, 1980). As summoning is usually a precursor to charging, the reduction in suspects summoned to court indicates that lockdown implications may have reduced the use of court disposals and affected the disposal types preferred by the police.

The pattern differed for out of court disposals (OOCD). Police cautions require admittance of guilt and are officially recorded, which provides a police record of perpetration, without attending court. Although these actions do not involve a charge, conditional cautions can result in a conviction where the attached conditions are ignored (MOJ, 2015). During lockdown the use of conditional cautions decreased by 0.6pp. Whereas simple cautions, which require suspects to admit their guilt and act as a formal warning for low-level offending (MOJ, 2015), increased by 0.5xspp. This inverted pattern may be explained by the difference between simple and conditional cautions. Considering the additional strain placed on the police during lockdown, through COVID-19 and ensuring that the public complied with lockdown restrictions (Justice Inspectorate, 2021b), it is likely that simple cautions were preferred as they do not require police to monitor compliance with conditions. Simple cautions enable police to resolve incidents, record IPV perpetration and may have been viewed as safer than conditional cautions for victims who were confined with their abusers and unwilling to pursue court action. Furthermore, the majority of IPV incidents reported during lockdown were classed as standard-risk, which are less likely to require a court disposal and may be suitable for less serious responses, such as an OOCD. These changes in disposal types were significantly correlated with lockdown (Table 3).

Overall, incidents ending in official action fell substantially by 1.9pp (7.5% to 5.6%), which was significantly correlated with lockdown (X2(1, N = 14191) = 0.621, p = <0.000). This occurred despite increases in total reported IPV and decreases in problems hindering official action. There was a 26.5pp reduction in instances where the victim would not support police action or identify the suspect (56.8% to 30.3%) and a 4.2 pp decrease in other prosecution problems (28.6% to 24.4%). The reduction in IPV incidents ending in official action likely occurred due to significantly lower rates of medium- and high-risk incidents reported, as higher-risk incidents are more likely to warrant an arrest, include evidential injuries and end in prosecution (Petersson and Strand, 2020). This suggests that police reactively altered their response to reflect changes identified in incident characteristics during lockdown, diverting official sanctions from incidents where they were deemed unnecessary (Joshi and Sorenson, 2010). Indeed, chi-squared tests confirmed that all changes in police response (arrest, DA risk level and disposal type) were significantly associated with changes in incident characteristics (whether a crime was recorded and suspect substance use).

Table 5:

Chi squared results for test of independence of police response with crimed

Table 5: Chi squared results are presented for whether an incident is crimed and the police response. All police responses (D A risk level, arrest, disposal type, official action) are significant, with the highest pearson value occurring for disposal type (13796.20) and the lowest occurring for official action (1078.72).
Crimed Significant changes Significance Pearson’s Chi 2 Likelihood Number Degrees of freedom Cramer’s V
Police response DA risk Level p<0.001 2275.30 2615.59 12620 2 0.425
Arrest p<0.001 2197.41 3054.55 14949 1 0.383
Disposal Type p<0.001 13796.20 16720.54 14150 3 0.987
Official Action p<0.001 1078.72 1562.98 14949 2 0.269
Table 6:

Chi squared results for test of independence of police response suspect substance Use

Table 6: Chi squared results are presented for substance use and the police response, all of which are significant (D A risk level, arrest, disposal type and official action). The highest pearson value occurs for arrest (631.81) while the other pearson values are all similar, with the lowest occurring for official action (99.39).
Substance use Significant changes Significance Pearson’s Chi 2 Likelihood Number Degrees of freedom Cramer’s V
Police response DA risk Level p<0.001 169.71 166.39 12620 2 0.116
Arrest p<0.001 631.81 588.18 14949 1 0.206
Disposal Type p<0.001 129.71 119.07 14150 3 0.096
Official Action p<0.001 99.39 91.67 14191 1 0.084

Summary of findings

During the first lockdown in England, IPV reported within this south-eastern police district increased, in line with initial fears. However, this increase was lower than expected, considering the rise in calls to DA services and charities. The age distribution changed for victims and perpetrators, with IPV decreasing among younger age groups and increasing within older age categories. Second, a shift occurred in the categorisation of IPV crimes with a decrease in physical crimes and an increase in crimes that can be committed remotely, most notably stalking and malicious communications. Finally, the risk level for reported IPV incidents differed, with a reduction in the number of incidents recorded as medium- and high-risk, while standard-risk incidents rose substantially, causing a change in the distribution of risk levels across reported incidents. This shift was reflected in arrest and official outcomes, with fewer court disposals and more simple cautions. However, despite the arrest rate decreasing overall, the arrest rate increased for medium- and high-risk incidents.

Discussion and research implications

IPV is an ongoing pattern of abusive behaviour that was recognised as a global epidemic prior to the pandemic (Stark, 2007). Rather than COVID-19 ‘causing’ IPV incidents, the rise in reported IPV mirrors annual increases and appears to be influenced by wider sociological, patriarchal and cultural factors that were exacerbated and overlapped during the pandemic, along with individual triggers such as isolation, anger and boredom (Williamson et al, 2020). This may have created ‘an enabling environment’ (Peterman et al, 2020) which may have contributed to increased IPV perpetration, following expectations that IPV rates would increase during lockdown (Bradbury-Jones and Isham, 2020).

Looking at the reduction in incidents reported to the police classed as medium- and high-risk, it should be remembered that IPV is a notoriously underreported crime (Bradbury‐Jones and Isham, 2020), with ONS (2017) finding that prior to the pandemic 79.0 per cent of victims did not report abuse to the police. Therefore, while it is very important to understand and recognise that the nature and severity of IPV reported to the police has changed during lockdown, this only represents one piece of the wider picture. Additionally, police lockdown policies altered risk classification by limiting the number of incidents regarded as medium- and high-risk, making it difficult to identify changes in IPV reported. More research is therefore required to understand how victim’s lived experience of IPV changed and if there was an underreporting of medium- and high-risk IPV incidents.

While these findings cannot confirm that more severe incidents of IPV were not reported, the large reduction in higher-risk incidents between lockdown and the equivalent timeframe indicates that victims of more severe IPV might not have reported to the police during lockdown (Moreira and Da Costa, 2020). Research by government, charities, academics and police agencies has similarly suggested that victims of more severe IPV may not have reported during lockdown (Justice Inspectorates, 2021b), as often occurs during periods of family proximity, such as holidays (Williamson et al, 2020). Further research is needed to determine how best to support victims’ needs, which could include alternative methods of reporting, such as online or community responses. Hence analysis of data from multiple DA services is needed to determine whether this suspicion can be confirmed.

This shift in crime type, from in-person to remote, makes it harder for the police due to the difficulties of regulating cybercrime and online platforms (Mayhew and Jahankhani, 2020). Additionally, the increase in low-level IPV reported may require consideration of alternative action, CJS responses and greater referrals to support services.

Contextualisation of quantitative data is necessary to understand how these findings intertwine and to investigate potential explanations for reductions in police recording of medium- and high-risk DA, use of arrest and disposal types during the first lockdown. Qualitative research would be beneficial to investigate what changes frontline police officers observed in victim/suspect demographics, incident characteristics and if their experience of responding to and negotiating IPV differed during the pandemic. As trends suggest a continued increase in IPV rates (Walklate et al, 2021), with COVID-19 shining a light on the pre-existing DA crisis (Hohl and Johnson, 2021), it is necessary that IPV remains high on police and government agendas.

Conclusion

Victim/suspect demographics, incident characteristics and police responses underwent significant change during the first English lockdown. Unsurprisingly, when restrictions prevented individuals from leaving their homes or meeting others, IPV increased between cohabiting partners. The greatest change within victim/suspect demographics occurred for victims and perpetrator age, with younger age groups experiencing a decrease, while IPV increased among those over 30. Incident characteristics demonstrated a rise in remote crimes, notably stalking which increased by 2.8pp, while physical and sexual abuse decreased. As non-cohabiting and ex-partners continued to constitute 53.7 per cent and 43.0 per cent of suspects respectively, this suggests that they found alternative means of abusing and controlling victims. Fewer incidents involved suspect substance use overall, especially alcohol, indicating that lockdown measures changed the nature of IPV reported to the police.

Police responses altered in reaction to these nuanced changes in victim/suspect demographics and incident characteristics. Overall reductions in reported IPV involving suspect substance use and physical abuse, alongside police prioritisation of high-risk incidents, were reflected in fewer incidents categorised as high- and medium-risk, with a large increase in the number of incidents classed as standard-risk. Subsequently, fewer suspects were arrested which contributed to fewer official outcomes. This substantial decrease in high- and medium-risk IPV reported to the police suggests that the already low reporting rates for IPV declined further during the first lockdown. This supports fears that higher-risk IPV victims were unwilling or unable to seek police help during lockdown.

Arrest rates increased for medium- and high-risk IPV, however, suggesting that police recognised the elevated dangers for IPV victims during lockdown. Court disposals reduced by 1.7pp, demonstrating greater disparity between reported IPV and official outcomes during the pandemic, as expected considering court closures and subsequent case backlogs. While simple cautions, reserved for low-level offending, were the only disposals to increase during lockdown. Overall, this article suggests that during lockdown, police officers altered their response to IPV reactively, to reflect changes identified in victim/suspect demographics and, most notably, incident characteristics.

Notes

1

The Home Office definition of DA is currently limited to individuals over 16 (Home Office, 2013).

2

Personal discussions with police force.

Acknowledgements

I would like to thank the Hampshire Constabulary (south-eastern police force) for kindly facilitating my research, especially the Domestic Abuse Strategic Lead, whose help has been invaluable in supporting my research.

Conflict of interest

The author declares that there is no conflict of interest.

References

  • Arenas-Arroyo, E., Fernandez-Kranz, D. and Nollenberger, N. (2021) Intimate partner violence under forced cohabitation and economic stress: evidence from the COVID-19 pandemic, Journal of Public Economics, 194: 104350. doi: 10.1016/j.jpubeco.2020.104350

    • Search Google Scholar
    • Export Citation
  • Bates, L., Hoeger, K., Stoneman, M.J. and Whitaker, A. (2021) Domestic homicides and suspected victim suicides during the covid-19 pandemic 2020–2021, 25 August, https://cdn.prgloo.com/media/02d412c416154010b5cebaf8f8965030.pdf.

    • Search Google Scholar
    • Export Citation
  • Bowstead, J. (2019) Women on the move: administrative data as a safe way to research hidden domestic violence journeys, Journal of Gender-Based Violence, 3(2): 23348. doi: 10.1332/239868019X15538575149704

    • Search Google Scholar
    • Export Citation
  • Bracewell, K., Hargreaves, P. and Stanley, N. (2020) The consequences of the COVID-19 lockdown on stalking victimization, Journal of Family Violence, doi: 10.1007/s10896-020-00201-0.

    • Search Google Scholar
    • Export Citation
  • Bradbury-Jones, C. and Isham, L. (2020) The pandemic paradox: the consequences of COVID-19 on domestic violence, Journal of Clinical Nursing, 29(13–14): 204749. doi: 10.1111/jocn.15296

    • Search Google Scholar
    • Export Citation
  • Brooks-Hay, O. and Lombard, N. (2018) ‘Home game’: domestic abuse and football, Journal of Gender-Based Violence, 2(1): 93108. doi: 10.1332/239868018X15155986580769

    • Search Google Scholar
    • Export Citation
  • Browne, A. (1989) When Battered Women Kill, New York: Free Press.

  • CPS (Crown Prosecution Service) (2013) Charging (The Director’s Guidance), May, https://www.cps.gov.uk/legal-guidance/charging-directors-guidance-2013-fifth-edition-may-2013-revised-arrangements.

    • Search Google Scholar
    • Export Citation
  • Criminal Justice Joint Inspection (2021) Impact of the Pandemic on the Criminal Justice System, London: Justice Inspectorates.

  • Dearden, L. (2021) Crown court backlog hits high, 24 June, https://www.independent.co.uk/news/uk/home-news/crown-court-backlog-coronavirus-cuts-b1872051.html.

    • Search Google Scholar
    • Export Citation
  • Dichter, M., Marcus, S., Morabito, M. and Rhodes, K. (2011) Explaining the IPV arrest decision: incident, agency, and community factors, Criminal Justice Review, 36(1): 2239. doi: 10.1177/0734016810383333

    • Search Google Scholar
    • Export Citation
  • Duerksen, K. and Woodin, E. (2019) Technological intimate partner violence, Computers in Human Behavior, 98(C): 22331. doi: 10.1016/j.chb.2019.05.001

    • Search Google Scholar
    • Export Citation
  • Friday, P., Lord, V., Exum, M. and Hartman, J. (2006) Evaluating the Impact of a Specialized Domestic Violence Police Unit, Washington, DC: National Institute of Justice.

    • Search Google Scholar
    • Export Citation
  • Graham, H. (2021) Meet the young people living with parents for lockdown, The Times, 28 February, https://www.thetimes.co.uk/article/meet-the-young-people-living-with-parents-for-lockdown-wkdlrpc9t.

    • Search Google Scholar
    • Export Citation
  • Grierson, J. (2020) Domestic abuse killings ‘more than double’ amid Covid-19 lockdown, The Guardian, 15 April, https://www.theguardian.com/society/2020/apr/15/domestic-abuse-killings-more-than-double-amid-covid-19-lockdown.

    • Search Google Scholar
    • Export Citation
  • Hartman, J. and Belknap, J. (2003) Beyond the gatekeepers: court professionals’ self-reported attitudes about and experiences with misdemeanor domestic violence cases, Criminal Justice and Behavior, 30(3): 34973. doi: 10.1177/0093854803030003005

    • Search Google Scholar
    • Export Citation
  • Hattery, A. (2009) Intimate Partner Violence, Washington, DC: Rowman and Littlefield Publishers.

  • Hester, M. and Radford, L. (2006) Mothering Through Domestic Violence, London: Jessica Kingsley Publishers.

  • Hester, M. and Westmarland, N. (2005) Tackling Domestic Violence: Effective Interventions and Approaches, Brighton: Home Office Research, Development and Statistics Directorate.

    • Search Google Scholar
    • Export Citation
  • Hohl, K. and Johnson, K. (2021) Home office preparedness for Covid-19 consultation, February, https://committees.parliament.uk/writtenevidence/22280/pdf/.

    • Search Google Scholar
    • Export Citation
  • Home Office (2013) Information for local areas on the change to the definition of domestic violence and abuse, March, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/142701/guide-on-definition-of-dv.pdf.

    • Search Google Scholar
    • Export Citation
  • Home Secretary (2020) Statement on domestic abuse and coronavirus, 11 April, https://www.gov.uk/government/speeches/home-secretary-outlines-support-for-domestic-abuse-victims.

    • Search Google Scholar
    • Export Citation
  • Johnson, M. (2010) A Typology of Domestic Violence: Intimate Terrorism, Violent Resistance, and Situational Couple Violence, Boston, MA: Northeastern University Press.

    • Search Google Scholar
    • Export Citation
  • Joshi, M. and Sorenson, S. (2010) Intimate partner violence at the scene: incident characteristics and implications for public health surveillance, Evaluation Review, 34(2): 11636. doi: 10.1177/0193841X09360323

    • Search Google Scholar
    • Export Citation
  • Justice Inspectorate (2021a) Policing in the pandemic, 20 April, https://www.justiceinspectorates.gov.uk/hmicfrs/publication-html/the-police-response-to-the-coronavirus-pandemic-during-2020/.

    • Search Google Scholar
    • Export Citation
  • Justice Inspectorate (2021b) Review of policing domestic abuse during the pandemic, 23 June, https://www.justiceinspectorates.gov.uk/hmicfrs/publication-html/review-of-policing-domestic-abuse-during-pandemic/.

    • Search Google Scholar
    • Export Citation
  • Magistrates’ Courts Act (1980) London: HMSO.

  • Mayhew, J. and Jahankhani, H. (2020) Current challenges of modern-day domestic abuse, in H. Jahankhani, B. Akhgar, P. Cochrane and M. Dastbaz (eds) Policing in the Era of AI and Smart Societies, New York: Springer International Publishing, pp 26782.

    • Search Google Scholar
    • Export Citation
  • MOJ (Ministry of Justice) (2015) Simple cautions for adult offenders, 13 April, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/708595/cautions-guidance-2015.pdf.

    • Search Google Scholar
    • Export Citation
  • Moreira, D. and da Costa, M. (2020) The impact of the Covid-19 pandemic in the precipitation of intimate partner violence, International Journal of Law and Psychiatry, 71: 101 606. doi: 10.1016/j.ijlp.2020.101606

    • Search Google Scholar
    • Export Citation
  • Myhill, A. (2019) Renegotiating domestic violence: police attitudes and decisions concerning arrest, Policing and Society, 29(1): 5268. doi: 10.1080/10439463.2017.1356299

    • Search Google Scholar
    • Export Citation
  • Myhill, A. and Johnson, K. (2016) Police use of discretion in response to domestic violence, Criminology and Criminal Justice, 16(1): 320. doi: 10.1177/1748895815590202

    • Search Google Scholar
    • Export Citation
  • Neale, J. (2018) Abused women’s perceptions of professionals’ responses: valued support, or collusion with perpetrator?, Journal of Gender-Based Violence, 2(3): 41127. doi: 10.1332/239868018X15366982612051

    • Search Google Scholar
    • Export Citation
  • Oliver, R., Alexander, B., Roe, S. and Wlasny, M. (2019) The economic and social costs of domestic abuse. London: Home Office.

  • ONS (Office for National Statistics) (2017) Domestic abuse in England and wales, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/bulletins/domesticabuseinenglandandwales/yearendingmarch2017.

    • Search Google Scholar
    • Export Citation
  • ONS (Office for National Statistics) (2019a) Domestic abuse, 22 November, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/articles/domesticabusefindingsfromthecrimesurveyforenglandandwales/yearendingmarch2017.

    • Search Google Scholar
    • Export Citation
  • ONS (Office for National Statistics) (2019b) Population estimates by marital status and living arrangements, 17 July, https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/bulletins/populationestimatesbymaritalstatusandlivingarrangements/2019.

    • Search Google Scholar
    • Export Citation
  • ONS (Office for National Statistics) (2020a) Homicide in England and Wales, 13 February, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/articles/homicideinenglandandwales/yearendingmarch2020#the-relationship-between-victims-and-suspects.

    • Search Google Scholar
    • Export Citation
  • ONS (Office for National Statistics) (2020b) Domestic abuse during the covid pandemic, 25 November, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/articles/domesticabuseduringthecoronaviruscovid19pandemicenglandandwales/november2020.

    • Search Google Scholar
    • Export Citation
  • ONS (Office for National Statistics) (2020c) Domestic abuse prevalence and trends, 25 November, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/articles/domesticabuseprevalenceandtrendsenglandandwales/yearendingmarch2020#understanding-domestic-abuse.

    • Search Google Scholar
    • Export Citation
  • Owens, C., Mann, D. and McKenna, R. (2014) The Essex BWV Trial, London: College of Policing.

  • Peterman, A., Potts, A., O’Donnell, M., Thompson, K., Shah, N., Oertelt-Prigione, S. and Van Gelder, N. (2020) Pandemics and violence against women and children. center for global development, https://www.cgdev.org/publication/pandemics-and-violence-against-women-and-children.

    • Search Google Scholar
    • Export Citation
  • Petersson, J. and Strand, S. (2020) Characteristics and recidivism in relation to arrest: differentiating between partner violent perpetrator subtypes, Nordic Journal of Criminology, 21(2): 20322. doi: 10.1080/2578983X.2020.1727119

    • Search Google Scholar
    • Export Citation
  • Sallie, S., Ritou, V., Bowden-Jones, H. and Voon, V. (2020) Assessing international alcohol consumption patterns during isolation from the COVID-19 pandemic using an online survey, British Medical Journal, 10(11): 110.

    • Search Google Scholar
    • Export Citation
  • Smith, K. (2020) Coronavirus doesn’t cause men’s violence against women, 15 April, https://kareningalasmith.com/2020/04/15/coronavirus-doesnt-cause-mens-violence-against-women/.

    • Search Google Scholar
    • Export Citation
  • Stalking Protection Act (2019) London: HMSO.

  • Stanko, B. (2007) From academia to policy making: changing police responses to violence against women, Theoretical Criminology, 11(2): 20919. doi: 10.1177/1362480607075848

    • Search Google Scholar
    • Export Citation
  • Stark, E. (2007) Coercive Control: The Entrapment of Women in Personal Life, Oxford: Oxford University Press.

  • Trujillo, M. and Ross, S. (2008) Police response to domestic violence: making decisions about risk and risk management, Journal of Interpersonal Violence, 23(4): 45473. doi: 10.1177/0886260507312943

    • Search Google Scholar
    • Export Citation
  • UNODC (United Nations Office on Drugs and Crime) (2018) Global study on homicide, November, Vienna: UNODC, https://www.unodc.org/documents/data-and-analysis/GSH2018/GSH18_Gender-related_killing_of_women_and_girls.pdf.

    • Search Google Scholar
    • Export Citation
  • UN Women (United Nations Women) (2020) COVID-19 and Ending Violence Against Women and Girls, New York: UN Women, https://www.unwomen.org/sites/default/files/Headquarters/Attachments/Sections/Library/Publications/2020/Issue-brief-COVID-19-and-ending-violence-against-women-and-girls-en.pdf.

    • Search Google Scholar
    • Export Citation
  • Virkki, T. (2007) Gender, care, and the normalization of violence: similarities between occupational violence and intimate partner violence in Finland, Nordic Journal of Women’s Studies, 15(4): 22032.

    • Search Google Scholar
    • Export Citation
  • Walby, S. and Allen, J. (2004) Domestic Violence, Sexual Assault and Stalking, London: Home Office.

  • Walklate, S., Godfrey, B. and Richardson, J. (2021) Changes and continuities in police responses to domestic abuse in England and Wales during the Covid-19 ‘lockdown’, Policing and Society, 32(2): 22133, doi: 10.1080/10439463.2021.1896514.

    • Search Google Scholar
    • Export Citation
  • Westmarland, N. (2015) Violence Against Women: Criminological Perspectives on Men’s Violence, Abingdon: Routledge.

  • WHO (World Health Organization) (2020) COVID-19 and Violence Against Women, Philippines: WHO.

  • Williamson, E., Lombard, N. and Brooks-Hay, O. (2020) Domestic violence and abuse, coronavirus, and the media narrative, Journal of Gender-Based Violence, 4(2): 28994.

    • Search Google Scholar
    • Export Citation
  • Women’s Aid (2020) A Perfect Storm: The Impact of the Covid-19 Pandemic on Domestic Abuse Survivors and the Services Supporting Them, Bristol: Women’s Aid, https://www.womensaid.org.uk/wp-content/uploads/2020/08/A-Perfect-Storm-August-2020-1.pdf.

    • Search Google Scholar
    • Export Citation
  • Zaami, S., Marinelli, E. and Varì, M.R. (2020) New trends of substance abuse during COVID-19 pandemic, Frontiers in Psychiatry, 11: 700, doi: 10.3389/fpsyt.2020.00700.

    • Search Google Scholar
    • Export Citation
  • View in gallery
    Figure 1:

    Victim age categories in lockdown and lockdown equivalent

  • View in gallery
    Figure 2:

    Suspect age categories in lockdown and lockdown equivalent

  • Arenas-Arroyo, E., Fernandez-Kranz, D. and Nollenberger, N. (2021) Intimate partner violence under forced cohabitation and economic stress: evidence from the COVID-19 pandemic, Journal of Public Economics, 194: 104350. doi: 10.1016/j.jpubeco.2020.104350

    • Search Google Scholar
    • Export Citation
  • Bates, L., Hoeger, K., Stoneman, M.J. and Whitaker, A. (2021) Domestic homicides and suspected victim suicides during the covid-19 pandemic 2020–2021, 25 August, https://cdn.prgloo.com/media/02d412c416154010b5cebaf8f8965030.pdf.

    • Search Google Scholar
    • Export Citation
  • Bowstead, J. (2019) Women on the move: administrative data as a safe way to research hidden domestic violence journeys, Journal of Gender-Based Violence, 3(2): 23348. doi: 10.1332/239868019X15538575149704

    • Search Google Scholar
    • Export Citation
  • Bracewell, K., Hargreaves, P. and Stanley, N. (2020) The consequences of the COVID-19 lockdown on stalking victimization, Journal of Family Violence, doi: 10.1007/s10896-020-00201-0.

    • Search Google Scholar
    • Export Citation
  • Bradbury-Jones, C. and Isham, L. (2020) The pandemic paradox: the consequences of COVID-19 on domestic violence, Journal of Clinical Nursing, 29(13–14): 204749. doi: 10.1111/jocn.15296

    • Search Google Scholar
    • Export Citation
  • Brooks-Hay, O. and Lombard, N. (2018) ‘Home game’: domestic abuse and football, Journal of Gender-Based Violence, 2(1): 93108. doi: 10.1332/239868018X15155986580769

    • Search Google Scholar
    • Export Citation
  • Browne, A. (1989) When Battered Women Kill, New York: Free Press.

  • CPS (Crown Prosecution Service) (2013) Charging (The Director’s Guidance), May, https://www.cps.gov.uk/legal-guidance/charging-directors-guidance-2013-fifth-edition-may-2013-revised-arrangements.

    • Search Google Scholar
    • Export Citation
  • Criminal Justice Joint Inspection (2021) Impact of the Pandemic on the Criminal Justice System, London: Justice Inspectorates.

  • Dearden, L. (2021) Crown court backlog hits high, 24 June, https://www.independent.co.uk/news/uk/home-news/crown-court-backlog-coronavirus-cuts-b1872051.html.

    • Search Google Scholar
    • Export Citation
  • Dichter, M., Marcus, S., Morabito, M. and Rhodes, K. (2011) Explaining the IPV arrest decision: incident, agency, and community factors, Criminal Justice Review, 36(1): 2239. doi: 10.1177/0734016810383333

    • Search Google Scholar
    • Export Citation
  • Duerksen, K. and Woodin, E. (2019) Technological intimate partner violence, Computers in Human Behavior, 98(C): 22331. doi: 10.1016/j.chb.2019.05.001

    • Search Google Scholar
    • Export Citation
  • Friday, P., Lord, V., Exum, M. and Hartman, J. (2006) Evaluating the Impact of a Specialized Domestic Violence Police Unit, Washington, DC: National Institute of Justice.

    • Search Google Scholar
    • Export Citation
  • Graham, H. (2021) Meet the young people living with parents for lockdown, The Times, 28 February, https://www.thetimes.co.uk/article/meet-the-young-people-living-with-parents-for-lockdown-wkdlrpc9t.

    • Search Google Scholar
    • Export Citation
  • Grierson, J. (2020) Domestic abuse killings ‘more than double’ amid Covid-19 lockdown, The Guardian, 15 April, https://www.theguardian.com/society/2020/apr/15/domestic-abuse-killings-more-than-double-amid-covid-19-lockdown.

    • Search Google Scholar
    • Export Citation
  • Hartman, J. and Belknap, J. (2003) Beyond the gatekeepers: court professionals’ self-reported attitudes about and experiences with misdemeanor domestic violence cases, Criminal Justice and Behavior, 30(3): 34973. doi: 10.1177/0093854803030003005

    • Search Google Scholar
    • Export Citation
  • Hattery, A. (2009) Intimate Partner Violence, Washington, DC: Rowman and Littlefield Publishers.

  • Hester, M. and Radford, L. (2006) Mothering Through Domestic Violence, London: Jessica Kingsley Publishers.

  • Hester, M. and Westmarland, N. (2005) Tackling Domestic Violence: Effective Interventions and Approaches, Brighton: Home Office Research, Development and Statistics Directorate.

    • Search Google Scholar
    • Export Citation
  • Hohl, K. and Johnson, K. (2021) Home office preparedness for Covid-19 consultation, February, https://committees.parliament.uk/writtenevidence/22280/pdf/.

    • Search Google Scholar
    • Export Citation
  • Home Office (2013) Information for local areas on the change to the definition of domestic violence and abuse, March, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/142701/guide-on-definition-of-dv.pdf.

    • Search Google Scholar
    • Export Citation
  • Home Secretary (2020) Statement on domestic abuse and coronavirus, 11 April, https://www.gov.uk/government/speeches/home-secretary-outlines-support-for-domestic-abuse-victims.

    • Search Google Scholar
    • Export Citation
  • Johnson, M. (2010) A Typology of Domestic Violence: Intimate Terrorism, Violent Resistance, and Situational Couple Violence, Boston, MA: Northeastern University Press.

    • Search Google Scholar
    • Export Citation
  • Joshi, M. and Sorenson, S. (2010) Intimate partner violence at the scene: incident characteristics and implications for public health surveillance, Evaluation Review, 34(2): 11636. doi: 10.1177/0193841X09360323

    • Search Google Scholar
    • Export Citation
  • Justice Inspectorate (2021a) Policing in the pandemic, 20 April, https://www.justiceinspectorates.gov.uk/hmicfrs/publication-html/the-police-response-to-the-coronavirus-pandemic-during-2020/.

    • Search Google Scholar
    • Export Citation
  • Justice Inspectorate (2021b) Review of policing domestic abuse during the pandemic, 23 June, https://www.justiceinspectorates.gov.uk/hmicfrs/publication-html/review-of-policing-domestic-abuse-during-pandemic/.

    • Search Google Scholar
    • Export Citation
  • Magistrates’ Courts Act (1980) London: HMSO.

  • Mayhew, J. and Jahankhani, H. (2020) Current challenges of modern-day domestic abuse, in H. Jahankhani, B. Akhgar, P. Cochrane and M. Dastbaz (eds) Policing in the Era of AI and Smart Societies, New York: Springer International Publishing, pp 26782.

    • Search Google Scholar
    • Export Citation
  • MOJ (Ministry of Justice) (2015) Simple cautions for adult offenders, 13 April, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/708595/cautions-guidance-2015.pdf.

    • Search Google Scholar
    • Export Citation
  • Moreira, D. and da Costa, M. (2020) The impact of the Covid-19 pandemic in the precipitation of intimate partner violence, International Journal of Law and Psychiatry, 71: 101 606. doi: 10.1016/j.ijlp.2020.101606

    • Search Google Scholar
    • Export Citation
  • Myhill, A. (2019) Renegotiating domestic violence: police attitudes and decisions concerning arrest, Policing and Society, 29(1): 5268. doi: 10.1080/10439463.2017.1356299

    • Search Google Scholar
    • Export Citation
  • Myhill, A. and Johnson, K. (2016) Police use of discretion in response to domestic violence, Criminology and Criminal Justice, 16(1): 320. doi: 10.1177/1748895815590202

    • Search Google Scholar
    • Export Citation
  • Neale, J. (2018) Abused women’s perceptions of professionals’ responses: valued support, or collusion with perpetrator?, Journal of Gender-Based Violence, 2(3): 41127. doi: 10.1332/239868018X15366982612051

    • Search Google Scholar
    • Export Citation
  • Oliver, R., Alexander, B., Roe, S. and Wlasny, M. (2019) The economic and social costs of domestic abuse. London: Home Office.

  • ONS (Office for National Statistics) (2017) Domestic abuse in England and wales, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/bulletins/domesticabuseinenglandandwales/yearendingmarch2017.

    • Search Google Scholar
    • Export Citation
  • ONS (Office for National Statistics) (2019a) Domestic abuse, 22 November, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/articles/domesticabusefindingsfromthecrimesurveyforenglandandwales/yearendingmarch2017.

    • Search Google Scholar
    • Export Citation
  • ONS (Office for National Statistics) (2019b) Population estimates by marital status and living arrangements, 17 July, https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/bulletins/populationestimatesbymaritalstatusandlivingarrangements/2019.

    • Search Google Scholar
    • Export Citation
  • ONS (Office for National Statistics) (2020a) Homicide in England and Wales, 13 February, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/articles/homicideinenglandandwales/yearendingmarch2020#the-relationship-between-victims-and-suspects.

    • Search Google Scholar
    • Export Citation
  • ONS (Office for National Statistics) (2020b) Domestic abuse during the covid pandemic, 25 November, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/articles/domesticabuseduringthecoronaviruscovid19pandemicenglandandwales/november2020.

    • Search Google Scholar
    • Export Citation
  • ONS (Office for National Statistics) (2020c) Domestic abuse prevalence and trends, 25 November, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/articles/domesticabuseprevalenceandtrendsenglandandwales/yearendingmarch2020#understanding-domestic-abuse.

    • Search Google Scholar
    • Export Citation
  • Owens, C., Mann, D. and McKenna, R. (2014) The Essex BWV Trial, London: College of Policing.

  • Peterman, A., Potts, A., O’Donnell, M., Thompson, K., Shah, N., Oertelt-Prigione, S. and Van Gelder, N. (2020) Pandemics and violence against women and children. center for global development, https://www.cgdev.org/publication/pandemics-and-violence-against-women-and-children.

    • Search Google Scholar
    • Export Citation
  • Petersson, J. and Strand, S. (2020) Characteristics and recidivism in relation to arrest: differentiating between partner violent perpetrator subtypes, Nordic Journal of Criminology, 21(2): 20322. doi: 10.1080/2578983X.2020.1727119

    • Search Google Scholar
    • Export Citation
  • Sallie, S., Ritou, V., Bowden-Jones, H. and Voon, V. (2020) Assessing international alcohol consumption patterns during isolation from the COVID-19 pandemic using an online survey, British Medical Journal, 10(11): 110.

    • Search Google Scholar
    • Export Citation
  • Smith, K. (2020) Coronavirus doesn’t cause men’s violence against women, 15 April, https://kareningalasmith.com/2020/04/15/coronavirus-doesnt-cause-mens-violence-against-women/.

    • Search Google Scholar
    • Export Citation
  • Stalking Protection Act (2019) London: HMSO.

  • Stanko, B. (2007) From academia to policy making: changing police responses to violence against women, Theoretical Criminology, 11(2): 20919. doi: 10.1177/1362480607075848

    • Search Google Scholar
    • Export Citation
  • Stark, E. (2007) Coercive Control: The Entrapment of Women in Personal Life, Oxford: Oxford University Press.

  • Trujillo, M. and Ross, S. (2008) Police response to domestic violence: making decisions about risk and risk management, Journal of Interpersonal Violence, 23(4): 45473. doi: 10.1177/0886260507312943

    • Search Google Scholar
    • Export Citation
  • UNODC (United Nations Office on Drugs and Crime) (2018) Global study on homicide, November, Vienna: UNODC, https://www.unodc.org/documents/data-and-analysis/GSH2018/GSH18_Gender-related_killing_of_women_and_girls.pdf.

    • Search Google Scholar
    • Export Citation
  • UN Women (United Nations Women) (2020) COVID-19 and Ending Violence Against Women and Girls, New York: UN Women, https://www.unwomen.org/sites/default/files/Headquarters/Attachments/Sections/Library/Publications/2020/Issue-brief-COVID-19-and-ending-violence-against-women-and-girls-en.pdf.

    • Search Google Scholar
    • Export Citation
  • Virkki, T. (2007) Gender, care, and the normalization of violence: similarities between occupational violence and intimate partner violence in Finland, Nordic Journal of Women’s Studies, 15(4): 22032.

    • Search Google Scholar
    • Export Citation
  • Walby, S. and Allen, J. (2004) Domestic Violence, Sexual Assault and Stalking, London: Home Office.

  • Walklate, S., Godfrey, B. and Richardson, J. (2021) Changes and continuities in police responses to domestic abuse in England and Wales during the Covid-19 ‘lockdown’, Policing and Society, 32(2): 22133, doi: 10.1080/10439463.2021.1896514.

    • Search Google Scholar
    • Export Citation
  • Westmarland, N. (2015) Violence Against Women: Criminological Perspectives on Men’s Violence, Abingdon: Routledge.

  • WHO (World Health Organization) (2020) COVID-19 and Violence Against Women, Philippines: WHO.

  • Williamson, E., Lombard, N. and Brooks-Hay, O. (2020) Domestic violence and abuse, coronavirus, and the media narrative, Journal of Gender-Based Violence, 4(2): 28994.

    • Search Google Scholar
    • Export Citation
  • Women’s Aid (2020) A Perfect Storm: The Impact of the Covid-19 Pandemic on Domestic Abuse Survivors and the Services Supporting Them, Bristol: Women’s Aid, https://www.womensaid.org.uk/wp-content/uploads/2020/08/A-Perfect-Storm-August-2020-1.pdf.

    • Search Google Scholar
    • Export Citation
  • Zaami, S., Marinelli, E. and Varì, M.R. (2020) New trends of substance abuse during COVID-19 pandemic, Frontiers in Psychiatry, 11: 700, doi: 10.3389/fpsyt.2020.00700.

    • Search Google Scholar
    • Export Citation
Lucy TraffordUniversity of Oxford, UK

Search for other papers by Lucy Trafford in
Current site
Google Scholar
PubMed
Close

Content Metrics

May 2022 onwards Past Year Past 30 Days
Abstract Views 44 44 0
Full Text Views 932 932 218
PDF Downloads 636 636 140

Altmetrics

Dimensions