Research and analysis

Key findings from analysis of domestic homicide reviews: September 2021 to October 2022 (accessible)

Updated 31 October 2024

Prepared for the Home Office by Analytics Cambridge and QE Assessments Ltd

Responsible author:

Richard Potter, Independent Researcher, BSc (Hons), MPhil, Dip Management, fellow of Royal Statistical Society at AnalyticsCambridge

Contact: info@analyticscambridge.co.uk

Reviewed by:

Simon Kerss, MA, BA (Hons.), PA Dip. Honorary Fellow (Policing Institute for the Eastern Region)

Gaynor Mears OBE, MA, BA (Hons), AASW, Dip SW

1. Executive Summary

Statistics from Domestic Homicide Reviews

This report summarises information from Domestic Homicide Reviews (DHRs) which went before the Home Office Quality Assurance Panel for the 12 months between September 2021 and October 2022. DHRs are multi-agency reviews into the deaths of adults which may have resulted from violence, abuse, or neglect by a person to whom they were related or with whom they were, or had been, in an intimate personal relationship. Reviews also take place where a victim took their own life (died by suicide) where there is a known history of domestic abuse.

This report provides information for each of the three types of victims: familial, intimate partner, or who died by suicide. It summarises learning the reviews identify as areas which can be improved. It also looks at recommendations in the reviews and family contributions to the reviews.

Overview

• In the 129 DHRs reviewed there are 132 victims: 24% had a familial relationship with the perpetrator(s), for 50% the relationship with the perpetrator was partner or ex-partner. Twenty six per cent were victims who died by suicide.

• The average age of familial abuse victims was 55 years, older than the average age of familial perpetrators which was 35 years. Intimate partner victims were on average younger (38 years) and also younger than preparators (43 years). The average age of victims who died by suicide was 36 years.

• Where victims were in an intimate partner relationship or who had died by suicide, 86% and 88% respectively were female. This was different where there was a familial relationship where 53% of the victims were female.

• Considering nationality, 69% of familial victims were British; 80% of intimate partner victims were British and where the victims died by suicide 91% were British.

Victims

• The DHRs include assessments of the vulnerabilities of victims, considering illicit drug use, mental ill-health, physical disability, pregnancy, problem alcohol use, as well as any other vulnerability. Overall, 70% of all victims were considered to have at least one vulnerability. Where victims were familial 47% were considered to have at least one vulnerability, this was 68% for intimate partner victims and 94% for victims who had died by suicide.

• The two vulnerabilities where there were the largest differences between the types of victim were mental ill-health (29% of familial victims and 49% of suicide victims) and physical disability (21% of familial and 2% of suicide victims).

• Fifty two per cent of all victims had been the target of an abuser before. In familial relationships this was 22%, for intimate relationships 56% and for victims who had died by suicide it was 70%.

• The DHRs considered whether victims had experienced aggravating factors: coercive control, digital stalking, financial abuse, forced marriage, honour-based violence, immigration issues, or physical stalking. At least one aggravating factor was identified for 26% of familial victims, for 78% of intimate partner victims and for 88% of victims who died by suicide. The most common aggravating factors were coercive control and financial abuse.

Perpetrators

• As with victims, vulnerabilities were recorded for perpetrators and 77% were considered as having at least one, with mental ill-health, problem alcohol use and illicit drug use being the most common.

• Forty per cent of perpetrators were managed or supervised by mental health services and 32% by Probation.

• Sixty four per cent of perpetrators had also abused previous partners or family members. And 56% of perpetrators were known to agencies as an abuser.

Key themes from lessons to be learned

• This report looks at the DHRs to summarise and give examples of lessons learned. These are given below, ordered with the lessons most referred to listed first:

  • where DHRs examined familial abuse the lesson themes identified were assessments, risk, information, awareness, family, support, care and children;

  • for intimate partner victims: risk, support, information, safeguarding, contact, family and children;

  • where victims died by suicide: support, information, risk, training, staff and review.

Recommendations in Domestic Homicide Reviews

• The main agencies with responsibility for recommendations were health, the Police and Community Safety Partnerships. Of agencies within health, those with the greatest responsibility for the recommendations were NHS Trusts and Clinical Commissioning Groups.

• The main themes across the recommendations are similar to the learning points identified: assessment, care, information, review, risk, staff and training. In addition, there were recommendations which referred to working with communities.

2. Introduction

1.This report summarises information from Domestic Homicide Reviews (DHRs) which were subject to the Home Office quality assurance process ^{[footnote 1]} for the twelve months from October 2021 to the end of September 2022. It follows previous reports for the similar years 2019/20 ^{[footnote 2]} and 20/21 ^{[footnote 3]}. The aim is to provide analysis from DHRs which involved intimate partner relationships, those where the relationships were familial, and those where the victim died by suicide.

2. Statutory guidance from the Home Office (2016) states ^{[footnote 4]} : a Domestic Homicide Review is a multi-agency review, commissioned by a Community Safety Partnership, into the circumstances in which the death of a person aged 16 or over has, or appears to have, resulted from violence, abuse, or neglect by a person to whom they were related or with whom they were, or had been, in an intimate personal relationship, or a member of the same household. Reviews should also take place where a victim took their own life (suicide) and the circumstances give rise to concern. The purpose of a DHR is to:

• Establish lessons to be learned from the domestic homicide for the way local professionals and organisations can work individually and together to safeguard victims;

• Identify lessons both within and between agencies, how and within what timescales they will be acted on, and what is expected to change as a result;

• Apply the lessons to service responses;

• Prevent domestic violence and homicide and improve service responses for all domestic violence and abuse victims and their children by developing a co- ordinated multi-agency approach to ensure that abuse is identified and responded to effectively at the earliest opportunity;

• Contribute to a better understanding of the nature of domestic violence and abuse; and

• Highlight good practice.

3. The Domestic Homicide Review does not replace the criminal or Coronial processes.

Structure

4.The report starts with information from the Office for National Statistics on trends in domestic homicide. It then follows questions in the management information reports (MIRs) which are submitted to the Home Office alongside DHRs:

a)The location, age, sex and ethnicity of victims and perpetrators. This also gives information on nationality;

b)Characteristics or experience of victims in terms of their vulnerability, mental health, and whether they had been the target of an abuser before;

c)Characteristics or experience of perpetrators, including vulnerabilities and mental health, any previous offending history, and details of criminal charges; and

d)Contributions from and support for families in the DHR process.

5. The data in the management information reports has been added to or edited if information from the DHRs indicates this is needed.

6. The report then summarises lessons learned and recommendations from the DHRs by showing which agencies are involved, by identifying themes, and giving examples. This is done for each group of victims.

7. Within this context, this report is of 129 Domestic Homicide Reviews which were reviewed by the Home Office Quality Assurance Panel from October 2021 to the end of September 2022.

8. Within these there have been 136 victims and this report looks in more detail at 132 of these ^{[footnote 5]}. Of these (132) victims, thirty two (24%) had a familial relationship with the perpetrator(s). For sixty six victims (50%) the relationship with the perpetrator was or had been intimate. Thirty four victims (26%) had died by suicide.

9. Information analysed here is on 92 perpetrators. This is smaller as information required for a perpetrator convicted of a homicide is not applicable where the victim died by suicide. Of the 92, 27 (29%) had a family relationship with the victim and 65 (71%) had or previously had an intimate relationship with the victim.

Interpretations of numbers

10. Information in the management information reports (MIRs) is used to show patterns and differences, but they are not precise. As one example of the difficulty of being certain, in six reviews the date of death is not exact e.g. “between 17th February and 28th September 2019”^{[footnote 6]}.

11. Not every piece of information asked for in the MIRs can be found from some reviews, and answers can be given as “not known” ^{[footnote 7]} or left blank. This varies between questions. For example, for 92 perpetrators, there are three where the nationality is marked as N/K. For the question ‘Any serious or life limiting illness?’ there are 20 marked as not known. The bottom rows in the tables in this report give the numbers of answers on which percentages in the rows above are calculated and where the information has been recorded (therefore where answer given as not known these are excluded).

12. The answers requested for some questions on the MIRs are Y, N or N/K. Answers for some questions (such as vulnerability and mental health issues) are “please mark ‘X’ for ALL that apply”. These do not ask for N/K. The figures used are from MIRs where vulnerabilities have been identified. It could be argued that these may have been slightly higher if the answers had been asked to separate N and N/K.

13. To help compare figures between the different descriptions of victims, they are given as percentages. There is a balance between putting the data into more categories to allow comparison against making differences dependent on small numbers. Caution should be applied where differences in percentages are relatively small.

14. Herein percentages are rounded to the nearest whole number, therefore there will be occasions when they do not add to 100 per cent.

3. Domestic Homicide Reviews: trends, location, and demography

15. This chapter begins with national trends in domestic homicides. It then describes information from the Domestic Homicide Reviews on the dates of death, and the time between the death and when the reviews were submitted to the Home Office. This is followed with the number of reviews in each region. The chapter then provides information on the victims and the perpetrators including their age, sex, and relationships.

Trends in domestic homicides in England and Wales, 2012/13 to 2022/23

16. For context, Figure 1 shows the number of victims of domestic homicide over the 10 year period 2012/13 to 2122/23. There is a fall in the average of 137 for the first three years compared to 118 for the last three years. Looking at these two periods, the proportion of female victims has fallen from 77% to 65% and the proportion of victims who are male has increased from 23% to 35%. These figures on domestic homicides do not include people who have died by suicide.

Figure 1 Number of domestic homicides in England and Wales: 2011/12 to 2021/22

Source: Office for National Statistics, Homicide in England and Wales: year ending March 2023 - Appendix Tables Table 32 [Accessed 2nd March 2024]

Domestic Homicide Reviews: date of death of victim ^{[footnote 8]}

17. The time between the date of a victim’s death and the completion of the review is influenced by a range of factors, including:

• Length of time of police investigation;

• Completion of the criminal trial;

• Coroner’s Inquest;

• Contact with family members and others to enable them to contribute to the review;

• Community Safety Partnership meetings; report sign off and submission to the Home Office; and

• Quality assurance process through the Home Office.

18. Figure 2 shows the years in which the victims in the reviews died. The largest number of reviews (45, or 35%) was for victims who died in 2019. For 2020 this was 36 reviews and these two years together account for 63% of the total.

19. Thirty of the domestic homicides took place after the COVID lock down date of 23rd March 2020. Eighty one per cent referred to COVID ^{[footnote 9]}. For some this included issues around COVID and domestic abuse – where the homicide or suicide took place during the lockdown and restrictions on movement and social interaction imposed during the COVID pandemic. Other references to COVID concern the pandemic’s impact on the review process including the extra time required for the completion of the DHR.

Figure 2 Year of death of victims in DHRs

Domestic Homicide Reviews where the victim died by suicide, over time

20. Of the 129 Domestic Homicide Reviews in this report, 26% (33) involved victims who died by suicide. This is a higher proportion than the two previous Analysis of DHRs reports ^{[footnote 10]}, and is an increase in DHRs reviewing where victims have died by suicide when the suicide rate for England and Wales has not changed ^{[footnote 11]}.

Location of the deaths of victims

21. Table 1 shows where the reviews took place by region in England and in Wales ^{[footnote 12]}. Relating these to the number of people aged 16 or over, the North West has the highest rate of DHRs (4.3) with the lowest being London (1.4). The boundaries of the regions are shown in Figure 3.

22. The 129 DHRs were in 83 Local Authorities in England and 5 in Wales. In 23 Local Authorities in England, and one in Wales, more than one DHR was completed for the 2021/22 period of this report. The number of reviews of victims who died by suicide range from none in Wales to nine in North West England (which feeds into the highest number of reviews for the numbers of people aged 16 or older).

Table 1 Number of Domestic Homicide Reviews by region or nation

Region / Nation	Number of reviews of victims who died by suicide	Total number of reviews	Number of reviews per (one million) population aged 16 and older
North East	1	8	3.7
North West	9	26	4.3
Yorkshire and the Humber	3	10	2.2
East Midlands	2	15	3.8
West Midlands	1	14	2.9
East	3	15	2.9
London	3	10	1.4
South East	4	13	1.7
South West	7	12	2.5
England	33	123	2.7
Wales	0	6	2.3
England and Wales	33	129	2.7

Note: The number of reviews per million persons aged 16 or over is calculated using the population aged 16+ from the 2021 Census, sourced from NOMIS

Figure 3 Map of regional boundaries

Source: Ordnance Survey election maps

23. How victims are related to place can also be examined through the urban to rural dimension. The Department for Environment, Food & Rural Affairs has put Local Authorities into types ranging from mainly rural to urban with major conurbation. Table 2 shows that when relating the number of victims to the number of adults, those Local Authorities with the highest rates are “urban with minor conurbation” and the second highest rates are in those which are “largely rural”.

Table 2 Domestic homicides in urban and rural areas

Urban to rural descriptions	Number of victims	Number of victims per million people aged 18 or over
Mainly rural	9	2.4
Largely rural	22	4.3
Urban with significant rural	9	1.7
Urban with city and town	32	2.9
Urban with minor conurbation	9	5.2
Urban with major conurbation	42	2.7
Total	129

Notes:

The number of reviews per million persons aged 18 or over is calculated using the population age 18+ from the Office for National Statistics population estimates for mid-2021 [Accessed 22nd February 2024].

The Department for Environment, Food & Rural Affairs classification of Local Authorities in Rural Urban Classification 2011 lookup tables for local authority areas, last updated on: 17 October 2023 [Accessed 22nd February 2024].

24. Another aspect recorded in the DHRs is whether reviews were of deaths at the home address of the victim ^{[footnote 13]}.

25. Table 3 shows the proportion of reviews where deaths have occurred at the home address: 86% for victims who were in a familial relationship and 78% for those who died by suicide. Where victims were or had been in an intimate relationship the lower proportion of 68% of their deaths were at the home of the victim.

Table 3 Whether death of victims at home address

Location	Familial	Intimate partner	Victims who died by suicide	Overall
At home address	86%	68%	78%	75%
Not at home address	14%	32%	22%	25%
Total reviews	28	60	32	120

Notes: above is the number of reviews and does not include all victims if there was more than one. It is reviews where the home address / not at home address is known.

Age of victims and perpetrators

26. As can be seen in Table 4 the average ^{[footnote 14]} age of victims where there was a familial relationship was 55 years, this is older than victims where there was an intimate relationship or where the victim died by suicide.

27. Where the relationship had been intimate, the age of perpetrators was on average five years older than the victim. For familial relationships the average age of the perpetrator was younger than that of the victims (on average 20 years younger). A following section in this report looks at relationships between victims and perpetrators, noting that the most frequent familial relationship was where the victim was a parent of the perpetrator.

Table 4 Average age of victims and perpetrators, by type of victim

Type of victim	Average age (years): Victims	Average age (years): Perpetrators
Familial relationship	55	35
Intimate partner relationship	38	43
Victim who died by suicide	36
All	40	39

Numbers of victims and perpetrators

28. The information on date and region where the victim(s) died was shown from counts of Domestic Homicide Reviews. The management information forms for each DHR, on which this report is based, then move on to information on victims and perpetrators as individuals. There are a small number of DHRs where there is more than one victim and some where there is more than one perpetrator.

29. From 129 DHRs this report examines information relating to 132 victims ^{[footnote 15]}. Four of these victims are in DHRs where in each there were two victims and one perpetrator.

30. The report also examines information on 93 perpetrators ^{[footnote 16]}. The number of perpetrators is from DHRs which are homicides and not those where victims died by suicide. It includes three DHRs where there were two perpetrators.

31. There can be some information on victims or perpetrators which has been marked as “not known” or is missing. Each table and figure give the totals on which the percentages have been calculated (and therefore exclude the “not known”).

32.One DHR reviewed two suicides.

Defining relationships between victims and perpetrators

33.This report seeks to improve understanding of the victims and perpetrators in Domestic Homicide Reviews by comparing information on victims and perpetrators who had a familial relationship, or who were or had been in an intimate relationship, and also those victims who died by suicide. Table 5 shows the number of victims within each of the three groups.

34. Sixty six (50%) of the victims had or were previously in an intimate relationship with the perpetrator ^{[footnote 17]}. For 25% (17) this was a former relationship.

Table 5 Victims in Domestic Homicide Reviews including relationship with perpetrators

Characteristics of victims associated in DHRs	Number of victims	Per cent
Familial relationship with perpetrator	32	24%
In or had been in an intimate relationship with perpetrator:	66	50%
of which current relationship	50
of which former relationship	16
Died by suicide	34	26%
Total	132

35. For the 32 victims who had a family relationship with the perpetrator the most common familial relationship is where the victim was a parent of the perpetrator (22 or 69% of the 32 victims). For the 22 victims who were a parent of the perpetrator 8 were fathers and 14 were mothers. More detail is given in Table 6.

Table 6 Victims by types of familial relationship with perpetrators

Type of familial relationship between victim and perpetrator	Number of victims	Per cent of victims
Parent:	22	69%
of which father	8
of which mother	14
Grandparent	2	7%
Filial (e.g. son or daughter)	4	14%
Sibling (e.g. brother or sister)	3	7%
Other	1	3%
Total	32

Notes: the groups include small numbers which are in-law relationships (one mother-in-law and one ex-son-in-law), and two of the fathers were step fathers.

Sex of victims and perpetrators

36. The sex of victims is shown in

37. Table 7. 79% were female and 21% were male. For intimate partner victims and those who died by suicide the proportions who were female were similar (86% and 88%). For victims who died by suicide the balance between genders is different to the national picture (of all who died by suicide), where “Males continued to account for three- quarters of suicide deaths registered in 2022”^{[footnote 18]}.

38. Where victims were familial there is a more even balance between female and male with 53% of the victims female and 47% male.

Table 7 Sex of victims, per cent by type of victim

Sex	Familial	Intimate partner	Victims who died by suicide	Overall
Female	53%	86%	88%	79%
Male	47%	14%	12%	21%
Total	32	66	34	133

39. The sex of both familial perpetrators and intimate partner is shown in Table 8. This is not an exact reverse of the sex of victims - as the sex of the victim and perpetrator can be the same. There are 12 victims within a familial relationship with the same sex as the perpetrator. Of these eight are where both were male, and four where both were female.

Table 8 Sex of perpetrators, per cent by type of perpetrator

Sex	Familial	Intimate partner	Overall
Female	19%	15%	16%
Male	81%	85%	84%
Total	27	65	92

40. The data on sex is from the management information forms. More detailed examination of the DHRs shows one where the question of sex is more complex for example: “XXX’s struggle with their gender identity …”

Ethnicity

41.The ethnicity of victims and perpetrators and the population of England and Wales aged over 18 (from the 2021 Census) is shown in Table 9. There are close similarities in the proportion of all ethnic groups given and that of victims. A difference with perpetrators in that the proportion who are Black, African, Caribbean or Black British is six percentage points higher than that for the populations aged 18 and over ^{[footnote 19]}. The proportion of white perpetrators is five percentage points lower.

42. The National Institute for Health and Care Excellence set out that “Domestic violence and abuse occurs across the whole of society, regardless of race, ethnicity, gender, religion, age, class and economic status, or where people live ^{[footnote 20]}. Aspects of these are not collected in the forms with the Domestic Homicide Reviews. The Office for National Statistics report on Homicide in England and Wales^{[footnote 21]} (page 20) makes a similar point: “differences in … figures are likely to be related to the ethnicity of the population differing by age, region, and socioeconomic factors which have not been taken into account”.

Table 9 Per cent of victims and perpetrators by ethnicity

Ethnicity	Percent of DHR victims	Per cent of DHR perpetrators	Per cent of population in 2021, aged 18+, by ethnic group
Asian / Asian British	10%	9%	9%
Black / African / Caribbean / Black British	4%	10%	4%
Other or multiple ethnic group	7%	7%	5%
White: any other white background	79%	75%	84%
Number of victims or perpetrators	131	92

Notes: the per cent of the population ages 18 or over are from the 2021 Census. Office for National Statistics (2023), Ethnic group by age and sex in England and Wales

43.The ethnicity of victims is shown in Table 10. The ethnicity is shown as either white or non-white to reduce the impact of small numbers on the data. Compared to the ethnicity of all victims, victims of a familial perpetrator are slightly less likely to be white and those who have died by suicide are more likely to be white.

Table 10 Ethnicity by type of victim

Ethnicity	Familial	Intimate partner	Victims who died by suicide	Total
White	66%	80%	91%	82%
Non-white	34%	20%	9%	18%
Number of victims	32	65	34	131

Nationality

44. The nationality of victims is that 80% (of 131) are British and 20% are non-British. In more detail of those not British, 12 victims are from Europe: Poland (7), Lithuania (3) and Romania (2). Eleven per cent of victims (14) are from 14 different counties outside Europe.

45. The nationality and type of victim is shown in

46. Table 11. Victims who have died by suicide are 91% British; where there has been a familial relationship between victim and perpetrator there is less likelihood of the victim being British (69%).

Table 11 Nationality and type of victim

Nationality	Familial	Intimate partner	Victim who died by suicide	Overall
British	69%	80%	91%	80%
Non-British	31%	20%	9%	20%
Number of victims	32	64	34	131

47. The nationality of the victims and perpetrators has been compared for 87 victims. These are victims who have not died by suicide and where the nationality of the victim and perpetrator are known. The results are shown in Table 12. In 74% of the domestic homicides both victim and perpetrator were British. The second largest proportion (18%) is where both were non-British. Eleven of these are where the victim had been an intimate partner, five where the relationship had been familial. Where the nationality of both victim and perpetrator was non-British, in 12 of these the nationality of both victim and perpetrator was the same e.g. both were Polish or both South African.

Table 12 Comparing nationality of victim and perpetrator

Comparative Nationality	Overall
Both British	74%
Both non-British	18%
One British	8%
Number of victims	87

Information on children aged under 18 years

48. One question on the management information form is “were there any children living, or regularly staying in the household?” There had been a change in the question wording between older and newer forms; the answers from the newer forms are used here.

49. In 42% of the DHRs there were children living, or regularly staying in the household (Table 13). There are differences between the different categories of victim. For familial victims there was one review where children aged under 18 were living or staying in the household (and 15 where there were no children in the household). Where victims died by suicide in 43% of the reviews there were children and, where the perpetrator was or formerly an intimate partner it was 55%.

Table 13 Children living, or regularly staying in the household

	Familial	Intimate partner	Victims who died by suicide	Overall
Yes	6%	55%	43%	42%
No	94%	45%	57%	58%
Number of reviews	16	44	23	83

50. An additional question is “were children present when the homicide occurred?” ^{[footnote 22]} The answer was “yes” (children were present) for 50% (17 of 34) of the reviews where the victim had or previously been in an intimate relationship with the perpetrator. Children were also present in 13% of the reviews where the victim had died by suicide (two of 16 reviews). No children were present where the victim had a familial relationship with the perpetrator. For this question answers were missing for 30 of the 90 newer forms used ^{[footnote 23]}.

51. Information was given for 99 reviews on the age and sex of children living or present in the home ^{[footnote 24]}. In 17 households there was one child, in 16 two children, and in 13 households there were three or more children.

52. The sex was given for 99 children: 55% were male and 45% were female.

53. The question was asked “Were children subject to Child Protection procedures due to Domestic Abuse prior to the homicide?”. Answers were given for 76 reviews and of these 29% reported there were children subject to Child Protection procedures due to Domestic Abuse prior to the homicide. The percentage was higher (35%) where the reviews were of victims who had died by suicide.

54. The last question asked about children was “any children removed into Care of Local Authority?” Answers were given for 47 reviews and 24% of children were removed into the Care of the Local Authority. The percentage (14%) was lower for reviews where there had been familial victims.

4. Characteristics of victims

55. This chapter summarises the information on the vulnerabilities and mental health issues identified as experienced by 132 victims. The figures are separated to show differences or similarities between 32 who had a familial relationship with the perpetrator(s), 66 who had or previously had an intimate partner relationship with the perpetrator(s), and 34 who died by suicide.

56. The chapter also looks at whether the victim was a carer or had a life limiting illness. This is followed by whether the victim had been the target of an abuser before and whether they had been referred to a Multi-Agency Risk Assessment Conference (MARAC) ^{[footnote 25]}. There is a summary of aggravating factors that many victims experienced.

Vulnerabilities

57. The DHR forms indicate the vulnerabilities victims may have experienced, in terms of:

• Illicit Drug Use;

• Mental ill-health;

• Physical disability;

• Pregnancy;

• Problem alcohol use;

• Any other vulnerability ^{[footnote 26]}.

58. Vulnerabilities by type of victim is shown in

59. Table 14. Overall victims who had a familial relationship with the perpetrator(s) had fewer vulnerabilities identified (53% had none) while those who died by suicide were more likely to have an identified vulnerability (94% had at least one noted).

Table 14 Victims and number of vulnerabilities

Number of vulnerabilities	Per cent by of victims with number of vulnerabilities: Familial	Per cent by of victims with number of vulnerabilities: Intimate partner	Per cent by of victims with number of vulnerabilities: Victims who died by suicide	All victims
0	53%	32%	6%	30%
1	28%	18%	26%	23%
2	9%	27%	32%	24%
3 or more	9%	23%	35%	23%
Number of victims	32	66	34	132

Figure 4 Victims and number of vulnerabilities

60. Where vulnerabilities have been identified there is further analysis in Table 15 ^{[footnote 27]}. As an example, there were 32 victims were there had been a familial relationship with the perpetrator and vulnerabilities had been identified for 17 familial victims with the total number of vulnerabilities being 24.

61. Comparing the types of victims, the largest difference are:

• The proportion of vulnerabilities which were illicit drug use was a little lower for intimate partner victims (17% compared to 21% and 22%);

• The vulnerability of mental health was 49% of the vulnerabilities of the victims who died by suicide;

• While being lower, mental health was the most identified vulnerability for both familial and intimate partner victims;

• Physical disability formed 21% of the vulnerabilities identified for familial victims;

• Pregnancy was a vulnerability for 8% of intimate partner victims; and

• The proportion of vulnerabilities which were problem alcohol use is similar for all three types of victims, being between 29% and 27% of the overall number of vulnerabilities.

Table 15 Vulnerabilities of victims

Vulnerability	Per cent of vulnerabilities by type of victim: Familial	Per cent of vulnerabilities by type of victim: Intimate partner	Per cent of vulnerabilities by type of victim: Victims who died by suicide	Total
Illicit drug use	21%	17%	22%	19%
Mental ill-health	29%	38%	49%	41%
Physical disability	21%	10%	2%	8%
Pregnancy	0%	8%	0%	4%
Problem alcohol use	29%	27%	28%	28%
Number of vulnerabilities	24	88	65	177

62. A number of other vulnerabilities had also been identified: untreated cancer; adverse childhood experience; epilepsy; homeless and was a ‘looked after child’; the victim’s insecure immigration status; learning difficulty; self-neglect; on-going physical health problems including a diagnosis of diabetes and hearing loss; physical health was poor/ulcered legs/sleeping in a chair; misuse of prescription drugs; and childhood sexual abuse & trauma.

Mental health issues

63. DHRs are asked to indicate mental health issues of victims and these are shown in Table 16 and

64.Table 17. The types of mental health issues listed to be identified were:

• Adjustment disorder;

• Anxiety;

• Dementia or Alzheimer’s;

• Depression;

• Low mood / anxiety;

• Panic attacks;

• Psychosis;

• Post-traumatic stress disorder (PTSD);

• Self-harm;

• Suicidal thoughts;

• Suicide attempts;

• Other.

65.There are differences in the number of mental health issues for the different types of victims:

• The majority (69%) of familial victims had no mental health issues identified;

• One third (33%) of intimate partner victims had no mental health issues, but 26% have three or more; and

• All victims who died by suicide had one or more mental health issues, and 79% have three or more.

Table 16 Victims and numbers of mental health issues

Number of mental health issues	Per cent of victims with number of mental health issues: Familial	Per cent of victims with number of mental health issues: Intimate partner	Per cent of victims with number of mental health issues: Victims who died by suicide	Total
0	69%	33%	0%	33%
1	9%	29%	15%	20%
2	9%	12%	6%	10%
3 or more	13%	26%	79%	36%
Number of victims	32	66	34	132

Notes: the category “3 or more” is those victims who had 3,4,5,6,7, or 8 mental health issues identified. This aggregates the relatively small numbers for each group.

66. Table 17 shows the different mental health issues. For psychosis there is a difference of 11 percentage points between the 13% of the mental health issues for familial victims compared to intimate partner victims (1%) of those who died by suicide (2%). The largest mental health issue for victims who died by suicide was depression (followed by suicidal thoughts and attempts). Suicidal thoughts are close to one in eight (13%) of the mental health issues experienced by familial and intimate partner victims.

Table 17 Mental health issues of victims

Mental health issue	Per cent of mental health issues by type: Familial	Per cent of mental health issues by type: Intimate partner	Per cent of mental health issues by type: Victims who died by suicide	Total
Anxiety	13%	13%	9%	11%
Depression	21%	19%	19%	19%
Low mood / anxiety	21%	23%	15%	19%
Psychosis	13%	1%	2%	2%
Self-harm	13%	10%	13%	11%
Suicidal thoughts	13%	12%	17%	15%
Suicide attempts	4%	9%	17%	12%
Other	4%	13%	9%	10%
Total number of mental health issues	24	105	128	257

Carer

67. Eleven per cent of victims had been identified as being carers ^{[footnote 28]}. There is a variation between the types of victims: 22% of familial victims were carers, whilst 11% of those in an intimate partner relationship were carers. None of the victims who died by suicide were identified as carers.

68. Of the seven victims who were carers, three of the seven familial victims and one (of the seven) intimate partner victims had received a carer’s assessment.

Life limiting illness

69. The DHR forms were asked to note if any victims had a life limiting illness ^{[footnote 29]}. Answers were given ^{[footnote 30]} for 111 victims. Of these, a life limiting illness was identified for 15% of familial victims, and 13% of intimate partner victims. None of the victims who died by suicide had a life limiting illness identified.

Target of abuser before

70. Information on whether the victim had been the target of an abuser before was given for 100 victims and 52% had previously been the target of an abuser. This was lower for victims in a familial relationship: 22% (of 23), but higher for victims of an intimate partner relationship; 56% (of 50). Victims who had died by suicide, at 70% (of 27) were the most likely to have been the target of abuser before.

71. The question asking whether the victim has been the target of an abuser before is followed by asking who this abuser had been. Information is given for 50 victims. For 42 this was a previous partner or partners and five included a family member as well as an intimate partner. There are also four references to family without referring to previous partners.

Multi-Agency Risk Assessment Conference

72. Information is available for 109 ^{[footnote 31]} victims on whether they had been referred to a Multi- Agency Risk Assessment Conference (MARAC) ^{[footnote 32]}. Where victims had familial or intimate partner relationships a minority (17% of 24 or 23% of 62) had been referred. Where victims died by suicide a greater proportion (52% of 23) were referred to a MARAC (see Table 18).

Table 18 Victims referred to a Multi-Agency Risk Assessment Conference

Referred to MARAC?	Familial	Intimate partner	Victims who died by suicide	Total
Yes	17%	23%	52%	28%
No	83%	77%	48%	72%
Number of victims	24	62	23	109

73. There was an additional question as to whether the case was heard at MARAC before the homicide^{[footnote 33]}. Where the relationship was intimate partner 94% (of 16) this was the case. The information was also provided for 9 victims who died by suicide and for these the case was heard at MARAC for 7. Information was available for only four instances of a familial relationship where one of the four was heard at MARAC before the homicide.

Aggravating factors

74.Information from the DHRs includes aggravating factors experienced by victims ^{[footnote 34]}. The question asks if any of the following have been observed:

• Coercive control;

• Digital stalking;

• Financial abuse;

• Forced marriage;

• Honour-based violence;

• Immigration issues;

• Physical stalking.

75.The number of aggravating factors experienced ranges from none through to five. Table 19 shows the percentage of victims^{[footnote 35]} looking at the number of identified aggravating factors (also shown by Figure 5). There are 30% of victims where no aggravating factors were identified and similar portions where two or three factors were indicated. There are differences depending on the type of victim. There were no aggravating factors identified for 74% (of 19) victims who were in a familial relationship with the perpetrator. The reverse is true for the other types of victim: for intimate partner victims 78% (of 45) victims had experienced aggravating factors and this was higher for victim who died by suicide where aggravating factors were identified for 88% (of 26).

Table 19 Aggravating factors experienced by victims

Number of aggravating factors identified	Per cent of victims: Familial	Per cent of victims: Intimate partner	Per cent of victims: Victims who died by suicide	Total
0	74%	22%	12%	30%
1	21%	36%	31%	31%
2	5%	36%	42%	31%
3, 4 or 5	0%	6%	15%	8%
Number of victims for which data available	19	45	26	90

Notes: the information is for 90 victims from the newer forms where financial abuse was an option for an answer

Figure 5 Aggravating factors experienced by victims

76.Coercive control forms 54% of the total (107) aggravating factors reported, and financial abuse (26%) is second (see

77.Figure 6). Together these account for 80% of all identified.

78.Immigration issues were asked to be identified if an aggravating factor, and these were noted in six reviews. As a small number this has been grouped with honour-based violence (in one review) under the category “other” in

79.Figure 6^{[footnote 36]}. Concerning the question of immigration as an aggravating factor it was also asked if this was experienced by the victim, perpetrator, or both. In the six relevant reviews three were related to the perpetrator only, one to the victim and two for both.

Figure 6 Aggravating factors as proportions of total

80. As Table 26 shows, the aggravating factors experienced by the type of victim varies. Where victim(s) had familial relationships with the perpetrator(s), 26% of reviews have identified coercive control as an aggravating factor and 5% identified financial abuse. The proportions are much higher where the victim was in, or previously in, an intimate relationship with the perpetrator(s) – coercive control was present in 67% of reviews and financial abuse in 33% (see

82. Table 20). Physical stalking is present in 11% of the reviews. The proportions for victims who died by suicide are higher: 88% had experienced coercive control, 48% financial abuse, and 16% physical stalking.

Table 20 Aggravating factors – by type

Type of aggravating factors	Per cent of reviews where aggravating factor identified: Familial	Per cent of reviews where aggravating factor identified: Intimate partner	Per cent of reviews where aggravating factor identified: Victims who died by suicide	Total
Coercive control	26%	67%	88%	64%
Digital stalking	-	7%	8%	6%
Financial abuse	5%	33%	48%	31%
Forced marriage	-	-	-
Honour-based violence	-	2%	-	1%
Immigration issues	-	9%	8%	7%
Physical stalking	-	11%	15%	10%
Number of victims	19	45	26	90

5. Characteristics of perpetrators

83. This chapter summarises information on 92 perpetrators^{[footnote 37]} from the Domestic Homicide Reviews. This includes six reviews which each had two perpetrators.

84. The vulnerabilities and mental health categories considered are the same as those for victims. The chapter looks at whether the perpetrator was a carer or had a life limiting illness. Information is then summarised on whether the perpetrator had abused previous partners or family members and whether this was known to agencies. It is followed by a section on Court verdicts and sentences.

Vulnerabilities

85. Twenty three per cent of perpetrators had no vulnerabilities identified and 77% at least one. As Table 21 shows, perpetrators who had a familial relationship with the victim were more likely to have three or more^{[footnote 38]} vulnerabilities than those who had been in an intimate partnership.

Table 21 Perpetrators and numbers of vulnerabilities

Number of vulnerabilities	Per cent of perpetrator with number of vulnerabilities: Familial	Per cent of perpetrator with number of vulnerabilities: Intimate partner	Total
0	15%	26%	23%
1	19%	31%	27%
2	30%	28%	28%
3 or more	37%	15%	22%
Number of perpetrators	27	65	92

86. The vulnerabilities are shown in Table 22. Forty three percent of the vulnerabilities of perpetrators in a familial relationship was mental ill-health and this formed 35% of the vulnerabilities of intimate partnership perpetrators. Problem alcohol use was 28% of the vulnerabilities of familial perpetrators, and 37% of the vulnerabilities of those who had been in an intimate partnership.

Table 22 Vulnerabilities of perpetrators

Type of vulnerability	Per cent of vulnerabilities by type and by type of perpetrator: Familial	Per cent of vulnerabilities by type and by type of perpetrator: Intimate partner	Total
Illicit drug use	30%	24%	26%
Mental ill-health	43%	35%	38%
Physical disability	0%	4%	2%
Problem alcohol use	28%	37%	34%
Number of vulnerabilities	47	83	130

Mental health issues

87. The management information forms also gave perpetrator’s mental health issues. Sixty three per cent of perpetrators had been identified with at least one mental health issue (Table 23). The likelihood of more than one mental health issue was greater for perpetrators who had a familial relationship (70%) compared to 61% for intimate partnership perpetrators. As another perspective, 48% (of 27) perpetrators who had a familial relationship with their victim had three or more mental health issues identified compared to 26% (of 65) perpetrators who had been in an intimate partnership.

Table 23 Perpetrators and numbers of mental health issues

Number of mental health issues	Per cent by type of perpetrator: Familial	Per cent by type of perpetrator: Intimate partner	Total
0	30%	38%	36%
1	11%	23%	20%
2	11%	12%	12%
3 or more	48%	26%	33%
Number of perpetrators	27	65	92

88. Table 24 compares the mental health issues identified. Both psychosis and self-harm form higher proportions of the mental health issues for perpetrators in familial relationships than intimate partnership perpetrators. Low mood / anxiety is 10% of the 73 mental health issues for familial perpetrators compared to 21% (of 98) of those of intimate partner perpetrators.

Table 24 Mental health issues of perpetrators

Mental health issue	Per cent by type of mental health issue: Familial	Per cent by type of mental health issue: Intimate partner	Total
Anxiety	10%	6%	8%
Depression	11%	15%	13%
Low mood / anxiety	10%	21%	16%
Psychosis	16%	4%	9%
Self-harm	16%	6%	11%
Suicidal thoughts	16%	16%	16%
Suicide attempts	8%	6%	7%
Other	12%	24%	19%
Total number of mental health issues	73	98	171

Carer

89. Information on whether the perpetrator(s) were carers was given for 88, and 16% of these were identified as carers. The percentages were similar for both familial and intimate partner perpetrators. Of the 14 who were carers four had received a carer’s assessment.

Life limiting illness

90. Information on whether or not a perpetrator had a life limiting illness was given for 75 perpetrators. No familial perpetrator had such an illness. Three (6%) of perpetrators who had or had previously been in an intimate relationship were recorded with a life limiting illness.

Has the perpetrator abused previous partner/s or family members?

91. Information was given on 67 of perpetrators on whether they had previously abused previous partners or family members. There are differences between those with a familial relationship (52% had previously abused) and those who had been in an intimate relationship (74%) (shown in Table 25).

Table 25 Abuse of previous partner/s or family members

Abuse of previous partner/s or family members	Familial	Intimate partner	Total
Yes	52%	74%	67%
No	48%	26%	33%
Number of perpetrators	21	46	67

Was the perpetrator known to agencies as an abuser?

92. The management information form asked whether the perpetrator was known to agencies as an abuser and, if the answer was yes, to state which agencies.

93. Table 26 shows that a higher percentage of offenders were known to agencies as an abuser than not known. There are differences between the two categories of perpetrators. While 36% of 25 familial perpetrators were known to agencies as an abuser there was a larger proportion (64% of 61) intimate partner perpetrators known.

Table 26 Was the perpetrator known to agencies as an abuser?

Abuse of previous partner/s or family members	Familial	Intimate partner	Total
Yes	36%	64%	56%
No	64%	36%	44%
Number of perpetrators	25	61	86

Notes: there are two familial and four intimate partner perpetrators where the answers given as not known.

94. The number of agencies where the perpetrator was known is shown in Table 27. Fifty nine per cent (of 63) perpetrators who had an intimate partnership relationship with the victim had been known to at least one agency, compared to 33% (of 27) familial perpetrators.

Table 27 Number of agencies to whom perpetrator was known

Number of agencies where perpetrator was known	Per cent by type of perpetrator: Familial	Per cent by type of perpetrator: Intimate partner	Total
0	67%	41%	49%
1	15%	22%	20%
2	7%	21%	17%
3	0%	8%	6%
4 or more	11%	8%	9%
Number of perpetrators	27	63	90

Notes: there are two intimate partner perpetrators where the answers were given as Not Known.

95. The agencies named in the management information forms have been placed into the following categories: Health, Police, Probation, Adult Social Care, Children’s Social Services, or other. These are broad descriptions - those in health include mental health, hospital, ambulance service and GP. The information does not show the number of agencies of the same type, as one example a perpetrator had been investigated for domestic abuse incidents by five different Police areas, in addition, to being investigated by another Police Force for an alleged sexual assault on a female. This would only be shown as known to the Police.

96. The agencies who knew the perpetrator had a prior history as an abuser are shown in Table 28. The Police account for 44% of the total (96) occasions when agencies knew of the perpetrator. The second highest proportion overall is Children’s Social Services (18%). Probation, and Health services and the range of “other” agencies^{[footnote 39]} have similar proportions (13%, 11% and 10%). The largest difference between familial and intimate partner perpetrators are Health services, which form 19% of the agencies with knowledge of familial perpetrators, compared to 9% of those who had knowledge of an intimate partner perpetrator.

Table 28 Agencies to whom perpetrator was known

Agencies where perpetrator was known	Per cent by type of victim: Familial	Per cent by type of victim: Intimate partner	Total
Health	19%	9%	11%
Police	38%	45%	44%
Probation	14%	12%	13%
Social Care – adult	0%	1%	1%
Social Services – Children’s	14%	19%	18%
Other	10%	11%	10%
Total number of agencies with knowledge	21	75	96

Was the perpetrator being managed or supervised?

97. The information forms were asked to record whether the perpetrator was being managed or supervised by any of five different types of service (the services are named in

98. Table 30). Fifty nine per cent of familial perpetrators were managed or supervised by a service, compared to 35% of intimate partner perpetrators.

Table 29 Number of perpetrators managed or supervised by or attending a service

Number perpetrators managed or supervised by, or attending a service	Per cent by type of perpetrator: Familial	Per cent by type of perpetrator: Intimate partner	Total
0	41%	65%	58%
1	37%	23%	27%
2	15%	6%	9%
3 or more	7%	6%	7%
Number of perpetrators	27	65	92

99. Table 30 shows the proportions of perpetrators attending or being supervised by a service. Fifty four per cent of the total attendances or supervision for familial perpetrators was Mental Health Services. This was 31% for intimate partner perpetrators. The proportion of intimate partner perpetrators being managed by National Probation was 36% compared to 25% of perpetrators in a familial relationship.

Table 30 Services perpetrator managed by, supervised or attending

Perpetrator managed or supervised by, or attending a service	Per cent by type of perpetrator: Familial	Per cent by type of perpetrator: Intimate partner	Total
Attend Perpetrator Programme	4%	6%	5%
Drug and Alcohol	17%	14%	15%
Multi-agency public protection arrangements (MAPPA)	0%	14%	8%
Mental Health	54%	31%	40%
National Probation	25%	36%	32%
Total services supervised by	24	36	60

Court verdict and sentence

100. The management information forms give information on the Court Verdicts and these are shown in Table 31. For both types of perpetrator the most common verdict is murder. The proportion is higher (65%) for intimate partner perpetrators than for those who are familial (41%). Alongside this, the proportion of manslaughter verdicts are higher for familial perpetrators (37% of verdicts) compared to intimate partners.

101. For 15% of familial perpetrators the verdicts are diminished responsibility.

Table 31 Court verdicts

Court Verdict	Familial	Intimate partner	Total
Diminished responsibility	15%	5%	8%
Manslaughter	37%	23%	27%
Murder	41%	65%	57%
Unfit to Plead	4%	0%	1%
Other	4%	8%	7%
Number of perpetrators	27	62	89

102. In addition, information forms are asked to include the length of sentences. For the 40 life sentences minimum tariffs ranged from 11 years to 35 years, with the average being 20 years.

Homicide followed by perpetrator suicide

103. Ten of the perpetrators (11%) died by suicide after the homicide. Nine were or had been intimate partners of the victims and one had been in a familial relationship. Nine were male.

6. Themes from lessons to be learned from DHRs for familial abuse victims

Introduction

104. This chapter looks at themes from lessons learned from a sample of 17 DHRs^{[footnote 40]} where the victims were familial. From the sample 101 lessons learned were extracted. These are examined firstly to identify the agencies and types of individual^{[footnote 41]} where learning was identified, then the lessons learned were reviewed to identify and give examples of themes (Appendix 2. Selection of lessons to be learned gives more detail on their selection).

Frequency of agencies and individuals in DHRs

105. The term agency is used for bodies or services identified in the DHRs. This covers a wide range of organisations (e.g. the Police, a school) including some who commission activity (e.g. Clinical Commissioning Groups, now changed to Integrated Care Boards^{[footnote 42]}), and partnerships (e.g. a Crime and Drugs Partnership). Another category is “professionals”, used in 15% of the lessons learned e.g. “professional curiosity to be used when a client with mental health issues mentions negative comments about a family member”.

106. Lessons to be learned can identify more than one agency e.g. “Public Health, CCG [Clinical Commissioning Group] and Children’s Services Access & Inclusion Team in xxx should ensure that GPs and education providers are aware of self-help resources that empower children, young people and families experiencing ADHD [Attention deficit hyperactivity disorder]”.

107. The approach gave 136 allocations of lessons learned to agencies or types of individuals.

108. Figure 7 shows the percentage of agencies or types of individuals. The largest proportion (48%) are those which are in the category “health agencies”. The next two highest proportions (15% each) included both those which attributed learning to many agencies and also those related to professionals e.g. “There is still a lack of awareness across agencies of the necessity and benefits of a carers assessment, this would have assisted professionals to understand the needs”. This is linked to professionals as well as “many agencies”.

109. Probation has 5% of the lessons learned e.g. “as outlined earlier, the NPS [National Probation Service] identified some issues about the accuracy and quality of the report while setting these in the important context of the challenges faced by probation officers in producing same day reports within the limited time allocated”.

110. There are a number of other agencies which have been linked to small numbers of lessons learned and these are in the category “other”. They are: Children’s Social Care, Court, Domestic Abuse Partnership or Service, Education, Housing, Local Authority^{[footnote 43]}, MARAC, Police and Prison.

111. Figure 8 also shows the type of health agency with lessons to be learned. The two types with the most lessons to be learned are Mental Health Services (29%) and GPs (25%). NHS Trusts are 12% of the total and Clinical Commissioning Groups 11%. Hospitals form 9%. Another 12% of the learning lessons are for other health related organisations, for example ambulance services and student health services.

Figure 7 Agencies or people in lessons to be learned from familial abuse DHRs

Figure 8 Health agencies in lessons to be learned from familial abuse DHRs

Themes from lessons to be learned

112. Figure 9 and Table 32 show eight themes within the lessons to be learned for familial victims. The following sections give examples.

Table 32 Themes in lessons to be learned from familial abuse DHRs

Lessons learned	Percentage
Assess	16%
Risk	16%
Information	13%
Aware	11%
Family	10%
Support	9%
Care	8%
Children	8%
Total occurrences	140

Figure 9 Themes in lessons to be learned from familial abuse DHRs

Lessons to be learned relating to assess / assessments:

113. Twenty four lessons learned relate to assess or assessments.

114. Six were on mental health assessments.

Mental health assessments (including for ADHD [attention deficit hyperactivity disorder]) should always be informed by a complete health history and by any previous assessments

115. Attention was also given for the need to communicate assessments with family members.

That there is timely feedback to family members who make referrals, that where appropriate their views are sought and form part of the assessment and decision-making process

Lessons to be learned relating to risk:

116. Twenty four lessons learned relating to risk were identified.

117. Seven referred to elements of risk that had not been recognised.

There is a need to equip case managers with the knowledge and practice guidance to inform their practice when exploring the risk posed between family members.

118. While a risk management assessment might have been started, three lessons gave examples where they had not been completed.

It is noted that the probation officer did not complete the correct layer of OASys [Offender Assessment System] Risk Assessment as directed in the xxx CRC [Community Rehabilitation Company] ‘Every Case Essentials’ practice guidance.

119. Three of the lessons were that risk assessments had not been carried out.

The GP practice did not have policies in place to support enquiries about domestic abuse or offer any risk assessment tools

Lessons to be learned relating to information:

120. Information was referred to in 20 lessons.

121. The largest number concerned information sharing, such as “accuracy of information sharing to inform multi-agency risk assessments and care planning.” The example below shows there can be more technical issues between systems and suggests ways round this.

GP2GP patient record transfers remain a national problem with the two predominant IT systems used by GPs still failing to transfer records seamlessly. Health professionals should consider direct conversations to achieve a verbal ‘handover’ of key concerns relating to vulnerable patients

122. The sharing of data is not only concerning transfer from one agency to another (or part of an agency to another part of the same agency). There is the need to ensure understanding goes with an exchange.

Where safeguarding concerns are identified they should be referred appropriately by the organisation identifying them and not passed to another organisation to be referred as this may lead to misinterpretation or the referral not being made.

123. Alongside this there are three references to information not being shared, or detail missing in shared information.

Information about risk which had been logged by police was not conveyed to mental health services in sufficient detail.

“The administration of depot medication was not recorded in the electronic clinical records.

Depot medication was missed, and there was no robust system for ensuring these were administered at the correct times or following up missed injections. This resulted in Mr X being unmedicated from August 2018 to January 2019.

Mr X was not stabilised on depot medication before discharge ….”

Lessons to be learned relating to aware / awareness:

124. There are 17 lessons learned which relate to aware or awareness.

125. These included the importance of awareness of the range of factors which needed to be assessed for safety/risk planning.

“The significant lessons highlighted by xxx were:

• the perpetrator’s history of domestic incidents was not taken in to account;
• the context of the coercive control was not recognised;
• the finality of the situation with the perpetrator having nothing to lose.

To raise awareness of these factors, xxx is requiring all its practitioners to receive training xxx to enable practitioners to understand the often hidden nature of coercive control, and the escalation of risk when it is challenged, so that this is taken into account into safety planning”

126. For the context of family domestic homicides, two DHRs refer to the rights of relatives in considering a mental health assessment of a family member.

Lessons to be learned relating to family:

127. There are 16 lessons learned which have been linked to a specific reference to family (all DHRs in this chapter consider familial relationships and frequently this is in or implied by other lessons learned). The key message from the family lessons learned is about the need for families to be included.

Professionals working with families where children are diagnosed with ADHD [attention deficit hyperactivity disorder] should be aware of the need to work with the whole family and identify their strengths as well as areas that require attention and support. A ‘whole family’ approach will always be helpful

Family education and interventions; as in NICE [National Institute for Health and Care Excellence] guidance ‘Psychosis and schizophrenia in adults: prevention and management’ (2014); were not provided.

The family was not involved in care planning for Mr X, despite their requests to be involved and informed.

There were no carer’s assessments requested or arranged for his parents, despite them specifically requesting this.

Risk management considerations were not applied to his family.

128. In addition to the need to take account of family’s views and needs, two lessons relate to connecting family information within systems.

Lessons to be learned relating to support:

129. The overall message from 14 lessons learned is that more support is needed. There are three which simply refer to no support being provided. There are four examples that while some support has been provided, other or different provision would be more beneficial. The first example below relates to the timing of support. Equally, networks are seen as important ways of helping to provide support, these can be described generically as “support networks” or, as also given below, there can be more specific examples.

Where a care and support package is required immediately but there are moving and handling concerns and a specialist assessment is indicated, consideration needs to be given as to how to provide support in the meantime rather than waiting for the outcome of that additional assessment

The role of faith leaders and the influence of the church in African and Caribbean communities … was a recurring theme in the review. The review highlighted the importance of working with faith leaders to ensure that they have access to training and support in matters related to mental health and domestic abuse, and that they can work effectively with a range of agencies to support members of their congregations and the wider community.

Lessons to be learned relating to care:

130. Thirteen lessons learned relating to care were identified.

131. Three concern care planning and three to the co-ordination of care. These two are brought together in one DHR.

“The CPA [child to parent abuse] Policy was not followed with respect to care coordinator provision, care planning and reviews, and Trust systems did not identify or address these deviations from expected Policy within CMHART.

Assistant Practitioners were assigned to take§ on the role of care coordinators within the original Trust. We have not made a recommendation that this should stop, because the new Trust has confirmed that this is no longer accepted practice.

There was no care coordinator cover provided for a six-month period in 2018.

132. Two DHRs refer to care / care receiver relationships.

Organisations should be more aware of domestic abuse in the form of coercive control and how this may present in a carer / care receiver relationship. This should be considered in assessments and contacts

Lessons to be learned relating to children:

133. There are 12 lessons learned which include a variety of references to children within seven DHRs. One reflects the importance of including this in a policy.

There is no implementation plan for the current xxx domestic abuse strategy. Within the strategy there is no mention of risk to parents of adult children

134. Another example is given below.

Professionals should be aware that CPA [child to parent abuse] can start early in a child’s life and become an entrenched problem

7. Themes from lessons to be learned from DHRs involving intimate partner victims

Introduction

135. This chapter examines lessons learned from a sample of 34 DHRs ^{[footnote 44]} where the victims have had intimate partnership relationships with the perpetrators. From the sample DHRs 244 lessons to be learned were extracted. The chapter shows the types of agency or individuals ^{[footnote 45]} where the need for learning has been identified. After this the lessons learned have been reviewed to identify and give examples of the main lessons.

Frequency of agencies and individuals in DHRs

136. In the sample of 34 DHRs 244 lessons learned were identified and 387 agencies were associated with these. They are shown in Figure 10.

137. Health agencies accounted for 19% of the 387 agencies linked to the lessons learned. Police have the next largest proportion: 16% of the total. Following this, 13% are related to “multi-agencies” e.g. “There was an opportunity on this case for agencies to have collectively considered the circumstances surrounding the abuse and commenced a multi-agency approach in terms of risk management and accessibility of services for the victim.” A slightly smaller proportion of lessons learned (11%) were related to professionals e.g. “Professionals must remember that not everyone understands what constitutes domestic abuse behaviour.”

138. Other types of organisations or partnerships identified with over 5% of the total were Children’s Social Care and MARAC.

139. There were a wide range of other organisations identified, each with relatively low numbers of lessons learned. Together they were 28% of the total and included: Adult Social Care, Community Safety Partnerships, Courts, drugs and alcohol services, domestic abuse partnerships or services, Education, Housing, local authorities, national organisations and Probation.

140. Health agencies were identified within 19% of the lessons learned. These are broken down in more detail, as shown in Figure 10.

Figure 10 Agencies or people in lessons to be learned from intimate partner DHRs

Figure 11 Health agencies in lessons to be learned from intimate partner DHRs

Themes from lessons to be learned

141. The lessons learned identified within intimate partner victim DHRs are shown in Table 33 and Figure 12. Examples follow in the rest of this section.

Table 33 Themes in lessons to be learned involving intimate partner victims

Lessons learned	Percentage
Risk	25%
Support	19%
Information	16%
Safeguarding	12%
Contact	10%
Family	10%
Children	9%
Total occurrences	390

Figure 12 Themes in lessons to be learned involving intimate partner victims

Lessons to be learned relating to risk:

142. Lessons learned which refer to risk form the largest number relating to intimate partner victim DHRs, with 97 being in this category (25% of the total).

143. Within risk, the most common references (23% of the total references to risk) are to risk changing over time.

A lack of response to concerns relating to risk of harm, specifically in relation to domestic abuse and that information and assessments were not reviewed contextually. Assessments were not updated after significant events. Record-keeping was not, at times, contemporaneous. Multi-agency working was not always co-ordinated and holistic in nature. Unconscious bias and overfamiliarity led to a lack of focus.

144. Following risk changing over time, 12 referred lessons learned related to risk assessment needed and action required following risk assessment.

This resulted in domestic abuse not being recognised and no further risk assessment being undertaken by children’s social care. A was not referred to a specialist domestic abuse service. Furthermore, children’s social care arranged for the perpetrator’s contact with the child to be supervised by A. Children’s social care response demonstrated that they did not perceive this case as one of domestic abuse and this was directly linked to the perpetrator’s call to the police.

145. A similar number (11) of references were made on the need for risk assessments and information for these to be shared across all the agencies.

Where information is missing from agency referrals, particularly for cases where there is evidence of complex needs and identified vulnerabilities, it creates a situation that the person or agency receiving that referral is not in possession of all the known facts and this can reflect on the level of service that they provide. Referrals for clients who have complex needs and identified vulnerabilities, should contain all relevant information, including vulnerabilities and areas of risk.

Lessons to be learned relating to support:

146. Seventy two lessons learned (19% of the 390 total) were identified which related to support. Thirty eight per cent (27 of the 72) were focused on support for the victim. These covered a wide range of needs. The most frequent were:

• support for the victim regarding use of drugs or alcohol (4);

• mental health needs (3);

• where support was needed by more than one agency (3), and

• where coercive control from the perpetrator made it more difficult for the victim to receive support (3).

147. Eleven of the lessons learned referred to support and the victim’s family.

The case has identified further learning around the consideration of sharing information with family members and/or named individuals, to allow those named persons to then provide advice and support to the person at risk. Information sharing with family members and/or named individuals can provide an opportunity for support and advice to be given to victims, in managing and understanding the risk.

148. A number of lessons learned relating to support (11) make references to professionals being alert to the available support for families.

Professionals need to be more alert to the fact that an underlying cause of depression, anxiety or other mental health conditions may be domestic abuse, and that interventions aimed at targeting domestic abuse are less likely to be effective if mental health needs are ignored. In addition, professionals need to be more aware of what domestic abuse support services are available to ensure victims are appropriately referred to specialist support.

149. Six references to ‘need for support’ have been grouped together in what is described as “cultural support”. A general example is given below. Another example refers to the need for interpreter provision to help statutory agencies better respond to need by referring survivors or perpetrators of domestic abuse into services or organisations when English is not their first language.

The Domestic and Sexual Violence and Abuse Strategic Framework and action plan - aims to ensure the reach and accessibility of both statutory and specialist support services for domestic violence is such that people in every community are clear where and how to seek help for themselves and others in a way which meets their needs.

Lessons to be learned relating to information:

150. Sixty two lessons to be learned (16% of the total) have been identified with regard to information. These have been grouped into four categories: sharing information (24), making information available (13), the use of information (13), and recording information (10)^{[footnote 46]}. A number of lessons fall into more than one category.

151. The lessons learned around sharing of information are in the largest category (41% of the 62). The need to share information is not just between different organisations but also includes similar organisations (e.g. information from one school to another school) and within organisations.

The learning from the GP Practice perspective is that they do not always have a full suite of information from outside health agencies. This can make consultations with patients challenging when a clear picture of other external consultations is not readily available

That schools have been historically ‘slow’ in sending files and information to the school to which a pupil has moved resulting in them having only the limited information included on the admission form

Equally significant, work was undertaken to bring about cultural change within the organisation, aimed at ensuring frontline operational officers and staff were able to recognise vulnerability in its various forms, including where it may result in domestic abuse and CSE [Child sexual exploitation] . Increased emphasis was placed on effective recording, information-sharing and risk assessment, and all officers were required to undergo mandatory training

152. The majority of the examples of lessons learned on making information available concern making information available to victims.

There were opportunities in this case for information to be shared …. This did not occur. Whilst processes have been implemented to address this area of learning, the case has identified further learning around the consideration of sharing information….

153. The use of all information lessons refers to where information is available but has not been accessed or examined to inform further actions.

Identified the need for greater professional curiosity and ensuring that full exploration of all information provided; especially the level of detail which could raise potential safeguarding concerns

154. Many of the lessons on recording information also cover ensuring that the information is used and shared. One example regarding the need to record information is given below.

There are examples of risk assessment taking place and in the main these were appropriate and of a good standard. However, the broader view of XXX and her risks did not permeate between agencies. This is despite there being evidence of information sharing and joint working. There were occasions where the use of DASH was not undertaken and these should be regarded as missed opportunities to identify and respond to risks

Lessons to be learned relating to safeguarding:

155. Forty six lessons (12% of the 390) which refer to safeguarding have been identified. Examples of those in the largest categories are given below.

156. The largest number of lessons learned (10 of the 46) were on making changes to safeguarding as a response to the DHR. The nature of the response included:

• the introduction of a safeguarding plan template;

• safeguarding which has received additional staff and established more effective ways of working;

• training around safeguarding has developed in response to the revised intercollegiate document and there is an ongoing rolling programme to keep staff refreshed;

• investments to improving policy, procedures, response, training and raising safeguarding standards;

• embedding domestic abuse awareness in both adult and child safeguarding training available to the Clinical Commissioning Group and Primary Care staff including; the importance of routine screening for domestic abuse, improved recognition, and response to domestic abuse including appropriate onward referral to specialist services; and

• staff from the Revenues & Benefits Service received training to identify vulnerable customers and to take appropriate action (e.g., referring potential safeguarding cases, and in respect of debt collection to pause recovery and refer customers to support services).

157. A similar number (10) of lessons were on safeguarding action not taken.

the multi-agency response did not follow Safeguarding Adults policy or procedure:

• Safeguarding meetings took place but were not properly minuted and communicated to multi-agency partners in line with local policy and procedure.

• Actions were agreed but without any clarity of who was responsible or when the actions would be completed by.

• Agreed actions were not implemented. This included an action that XXX’s mental capacity (in relation to reporting alleged abuse and seeking protection from further abuse) should be assessed. This assessment did not take place.

• Despite clear evidence that abuse had taken place, this was never followed up, or formally recorded as substantiated / unsubstantiated or inconclusive

158. Six of the examples focus on the need for safeguarding to be shared between agencies.

Staff in A&E must raise safeguarding concerns to children’s social care in accordance with their safeguarding policy and procedure, when deteriorating mental health and alcohol misuse is disclosed

159. The need to share information between agencies is also (in five lessons learned) related to Multi-Agency Safeguarding Hubs (MASH).

The consequence of his call was that, as there was a young child in the household, a report was sent to the multi-agency safeguarding hub (MASH) and copies were received by children’s social care, the GP and the health visiting team. The report described him as the aggrieved and X as the suspect. This notification, together with the notification of X’s call to the police, gave professionals the impression that X and the perpetrator were making ‘tit for tat’ allegations during an acrimonious separation

Lessons to be learned relating to contact:

160. In the sample of DHRs where the relationship between the victim and perpetrator was (or had been) an intimate partnership 40 of the lessons learned concerned contact with the victim.

161. An important part of this (18 of the 40 lessons) is the contact between the victim and services.

XXX’s contact with services is characterised by an initial desire to engage, but she experienced difficulties and concerns about sustaining engagement with some of those services. This meant it was sometimes difficult for those services to sustain a meaningful programme of support.

162. Another aspect of contact with the victim is understanding and responding to cases where contact is not being continued.

Whilst there can be legitimate reasons for people not attending appointments or engaging with services, professionals need to incorporate professional curiosity when reviewing non-contact, and take cognisance of the circumstances of individual cases, particularly when an individual is being ‘cared’ for or has additional needs. Professionals must ensure that they have robust policies and processes to respond to incidents of non-engagement or contact.

163. There were nine lessons learned on the need for services to contact each other. One DHR gives an example of a single point of contact as a way of addressing this.

XXX was not known to Adult Social Care [ASC] and ASC were unaware of the relationship between her and the Perpetrator during their interactions with him. It appears that from 2019, XXX had at least 6 VPA’s [Vulnerable Person Assessment] activated, yet none of these appear to have been received by Adult Social Care. ASC noted a comment from a meeting of the MARAC suggesting that the IDVA [Independent Domestic Violence Adviser] service was waiting for a joint visit to XXX with Adult Social Care, but that she did not hear from them. As ASC had no information on XXX, or received any VPAs, this contact was obviously not made and there was no follow up from the IDVA. Improved sharing of information between the Police, CMHT [Community Mental Health Team], IDVA and ASC may have been beneficial in this case.

Lessons to be learned relating to family:

164. Thirty eight lessons learned have been identified which relate to family.

165. Four use the term “family dynamics” as the way of expressing the learning needed.

Throughout the brief contact that XXX and XXX had with agencies, there was little professional curiosity to tease out the dynamics of the family unit or understand the complex relationship between XXX and XXX. Staff need to be more inquisitive, adopting a ‘believe but verify’ approach to information received and make further enquiries as and when necessary. This will help build a more accurate picture of individuals, family dynamics and aid risk assessment.

166. Three lessons learned relate to the need to consider the wider family.

Domestic abuse relationships are very complex, as is the impact on everyone in the family who experiences or witnesses it. Clear reflective supervision is needed to support professionals in this environment to remain curious and to question their own assumptions… The historical information regarding the family is clearly recorded in the last two referrals made in the scope of the review. It is acknowledged and discussed within the referrals; however, it does not appear to have changed the way that the referrals were managed.

167. There are lessons learned which have each occurred twice. These relate to awareness that financial issues can affect the family, the dangers of using family members as interpreters, and action needed when families move.

gambling and alcohol abuse led to financial problems for the family. Agencies need to be alert to financial hardship as being an indicator of economic and domestic abuse.

“Not being able to match a patient’s first or preferred language can impact on patient experience and health outcomes, the frequency of missed appointments and the effectiveness of consultations. It may have serious implications such as misdiagnosis and treatment, ineffective interventions and, in extreme circumstances, preventable deaths. The use of an inadequately trained (or no) interpreter poses risks for both the patient and health care provider. The error rate of untrained interpreters (including family and friends) may make their use higher risk than having no interpreter at all.”

Where families move areas, it is imperative that relevant information moves with them. This is particularly important where the family are not receiving statutory support. The information of the child protection concerns did not follow the children with Education and Health. In cases such as this the only opportunity to be able to provide the victim and wider family with support will be by ensuring that universal services effectively share what limited information there is

168. There are also two lessons which refer to ways in which families can understand domestic abuse. One explains “It is understandable that families want to help couples experiencing difficulties but there are real dangers in mediating where domestic abuse is involved” and a second learning point observed “The review identified that coercion and control was not known as a form of domestic abuse by the family.”

Lessons to be learned relating to child / children:

169. Thirty five lessons were identified which referred to children.

170. The lesson learned with the largest number of references (12) was safeguarding.

During the period prior to the removal of her children, XXX’s needs as a victim of domestic violence and abuse may have been obscured by the safeguarding needs of her children. It was right that priority was afforded to safeguarding XXX’s children because, by reason of their age and maturity, they were unable to protect themselves from harm. However, the need to safeguard XXX from domestic violence and abuse appears to have been overlooked at times. It is acknowledged that this is a challenging area of work for practitioners who work diligently to support parents who are victims of domestic violence and abuse to keep their children, and themselves, safe.

171. There were two additional lessons where four examples were found. These included meeting children’s needs.

There may be a barrier preventing victims of domestic abuse accessing support where they are concerned for the implications such disclosure would have on their perceived ability to meet their children’s needs.

8. Themes from lessons to be learned from DHRs involving victims who died by suicide

Introduction

172. This chapter looks at lessons learned from a sample of 18 DHRs^{[footnote 47]} where the victims died by suicide. From the sample 173 lessons have been extracted. These are examined firstly to identify the agencies and types of individuals^{[footnote 48]} for whom learning was identified. Following this, themes have been identified through the most common words used and examples given for these. More detail on how the lessons to be learned have been identified is in Appendix 2. Selection of lessons to be learned.

Frequency of agencies and individuals in DHRs

173. The 173 lessons learned contain 361 references to agency (or the term professional) (see Figure 13). The largest proportion are health agencies (30%). The next largest proportions are where many agencies have been identified and also where the term ‘professional’ has been used: 10% each. An example of many agencies is “Information sharing between agencies and the need for the whole person/family to be visible” and for professionals: “knowledge of the link between domestic abuse and suicide will enable professionals to formulate appropriate risk assessments and risk management plans.” There are a number of other agencies with close proportions: Police 9%, and domestic abuse services or partnerships with eight per cent. Both Children’s Social Care and MARAC each form 6% of agencies.

174. In looking at the references to health agencies in more detail (Figure 14) GPs have the highest proportion – being 30 or 27% of the 110 references to health agencies. At 20% NHS Trusts are the second largest group, with Mental Health Services having 18% and psychologists or psychotherapists have 7%.

Figure 13 Agencies or people in lessons to be learned where victims died by suicide

Figure 14 Health agencies in lessons to be learned where victims died by suicide

Themes from lessons to be learned

175. From the sample of DHRs for victims who died by suicide, 340 lessons learned were identified ^{[footnote 49]}. Themes from these were identified through the most commonly used words. Figure 15 and

176. Table 34 show these themes. After this the following sections illustrate these with examples.

Table 34 Themes in lessons to be learned for victims who died by suicide

Learning lessons	Percentage
Support	26%
Information	22%
Risk	22%
Training	14%
Staff	9%
Review	7%
Total occurrences	210

Figure 15 Themes in lessons to be learned for victims who died by suicide

Lessons to be learned relating to support:

177. Fifty five lessons learned (26% of the total) were found which related to support.

178. Within these, the largest proportion (38%) related to the need for support not being identified. As part of this, the following example shows the complexity of elements encompassed within domestic abuse for which support can be required.

The need for trauma-informed services and assertive outreach for those with multiple disadvantages, with a pathway of support in place for repeat victims of multiple perpetrators

179. The second largest number of lessons (20%), referred to the need for support related to families and children. An example of this is given below.

Impact on Children living with Domestic Abuse. It is important that professionals working with families understand the impact on children of living with domestic abuse and have the tools to identify behaviours, listen and interpret the voice of the child to enable the whole family’s needs to be better supported

180. Fifteen per cent of the lessons give examples of support not being provided by all agencies. Those which relate to information are mentioned (later) in that section. The work between agencies can be seen to be wider than only sharing information.

XXX’s GP referred her to the Health Visiting service requesting a support visit as a result of the acrimonious divorce; successful contact was not made - an attempt was made via phone, no message left because it was not a personalised answerphone and no further follow up was documented. Panel concluded that this may have been a possible missed opportunity to provide support and gather further information about their relationship and likely impact on the safety and wellbeing of XXX.

181. Another issue raised in relation to support was concerning the victim accessing support, many examples refer to a wide range of agencies.

xxx did not accept services from Children’s Social Care intended to support her, and referrals to Adult Social Care did not result in support being offered. People who feel threatened or alienated are unlikely to engage in services. A separate offer of support outside child protection arrangements could be more helpful

182. Five lessons learned referred to resourcing and support. Many related to the number of staff, but one saw the position as wider.

XX have secured funding and support for a new post for the next 12 months to employ a Recovery Coordinator specialising in domestic abuse. The new post will provide the capacity to assertively engage and manage the complexities and risk of victims that have significant substance misuse issues. That being to improve victim access and engagement in safety planning, risk management, and linking them with specialist support such as IDVA by using an intensive engagement approach not previously possible

183. Other lessons included the need for support to be given sooner.

Lessons to be learned relating to information:

184. Forty seven lessons were identified which related to information.

185. The need to share information formed the majority (53%) of these. One example, given below, shows the importance of information sharing between all organisations.

Information sharing: It is vital that agencies share information where there are issues relating to safeguarding including mental health and domestic abuse. There were occasions picked up in the review where information sharing did not happen

Currently the onus is on the professional to review the historic records systems including scanned paper records to access historic information and to information share appropriately across teams and services, this is unrealistic as it is time consuming and not all staff have access to all systems.

186. That more information was needed accounted for 23% of the lessons learned.

The failure of some representatives to research available information relating to XX and XX prior to the MARAC meeting resulted in the MARAC being unable to agree what support services were appropriate for XX. This reinforced the need for the MARAC Operating Protocol and the implementation of the proposed MARAC Improvement Plan.

187. Other references to information give examples relating to when information was provided but not used and when a summary of the detail was needed to help its use. Other examples include when information was required to be provided to the victim, and the timing of when the information was used.

Lessons to be learned relating to risk:

188. The lessons learned relating to victims who died by suicide included 46 on risk. The largest group of lessons (14) were around the need to have correct risk assessments made.

Cases where there is a high volume of repeat domestic abuse incidents, combined with other risk factors, should be recognised as high risk and generate a MARAC referral

189. That risk assessments were not carried out were in eight of the lessons learned.

given XX had made several disclosures of domestic abuse during this contact with this organisation a DASH should have been considered to assess the level of risk and offer support

190. At a similar frequency (eight lessons) was the importance of understanding the risk assessments.

Referral to specialist domestic abuse services is an essential part of providing support and developing a more informed understanding of risk; their practitioners are also more likely to be trained in appropriate techniques such as motivational interviewing and opt-in language to enhance the opportunity for engaging with victims of domestic abuse and eliciting information and providing advocacy;

191. A number of other lessons are on the importance of organisations working together.

XX did not share their concerns of the risk regarding potential development of a relationship with ex-partner, who has previously been physically abusive’, with any other agencies, nor did they carry out any form of risk assessment.

192. Other lessons refer to different aspects of “risk”, including the need for a plan to mitigate risk (four lessons) and two referred to the need to include children in the management of risk.

Lessons to be learned relating to training:

193. Twenty nine lessons found refer to training. A wide range of topics for training were identified. Two were each about:

• Professional curiosity;

• Referral to a new service;

• Training on suicide prevention; and

• Training to work in a trauma informed way.

194. Other training requirements identified were on:

• Behaviours of domestic abuse perpetrators;

• Common understanding of domestic abuse;

• Documentation regarding consent;

• Effects of domestic abuse and coercive control on adult victims;

• Legislation and practice relating to grooming, stalking and coercive control;

• Referral routes;

• Spotting the signs and proactive enquiry;

• Suicide prevention; and

• To support private counsellors.

195. Four lessons identified training in agencies or sectors: Adult Community Mental Health, Children’s Social Services, trainee GP’s, Health Visitors, and school. Regarding how training can be delivered, one DHR noted the co-operation taking place between agencies.

Lessons to be learned relating to staff:

196. Eighteen references to staff were identified, four of these to the number of staff.

In March 2020 only one of these post holder(s) was in post, whilst the other two were vacancies actively being advertised and suitable candidates sought^{[footnote 50]}

Lessons to be learned relating to review:

197. In fifteen lessons review was identified.

198. Examples were also given on the need to review the understanding of the circumstances. These were linked to groups as well as individuals.

“Whether professionals were aware of unconscious bias when reviewing the support XX, XX and the children received / were offered to ensure scenarios were interpreted from a neutral standpoint.”

199. Other references to review related to process, including the need to review records. One lesson concerned a review of outstanding risks in MARAC.

Recent changes have resulted in the MARAC minutes now being recorded, and actions agreed at previous meetings which have been completed are shared by agencies during the meeting. A review of outstanding risks takes place, and a follow-on action plan is recorded including which agency is accountable for the action’s completion. A MARAC action tracker has been introduced to record the progress of assigned MARAC actions to ensure there is evidence of outcomes being achieved via the MARAC process …

9. Comparing themes from lessons to be learned across the different types of victims

200. This chapter examines the lessons learned from chapters 6, 7 and 8 to compare themes for each type of victim.

Frequency of agencies and individuals in DHRs

201. The proportion of different groups of agencies with lessons learned are shown in both Table 35 and Figure 16.

• Health agencies are the largest proportion of agencies for DHRs with familial abuse victims or victims who died by suicide (48% and 30%). For intimate partners the proportion of lessons learned for health agencies (19%) is smaller. And this is close to the 16% in which the Police are identified;

• The Police feature in 16% of the lessons learned involving intimate partner victims. For victims who died by suicide it is 9%, and 4% for familial abuse victims; and

• The proportion of other agencies (those outside the named categories in the table) is 22% and 20% for DHRs involving intimate partner victims or victims who died by suicide compared to 7% for familial abuse victims.

Table 35 Agencies identified in lessons to be learned, by type of victim

Agency	Percentage: Familial	Percentage: Intimate Partner	Percentage:Victims who died by suicide
Children’s social care	4%	7%	6%
Domestic Abuse Partnership or Service	1%	4%	8%
Health	48%	19%	30%
Many	15%	13%	10%
MARAC	1%	5%	6%
Other	7%	22%	20%
Police	4%	16%	9%
Probation	5%	2%	1%
Professionals	15%	11%	10%
Total occurrences	136	387	361

Figure 16 Agencies identified in lessons to be learned, by type of victim

202. The types of health agency identified in the lessons learned are shown in Table 36 and

203. Figure 17. The larger differences are:

• Mental Health Services are identified for 31% of the health lessons within DHRs with familial abuse victims, while they feature in 14% and 18% respectively for intimate partner victims and victims who died by suicide;

• Services from psychologists or psychiatrists are identified for higher proportions of DHRs with intimate partner victims or victims who died by suicide (7% for each) compared to none of the familial DHRs examined;

• Ambulance Services are identified for 8% of the lessons learned for DHRs involving intimate partner victims, and 2% of both familial abuse and death by suicide DHRs;

• Clinical Commissioning Groups are associated with 11% of familial abuse health lessons and 3% of DHRs with victims who died by suicide;

• The proportion of health lessons learned relating to GPs vary from 31% for DHRs involving intimate partner victims to 25% of those for familial abuse victims;

• Hospitals are linked to 9% of lessons for DHRs with familial abuse victims and 2% for victims who died by suicide;

• NHS Trusts are highlighted in 20% of the lessons learned for victims who died by suicide compared to 12% of DHRs for victims of familial abuse; and

• The group of agencies with only small numbers of references to each (the group “other”) account for 21% for DHRs involving victims who died by suicide compared to 11% for victims of familial abuse and 12% for intimate partner victims.

Table 36 Health agencies identified in lessons to be learned

Agency	Percentage: Familial	Percentage: Intimate Partner	Percentage: Victims who died by suicide
Ambulance Service	2%	8%	2%
Clinical Commissioning Group	11%	7%	3%
GP	25%	31%	27%
Hospital	9%	7%	2%
Mental Health Service	31%	14%	18%
NHS Trust	12%	15%	20%
Other	11%	12%	21%
Psychologist / Psychiatrist	0%	7%	7%
Total occurrences	65	74	110

Figure 17 Health agencies identified in lessons to be learned

Lessons to be learned

204. The lessons to be learned for the three types of victims are shown in Figure 18.

205. There are three lessons learned common across the three types of victims: information, risk and support.

206. The lessons for both familial and intimate partner victims relate to children and family.

207. The other identified lessons learned are different for each of the victim types.

Figure 18 Lessons to be learned by type of victim

10. Analysis of recommendations in Domestic Homicide Reviews

Introduction

208. Domestic Homicide Reviews should make specific, measurable, achievable, relevant, and time-bound (SMART) recommendations identifying actions to improve responses. This chapter summarises agencies who have been given responsibility for the recommendations.

209. The analysis is from randomly selected samples of reviews for each type of victim. These are 13 reviews (46% of total) of familial victims, 31 reviews (49% of total) of intimate partner victims and 16 reviews (48% of total) for victims who died by suicide. More information on the method used is given in Appendix 3. Selection of recommendations.

210. Reviews of victims who have died by suicide have an average of 18 recommendations and reviews of intimate partner victims the average 16 recommendations per review. Reviews of familial victims have an average of nine recommendations.

211. This chapter first looks at agencies given the responsibility for recommendations and then themes from the recommendations. Themes come from the most commonly used words in the recommendations. Appendix 3. Selection of recommendations gives more detail on the method for identifying the themes.

Recommendations by agency

212. Recommendations are made for a range of agencies or partnerships to implement. These are shown in Figure 19 and Table 37.

213. In terms of recommendations for different agencies for the different types of victim:

• Children’s Social Care feature in a higher proportion of recommendations (9%) for intimate partner victims and those who have died by suicide than for familial victims;

• Community Safety Partnerships have responsibility for 19% of the recommendations for familial victims, compared to 12% for intimate partner victims and 9% for victims who have died by suicide;

• Domestic Abuse Partnerships or services have responsibility to 11% of the recommendations where victims have died by suicide compared to 3% where victims are familial;

• Probation is named in a larger proportion of recommendations for intimate partner victims (9%) and familial victims (7%) than for victims who have died by suicide (1%); and

• Health agencies are responsible for the highest proportion for all three groups of victims. There is a relatively large difference between 37% of recommendations for familial victims compared to 23% for intimate partner victims.

214. For other types of agency, housing occurs in a small proportion of recommendations for familial victims and Probation is in a small proportion of recommendations where victims have died by suicide.

Table 37 Agencies with responsibility for recommendations, by type of victim

Agencies with responsibility for recommendations	Percentages: Familial	Percentages: Intimate partner	Percentages: Victim of suicide
Adult Social Care	3%	4%	3%
Children’s Social Care	4%	9%	9%
Community Safety Partnership (CSP)	19%	12%	9%
Domestic Abuse (DA) Partnership (or service)	3%	8%	11%
Education	0%	1%	4%
Health	37%	23%	28%
Housing	1%	5%	4%
Multi-Agency	7%	6%	7%
National	2%	2%	5%
Other	5%	8%	6%
Police	13%	15%	14%
Probation	7%	9%	1%
Total number of agencies with responsibility	124	529	312

Figure 19 Agencies with responsibility for recommendations

Recommendations by health agencies

215. The recommendations for health agencies are shown in Table 38 and Figure 20 for the different types of victim.

Table 38 Health agencies with responsibility for recommendations, by type of victim

Agencies with responsibility	Percentages: Familial	Percentages: Intimate partner	Percentages: Victim of suicide
Ambulance service	0%	7%	5%
Clinical Commissioning Group (CCG)	26%	20%	21%
Community Trust	0%	0%	6%
GP	0%	21%	12%
Hospital	0%	5%	1%
Mental Health	11%	7%	3%
NHS Trust	54%	37%	41%
Other	9%	3%	5%
Public Health	0%	1%	7%
Total number of health agencies with responsibility	46	122	87

Figure 20 Health agencies with responsibility for recommendations

216. Where victims are familial the recommendations are for NHS Trusts, CCGs, Mental Health Services and other health services (those not in the main categories). This also means that there are no recommendations for ambulance services, community trusts, GPs, Hospitals or Public Health.

217. There are large differences between types of victim when looking at recommendations for GPs: these are 20% of the recommendations for intimate partner victims, 11% where victims have died by suicide, and there are none for familial victims.

218. To say that a health service has no recommendations does not mean that it has was not involved.

Themes in recommendations

219. Themes in recommendations are shown in

220.

221. Table 39 and

223. Figure 21. The themes listed here are those which are in the top themes of at least one type of victim. They show 12 themes which have some relationship with each of the three types of victim, even if not appearing in the top five for every type of victim. From this:

• Eight themes occur for each type of victim: assessment; care; community; information, review, risk, staff, and training;

• Guidance appears for familial and intimate partner victims;

• Service appears for intimate partner victims and victims who have died by suicide;

• Family is a theme in the recommendations for familial victim; and

• Victim is itself the most frequently occurring theme in the recommendations for victims who died by suicide.

Table 39 Themes in recommendations

Theme	Percentages: Familial	Percentages: Intimate partner	Percentages: Victim of suicide	Rank of added percentages
Assessment	4%	2%	2%	9
Care	4%	4%	2%	8
Community	5%	3%	2%	7
Family	4%	0	0	12
Guidance	4%	2%	0	11
Information	4%	4%	5%	3
Review	3%	5%	5%	2
Risk	9%	5%	4%	1
Service	0	6%	6%	6
Staff	2%	6%	4%	5
Training	2%	5%	6%	4
Victims	0	0	7%	10
Total	542	552	341

Figure 21 Themes in recommendations

224. The most frequently occurring themes for each type of victim are examined in more detail with examples given.

Themes from recommendations for familial victims

225. Table 40 shows themes from the recommendations for familial victims (each of the themes here has at least 5% of the total^{[footnote 51]}).

Table 40 Themes in recommendations for familial victims

Theme	Percentage of total
Risk	9%
Community	5%
Information	4%
Assessment	4%
Care	4%
Family	4%
Guidance	4%
Sum of all words counted and used	542

226. Examples of recommendations which use these words are given below.

Risk

xxx Constabulary must ensure that safeguarding plans are created for offenders identified as ‘adults at risk” and/or vulnerable

Community

Better publicity is also needed about who to talk to if someone has concerns about another’s mental health. This needs to be particularly focused on reaching African and Caribbean communities who are currently over-represented as patients within mental health services.

Information

To highlight the importance of historical information and that this is an indicator of future risk.

Assessment

Where possible, health providers involved in this review should assure themselves that in assessing risk to others, their tools and practices embrace all assessments,

Carer

People identified as carers should have an assessment completed by Adult Services. This should not only assess whether they are coping with their caring responsibilities but should also probe to identify if other events in their lives may be undermining their caring role.

Family

That the CCG undertakes further analysis to identify the barriers for GPs in completing details of family groups and relationships to identify ways of improving practice.

Guidance

Adult Social Care should ensure and provide assurances that pathways and online guidance to section 13(4) MHA ‘nearest relative’ assessments requests are clear and accurate, properly publicised and understood by call handlers receiving requests for such support as well as those managing and providing this service within Adult Social Care.

Themes in recommendations for victims of intimate partnerships

227. Themes from the recommendations for intimate partner victims are shown in

228. Table 41. Together they account for 28% of the total occurrences of the most common words (which total 1,983).

Table 41 Themes in recommendations for intimate partner victims

Theme	Percentage of total
Staff	6%
Service	6%
Training	5%
Review	5%
Risk	5%
Sum of all words counted and used	1,983

229. Examples of recommendations with these words are given below.

Staff

Children’s Social Care staff are supported to reflect on the importance of professional curiosity in practice. To ensure that every contact with a family counts. That staff are alert to the possibility of domestic abuse and follow up all seemingly unimportant pieces of information ensuring information is triangulated.

Service

c) Consider barriers to Patients access to service: these should be clearly recorded on their records. For example,

• Being vulnerable
• Being housebound
• Repeated DNA’s [Did Not Attend]

Training

That Foundations undertakes refresher training with all staff about MARAC aims and procedures.

Review

That xxx Strategic Partnership shares this DHR report with xxx Safeguarding Children Partnership in order that the latter partnership may consider whether children’s social care should review the arrangements for the grant of Child Arrangement Orders, in which children’s social care is involved, or invited by Cafcass [Children and Family Court Advisory and Support Service] to be involved….

Risk

Probation staff to be reminded to use all available sources of information, e.g. previous records when making risk assessment and sentence plan recommendations.

Most common themes in recommendations for victims who died by suicide

230. The most commonly appearing themes in the recommendations for victims who died by suicide are given in Table 42.

Table 42 Themes in recommendations for victims who died by suicide

Theme	Percentage of total
Victim	7%
Training	6%
Service	6%
Review	5%
Information	5%
Sum of words counted and used	1,186

231. Examples of recommendations containing these words are given below.

Victim

Information is passed to all victims of domestic abuse about domestic abuse support services and victims are encouraged to contact services for support

Training

Training for Children’s Social Workers in relation to adult safeguarding and procedures.

Service

Development of domestic abuse care bundles:…. This will include the domestic abuse pathway, clinical photography prompts, body maps and literature for safety planning and onward referrals to IDVA and safeguarding services.

Review

When members of the public contact the Police with information regarding vulnerability or domestic abuse, this should not be filed at source but forwarded for review by specialists to ascertain who is the most appropriate recipient of the information and ensure they receive it

Information

Children’s Services information requests from multi-agency partners should include detail relating to the specific safeguarding concern as domestic abuse to ensure all agencies can flag on their respective systems

11. Family contribution and support through the DHR process

232. The Multi-Agency Statutory Guidance for Domestic Homicide Reviews (2016)^{[footnote 52]} sets out the importance of contributions and engagement with family, friends, work colleagues, neighbours, and the wider community. The management information form requested with the DHRs included questions to record family contributions to the review process. This chapter looks at answers for 123 reviews classed as either familial (28 DHRs), intimate partner (62) or where victims had died by suicide (33).

Did the family contribute to the DHR?

233. Families contributed to 84% of the DHRs. The proportion contributing was 82% where the victim had been an intimate partner, or the victim died by suicide. The proportion was higher (89%) where the relationship had been familial.

234. In 16 reviews the invitation had been made but was declined. There were four DHRs where no contact was achieved.

Were the family consulted about the terms of reference?

235. For 77% of the reviews the family were consulted about the terms of reference. For reviews where the victim had died by suicide 72% of the families were consulted. Where the DHR involved a victim of an intimate partner or former partner this was 78%, and for a victim of familial abuse this was 80%.

Did the family have the support of an expert specialist advocate?

236. Support from an expert specialist advocate was taken up by 57% of the families^{[footnote 53]}. It was noted that in 5% of the reviews support was offered but was declined.

237. Some forms gave more detail on the organisation chosen by the family. In 33 reviews support was through Victim Support ^{[footnote 54]}. In 26 reviews support was provided through Advocacy After Fatal Domestic Abuse (AAFDA^{[footnote 55]}). All support used by families of victims who died by suicide was by AAFDA.

Did the family receive the draft report to comment on?

238. In 78% of the DHRs the family received the draft report on which to comment ^{[footnote 56]}.

Did the family attend the DHR panel?

239. In 14% of the reviews the family attended the DHR panel. There was a difference where the relationship had been familial, where in 7% of the reviews a member of the family attended.

240. As was noted earlier in the report, 81% of the DHRs referenced COVID and this is likely to have had an impact on the ability to attend a face to face or virtual DHR Panel.

Appendix 1. Questions on information forms

Guidance or definition given with some questions are placed at the end of the Appendix.

The form uses the following abbreviations:

CSP	Community Safety Partnership
DHR	Domestic Homicide Review
PTSD	Post-Traumatic Stress Disorder
MAPPA	Multi-Agency Public Protection Arrangements
MARAC	Multi-Agency Risk Assessment Conference

PLEASE MARK EACH BOX: IF QUESTION IS NOT APPLICABLE PLEASE STATE: N/A IF ANSWER IS NOT KNOWN PLEASE STATE THIS OR PUT: N/K

Name of Community Safety Partnership

Local Authority

Police Force Area

Date of death

Location of death

Is location victim’s home address? (Y, N or N/K)

Review Panel Chair

Review Author

Date Home Office notified of DHR

Local DHR Reference

Date report completed by author

Date signed off by CSP Board

Date submitted to Home Office by CSP Board

Home Office Reference Number given for report

1. Victim/s

	Victim 1	Victim 2	Victim 3
Sex of victim/s
Age at time of death
Relationship to perpetrator
Ethnicity
Nationality
Is or was the victim a Carer? (Y, N or
N/K)
If Yes, had they had a Carer’s Assessment under the Care Act? (Y, N or N/K)

Vulnerabilities. Please mark (e.g. X) for ALL that apply

	Victim 1	Victim 2	Victim 3
Illicit Drug Use
Mental Ill-Health
Physical Disability
Pregnancy
Problem Alcohol Use
Other - Please state

Mental health Issue/s identified in the DHR. Please mark ‘X’ for ALL that apply

	Victim 1	Victim 2	Victim 3
Adjustment Disorder
Anxiety
Dementia or Alzheimer’s disease
Depression
Low mood / anxiety
Panic attacks
Psychosis
PTSD
Self-harm
Suicidal thoughts
Suicide attempt/s
Not specified (please state)
Any serious or life limiting illness? (Y, N or N/K)
If Yes please describe
Has the victim been a target of an abuser before? (Y, N or N/K)
if Yes please state by whom?

2. Perpetrator/s

	Perpetrator 1	Perpetrator 2
Sex of perpetrator
Age at time of death
Relationship to victim/s
Ethnicity
Nationality
Is or was the perpetrator a Carer? (Y, N or N/K) If YES state for whom they were a carer?
If Yes, had they had a Carer’s Assessment under the Care Act? (Y, N or N/K)

Vulnerabilities. Please mark (e.g. X) for ALL that apply

	Perpetrator 1	Perpetrator 2
Illicit Drug Use
Mental Ill-Health
Physical Disability
Problem Alcohol Use
Other - Please state

Mental health Issue/s identified in the DHR. Please mark ‘X’ for ALL that apply

	Perpetrator 1	Perpetrator 2
Adjustment Disorder
Anxiety
Dementia or Alzheimer’s disease
Depression
Low mood / anxiety
Panic attacks
Psychosis
PTSD
Self-harm
Suicidal thoughts
Suicide attempt/s
Not specified (please state)
Any serious or life limiting illness? (Y, N or N/K)
If Yes please describe
Had the perpetrator abused previous partner/s or family member before? (Y, N or N/K)
If Yes please state who the victim was
Was the perpetrator known to agencies as an abuser? (Y, N or N/K)
If Yes please state which agencies
Has the perpetrator any previous offending history? (Y, N or N/K)
If Yes please state offences committed

Was the perpetrator being managed or supervised by, or attending any of the following? Please mark (e.g. X) for ALL that apply

	Perpetrator 1	Perpetrator 2
Attending or had attended a Perpetrator Programme
Drug and Alcohol Services
MAPPA
Mental Health Services
National Probation

3. Crime details, MARAC and Outcome of Trial

	Y, N or N/K
Had the victim been referred to MARAC? (Y, N or N/K)
Was the case heard at MARAC before the homicide? (Y, N or N/K)

Method of killing. If relevant please state weapon used

	Details
Blunt Force trauma
Fire Arm
Stabbing Knife
Strangulation
Other, please state
Cause of death - results from Post-Mortem

Details of Court verdict. Please mark (e.g. X) for ALL that apply

	Details
Murder
Manslaughter
Diminished responsibility
Unfit to Plead
Not Guilty
Details of sentence/s AND sentence tariff/s

4. Details, if reviewing suicide or murder / suicide

	Details
Is DHR reviewing a murder and suicide? (Y or N)
If DHR is reviewing a death by suicide, please answer the following about the Person who took their life by Suicide
Sex and Age of deceased
Method of suicide
Is the suicide by the perpetrator who is responsible for the victim’s homicide? (Y, N, N/K)

5. Aggravating factors

Aggravating factors in DHR. Please mark (e.g. X) for ALL that apply

	Details
Coercive control
Digital Stalking
Forced Marriage
Honour Based Violence
Financial Abuse
Immigration issues (V if relevant for victim and / or P if relevant for perpetrator)
Physical stalking

6. Details of children if relevant (0-18yrs)

	Child/Children’s details
Were there any children living, or regularly staying in the household? (Y, N or N/K)
Were children present when the homicide occurred?
If YES, please give sex of child/ren
If YES, please give age of child/ren
Were children subject to Child Protection procedures due to Domestic Abuse prior to the homicide? (Y, N or N/K)
Any children removed into Care of Local Authority? (Y, N or N/K)

7. Family contribution and support though DHR process

	Details
Did the family contribute to the DHR? (Y, N or N/K)
If answer is N, please comment
Were the family consulted about the terms of reference? (Y, N or N/K)
If answer is N, please comment
Did the family have the support of an expert specialist advocate? (Y, N or N/K)
If answer is Y, please specify
Did the family receive the draft report to comment on? (Y, N or N/K)
If answer is N, please comment
Did the family attend the DHR panel? (Y, N or N/K)
If answer is N, please comment

For Ethnicity (Office for National Statistics)

White

1. English/Welsh/Scottish/Northern Irish/British

2. Irish

3. Gypsy or Irish Traveller

4. Any other White background, please describe

Mixed/Multiple ethnic groups

5. White and Black Caribbean

6. White and Black African

7. White and Asian

8. Any other Mixed/Multiple ethnic background, please describe

Asian/Asian British

9. Indian

10. Pakistani

11. Bangladeshi

12. Chinese

13. Any other Asian background, please describe

Black/ African/Caribbean/Black British

14. African

15. Caribbean

16. Any other Black/African/Caribbean background, please describe

Other ethnic group

17. Arab

18. Any other ethnic group, please describe

Notes given in the form, next to relevant questions

• Ethnicity: please use codes / descriptions given at foot of the form.

• Carer: the definition of a carer in this context refers to an adult or young person who is caring for someone due to their health and social care needs. This includes mental health as well as physical health support, which would entitle the carer to a Carer’s Assessment under the Care Act 2014. The Children and Families Act 2014 also includes duties for the assessment of young carers and parent carers of children under 18.

• Physical disability: a person is considered to have a disability if they have a long- standing illness, disability or impairment which causes difficulty with day-to-day activities (Equality Act 2010).

• Life-limiting illness is a term used to describe an incurable condition that will shorten a person’s life, though they may continue to live active lives for many years. There is a wide range of life-limiting illnesses, including heart failure, lung disease, neurological conditions, such as Parkinson’s and Multiple Sclerosis, and cancer that is no longer responding to treatment intended to cure. stclarehospice.org.uk/what-does-that-mean/

• Details of sentence/s AND sentence tariff/s: i.e. Guilty of Murder, Manslaughter, or Manslaughter Diminished Responsibility etc, then the sentence tariff i.e. minimum 25yrs, Hospital Order with Restriction etc.

Appendix 2. Selection of lessons to be learned

The approach to selecting lessons to be learned has been to take random samples of DHRs (for each of the three types of victim) and, within each sample, to extract the lessons learn from the DHRs. On lessons to be learned the Home Office (2016) Statutory Guidance for the Conduct of Domestic Homicide Reviews^{[footnote 57]}, page 38, states “this part of the report should summarise what lessons are to be drawn from the case and how those lessons should be translated into recommendations for action”.

Lessons learned are examined to show which agencies they were related to. These can be in the learning text or in a heading under which they appear in the DHR. The names given to the categories comes from the words used in the DHRs. The term agency is used to cover the area of work and includes partnerships or organisations. The categories used in this report can be specific e.g. GP or can be a range e.g. the term education frequently refers to schools but sometimes referred to University.

Lessons are attributed to the agency or agencies where specifically mentioned e.g. The impact of the loss of her children on the victim’s mental health. The permanent loss of Child 1 and 2 from her care and her limited access to Child 3 appears to have contributed to xx’s mental health issues. Children’s social care have included a single agency recommendation to ensure appropriate emotional support is offered to parents whose children are removed from their care is allocated to children’s social care.

If more than one agency is referred to then the lesson is allocated to each e.g. The GP appropriately responded in signposting xx for help. However, there was no assistance nor advice provided to the partner who had attended with him. In later case discussion with GPs at the practice, it was clear there was no consensus as to what would have been good practice in this situation with both victim and perpetrator present. This uncertainty may also exist in other GP surgeries. Therefore, the Panel would suggest the CCG establish with a subject matter expert what good practise would be in such a scenario and communicate that to all GP surgeries and other similar frontline services. This is attributed to both health categories: GP and Clinical Commissioning Group.

There are lessons for numbers of agencies but these are not individually referred to. Here the category “many” is used e.g. “It is feasible that xx and her family may have been either reluctant to contact agencies to seek help with marital/domestic concerns prior to 2019 or may have been unaware of the availability of services”.

In addition to references to agencies there are also references to professionals e.g. “Professionals to be made aware that the watching of very violent media content may lead to increased aggression in individuals”. For these lessons learned the classification “professional” is used. If the term “professional” is used with an agency reference then both relevant categories are used.

References to agencies are to those of the type which existed at the date of the homicide e.g. Clinical Commissioning Groups, which ceased in July 2022 ^{[footnote 58]}. References to Community Rehabilitation Companies have been classified as Probation ^{[footnote 59]}.

Categories for agencies in lessons learned

Adult Social Care	Health - GP
Children’s Social Care	Health - Many
Community based organisations	Health - Mental Health Service
Community Safety Partnership	Health - NHS Trust
Court	Health - Other
Crime and Drugs Partnership	Health - Psychologist / Psychiatrist
Crown Prosecution Service	National
Drugs and alcohol service	Housing
Domestic Abuse Partnership or Service	Local Authority
Education	Many
Health - Ambulance Service	MARAC
Health - Clinical Commissioning Group	Police
Health - Community Trust	Prison
Health - Health and Well-being Board	Probation
Health - Hospital	Professionals

Lessons to be learned were identified though a word count^{[footnote 60]} to give the words most often used. For each of the three types of DHR there follows a table to give more clarity on words used or not used.

The categorisation of words to lessons learned moves different variations of words into the same category. For example: consider, considered and consideration are all allocated to the category “consider”.

Lessons to be learned for familial abuse DHRs

Seventeen of the 28 familial DHRs were randomly selected. One hundred and thirty six lessons learned were extracted. These contained 4,899 words.

These were searched to see which occurred most often. Following adjustments the most relevant used eight lessons were selected. The most common words are shown in the table below, together with reasons for adjustment or why not used. If a word was used many times in a lesson learned it would only be counted once.

Words	Initial number of times counted	Comment relating to non-inclusion of word as a lesson learned	Number of adjusted lessons
Assessment	34	Included. Occurrence where link to risk or care not used as they are included with those words.	24
Risk	33	Lesson learned.	24
GP	30	Not included. The word often used to show where the learning was needed rather than the type of learning.
Health	28	Not included as occurrence is with agencies where learning is recommended rather than the type of learning.
Domestic	23	Not included. The majority of references were as Domestic Homicide Reviews and not as part of learning activity.
Mental	21	Not included. Occurrence was frequently with mental health service (agency).
Case	21	Not included. Although an initial number of 21 times counted, final number (11) was reduced as the word case was often used to mean example e.g. “in this case”.
Abuse	21	Not included. The term abuse was most frequently used linked to domestic abuse strategy or similar. This is used in chapter on Recommendations.
Professionals	20	Not included. The use was for where the learning is needed rather than the type of learning.
Support	19	Lesson learned.	14
Family	19	Lesson learned.	16
Care	18	Lesson learned.	13
Practice	17	Not included. Word most often used in General Practice.
Service	15	Not included as use was where the learning is needed (e.g. mental health service) rather than the type of learning. This was analysed in other parts of the report.
Mr	13	Not included. Referred to victim or perpetrator and not as a lesson learned.
Lack	13	Not included as closely related to other uses e.g. “The system for allocation of medical reviews was reactive and not fit for purpose, and waiting lists were lengthy and unmanaged. This resulted in a lack of medical oversight of Mr X’s care for 18 months”.
Ensure	13	Not included as term relates to recommendations and is covered in that chapter.
Children	13	Lesson learned.	12
Information	12	Lesson learned.	20
Aware	12	Lesson learned.	17
Agencies	12	Not included as general use on where the learning needed e.g. “appropriate agency support” is covered through lesson “support”.

Lessons to be learned for intimate partner DHRs

Thirty five of the 63 intimate partner DHRs were randomly selected. One hundred and thirty six lessons learned were extracted. These contained 4,899 words.

These were searched to see which words occurred most often, further information given below. If a word was used many times in a lesson learned it would be only counted once.

Words	Initial number of times counted	Comment relating to non-inclusion of word as a lesson learned	Number of adjusted lessons
Abuse	299	Not included. Many uses were as an aggravating factor and this covered in separate section in report.
Domestic	288	Not included. The majority or references were as Domestic Homicide Reviews and not as part of learning activity.
Risk	213	Lesson learned.	97
Victim/s	150	Not included. Frequent references to agencies (e.g. Victim Support). These are covered by counts of agencies.
Police	122	Not included as these are where the learning is needed rather than the type of learning.
Services	121	Not included as these are where the learning is needed rather than the type of learning.
Agency	118	Not included as reference to agencies where learning is recommended rather than type of learning.
Support	117	Lesson learned.	72
Child / Children	115	Lesson learned. Number reduced by omitting reference to children’s social care as this refers to agencies.	35
Information	114	Lesson learned.	62
Care	96	Not included as many examples outside lesson learned (e.g. children’s social care or Social Care Act).
Assessment	90	Not included. Large proportion connected to risk assessment and through this.
Safeguarding	87	Lesson learned.	46
Health	76	Not included as occurrence is with agencies where learning is recommended rather than the type of learning.
Case	70	Not included as many references covered through other lessons.
Review	67	Not included as the majority of references are to the review (DHR) itself.
Contact	64	Lesson learned.	40
Violence	64	Not included as inherent in Domestic Homicide Reviews.
Betty	61	Not included as individual’s pseudonym.
Family	61	Lesson learned.	38
Professionals	56	Not included as these are where the learning is needed rather than the type of learning.
Social	55	Not included as related to social care and covered by the information on service provision.
Perpetrator	54	Not included as if directly related to action
was within other lesson learned.
Need	53	Not included as frequent use in relation to other lessons learned.
Mental	21	Not included as use was frequently to service agency and so covered elsewhere.

Lessons to be learned from DHRs where victim died by suicide

Eighteen of the 63 DHRs where victims died by suicide were randomly selected. 173 lessons learned were extracted. These contained 1,845 words.

These were searched to see which occurred most often. For words not selected as lesson learned the reasons are given in the table below. If a word was used many times in a lesson learned it would be only counted once.

Words	Initial number of times counted	Comment relating to non-inclusion of word as a lesson learned	Number of adjusted lessons
Abuse	245	Not included. General reference to domestic abuse, lessons from this included in other categories.
Domestic	186	The majority or references were as Domestic Homicide Reviews and not as part of learning activity.
Service/s	162	Not included as these are where the learning is needed rather than the type of learning.
Rosie	122	Pseudonym of victim.
Risk	102	Lesson learned. Those lessons where the references to risk training are counted under training.	46
Health	99	Not included as occurrence is with agencies where learning is recommended rather than the type of learning.
Information	93	Lesson learned. The use of the term information to support the DHR has not been included e.g. “although in reviewing further information as already describe in paragraph 13.5.2”	47
Support	92	Lesson learned. Reduced number to avoid duplication e.g. support for training is counted under training.	55
Victim/s	90	Not included as lessons learned are elements which relate to victims. Many references are to agencies (e.g. Victim Support).
Agencies	77	Not included as a separate lesson as these are in general where the learning is needed.
Police	68	Not included as these are where the learning is needed rather than the type of learning.
Dave	66	Not included. pseudonym of individual.
Mental	61	Not included as most were references associated directly with agencies.
MARAC	60	Not included as these are where the learning is needed rather than the type of learning.
Children	58	Not included following exclusion of direct service-related instances (e.g. children’s social care).
Professionals	54	Not included as these are where the learning is needed rather than the type of learning.
Social	51	Not included as these are where the learning is needed (e.g. social services or social workers) rather than the type of learning.
Suicide	49	Not included as this is the context for the DHRs in this section.
Training	48	Lesson learned. Removed duplication in a lesson learned.	29
Care	47	Not used after count following exclusion of direct service-related instances (e.g. children’s social care or Social Care Act).
GP	46	Not included as these are where the learning is needed rather than the type of learning.
Assessment	46	A large proportion of the use of assessment is connected to risk assessment and will be covered through risk lesson learned.
Consider	42	Not included. Link was to lessons covered in other categories i.e. “consider risk assessment”.
Review	41	Lesson learned. The reduction is to exclude the reference to the DHR e.g. “as in this review”.	15
Staff	40	Lesson learned. Number reduced by excluding those which would be part of other lesson (e.g. staff training is counted in training).	18
Case	39	Not included directly as would be part of other categories e.g. “Through reflective supervision, a social worker’s manager will be able to review case worker and challenge any perceived professional grooming” would be with review.
Need	38	Not included as references included in other categories on how need might be met.

Appendix 3. Selection of recommendations

241. This Appendix describes the approach to Chapter 10: Analysis of recommendations in Domestic Homicide Reviews.

242. The recommendations are from a random selection of DHRs from 124 which could be classed as either familial, intimate partner or where the victims died by suicide.

Victim type	Number of DHRs	Number in sample
Familial	28	13
Intimate partner	63	31
Victims who died by suicide	33	16
Total	124	60

243. The samples of DHRs were searched for recommendations and these were extracted.

Victim type	Number of recommendations in DHRs: Maximum	Number of recommendations in DHRs: Minimum	Number of recommendations in DHRs: Average (mean)
Familial	28	4	9
Intimate partner	46	0	16
Victims who died by suicide	45	2	18

244. The recommendations were searched to identify the agencies or partnerships given responsibility. The focus has been on agencies required to take action, e.g. “XX Drug and Alcohol Services to provide awareness sessions to the Safeguarding Adult Board” has been put with Drug and Alcohol services only.

245. Where more than one agency is given responsibility for undertaking action then this is allocated to each e.g. “… Clinical Commissioning Groups, Safeguarding Children Partnership, Safeguarding Adults Boards and health providers (whether NHS or private) should, as a matter of urgency, agree a consistent policy and practice to support routine and targeted enquiry for domestic abuse which ensures that every opportunity is taken to identify where such abuse may be being perpetrated and to signpost or offer services appropriate to need.” This is marked for Clinical Commissioning Group, Children’s Social Care, Adult Social Care and health – other.

246. The classification “multi agency” is used where more than one agency is implied but named; for example, “That all local agencies raise awareness amongst staff about the risks posed in sibling relationships so that they are more alert to the warning signs. It is recommended that this is overseen by the Community Safety Partnership to ensure a consistent approach across agencies” has been classified as both multi agency and Community Safety Partnership.

247. The second part of the chapter identifies commonly used words in recommendations which were not agencies.

248. This can be considered a five stage process for each type of victim.

249. The first part is the selection of the 40 most common words from the samples for category of victim ^{[footnote 61]}.

250. After this, words which directly related to agencies e.g. partnership, police, social were removed.

251. As a third stage where words had the same base these were combined e.g. include and including; service and services.

252. This revised list of words gives the counts shown in each table of the most common words used e.g. 542 total words for reviews with familial victims.

253. The fourth stage was the exclusion of a small number of words which did not describe a main area of work were removed e.g. “ensure” as it was regarded as a general reference to the purpose of the recommendation rather than word related such as “information” or “training”.

254. The aim was to give examples of a small number of recommendations for each type of victim. This was the use of words from the fourth stage which were counted for more than 5% of the total of revised words. For words for intimate partner victims this was five words. For victims who died by suicide this also came to five words with the fifth word (information) being 4.8%. Where the victims were familial victims there was less concentration in separate words. Risk and community were 8.7% and 4.6%. The fifth word, care, was counted for 3.7% of the total. This was the same as another four words; so, all five words have been included – giving a larger number of words than for intimate partner or suicide victims.

Glossary

Acronyms	Full description
ADHD	Attention deficit hyperactivity disorder
AHMP	Approved Mental Health Professionals
Cafcass	Children and family court advisory and support service
CCG	Clinical Commissioning Group
CMHT	Community Mental Health Team
CPA	Child to parent abuse
CRC	Community Rehabilitation Company
CSE	Child sexual exploitation
CSP	Community Safety Partnership
DA Partnership	Domestic Abuse Partnership
DASH	Domestic abuse, stalking and ‘honour’- based violence (risk assessment tool)
DHR	Domestic Homicide Review
DNA	Did not attend
GP	General Practitioner / Doctor
HBA	Honour Based Abuse
ICB	Integrated Care Board
IDVA	Independent Domestic Violence Advisor /Advocate
MARAC	Multi-Agency Risk Assessment Conference
MASH	Multi-Agency Safeguarding Hub
MHA	Mental Health Act
MIR	Management Information Report
N/A	Not applicable
NICE	National Institute of Health and Care Excellence
N/K	Not known
NPS	National Probation Service
OASys	Probation Offender Assessment System
PTSD	Post-Traumatic Stress Disorder
VPA	Vulnerable Person Assessment

1. Home Office (no date) Criteria for considering Domestic Homicide Review reports https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/207602/cri teria-DHR-web-v2.pdf [Accessed 30th April 2023] ↩

2. Potter, R. (2022), Key findings from analysis of domestic homicide reviews, https://www.gov.uk/government/publications/key-findings-from-analysis-of-domestic-homicide-reviews/key- findings-from-analysis-of-domestic-homicide-reviews [Accessed 30th April 2023] ↩

3. Potter, R. (2022), Domestic homicide reviews, quantitative analysis of domestic homicide reviews October 2020 - September 2021, https://www.gov.uk/government/publications/key-findings-from-analysis-of- domestic-homicide-reviews [Accessed 30th April 2023] ↩

4. Home Office (2016) Multi-agency Statutory Guidance for the Conduct of Domestic Homicide Reviews, from paragraph 13, page 7 and paragraph 18, page 8. Source: https://www.gov.uk/government/publications/revised-statutory-guidance-for-the-conduct-of-domestic- homicide-reviews [Accessed 30th April 2023] ↩

5. The omission of four deaths from this report is from the complexity of relationships and cause of death, and the difficulty of placing these into the three main categories of victim. ↩

6. This may be due to the victim not being found for some time. ↩

7. In the forms some questions are asked to indicate Y, N or N/K. ↩

8. The term victim includes those who have died by suicide. ↩

9. From a search of 124 reviews. ↩

10. Referenced in footnotes 3 and 4. ↩

11. “In 2022, there were 5,642 suicides registered in England and Wales (10.7 deaths per 100,000 people); this is consistent with 2021 (5,583 deaths; 10.7 per 100,000).” (Office for National Statistics, 2022, Suicides in England and Wales: 2021 registrations, page 2 Office for National Statistics (2023) Suicides in England and Wales: 2022 registrations: https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/suicidesi ntheunitedkingdom/2022registrations [Accessed 7th March 2024]. ↩

12. The regions used are those as shown by the Office for National Statistics, England, Detailed information on the administrative structure within England:https://www.ons.gov.uk/methodology/geography/ukgeographies/administrativegeography/england [Accessed 22nd February 2024]. ↩

13. Not at home address can include in hospital, in the street, or perpetrator’s address. ↩

14. Average age as median. ↩

15. This report omits information on 4 victims as the information given does simply place them into the three categories used (familial, intimate partner or victim who died by suicide) e.g. “no relation (new partner of perpetrator’s wife)”. ↩

16. From the 129 DHRs information on five perpetrators has been omitted in this report due to the complexity of aligning with the two categories of perpetrator: familial or intimate partner. For example, three perpetrators have been given a relationship as “lodger”. Information on four perpetrators has also been omitted from the Court verdicts e.g. “xxx was arrested and claimed that she had acted in self-defence. She was charged with her father’s murder but following a two-week trial, the jury found her not guilty of all charges.” ↩

17. As examples, the terms used include: ex intimate partner, ex-partner, ex-partner and former lodger, ex- wife, former partner, girlfriend, husband, partner, partner / ex-partner, recent intimate relationship, wife, and wife or partner. ↩

18. Office for National Statistics, 2022, Suicides in England and Wales: 2021 registrations, page 3, Office for National Statistics (2023) Suicides in England and Wales: 2022 registrations: https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/suicidesi ntheunitedkingdom/2022registrations [Accessed 7th March 2024]. ↩

19. This does not take into account the full age structure of the population e.g. whether an ethnic group has a higher proportion aged 65 and over. ↩

20. Department of Health (2013) Guidance for health professionals on domestic violence. ↩

21. Office for National Statistics (ONS), released 9 February 2023, Homicide in England and Wales: year ending March 2023, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/articles/homicideinenglandandwale s/yearendingmarch2023 ↩

22. Answers were given where the victim died by suicide. ↩

23. In two reviews children were present at the time of the homicide (both were “intimate partner”) but did not live or regularly stay. ↩

24. There were small numbers of DHRs where the sex of children was given but the age was marked as unknown. There were also a small number of reviews where the sex of children was given as unknown, but the age was given. ↩

25. A local, multi-agency victim-focused meeting where information is shared on the highest risk cases of domestic violence and abuse (includes considerations of child protection) between different statutory and voluntary sector agencies (available through Ministry of Justice (n.d.) Search for an Acronym https://ministry- of-justice-acronyms.service.justice.gov.uk/ [Accessed on 22nd February 2024]). ↩

26. The information came on newer forms for 94 victims and for older forms for 39. One of the small differences is differences in vulnerabilities which in the older form had a category “learning difficulties” and no category for “other”. The older form containing learning difficulty was counted as “other” and all forms then used for analysis. This will result in a slight undercount of number of vulnerabilities in the older forms. ↩

27. The vulnerability identified as “other” has not been included as it is not available for newer forms. Nine victims of the 95 on the newer forms had a vulnerability marked as “other”. ↩

28. Defined as an adult or young person who is caring for someone due to their health and social care needs. This includes mental health as well as physical health support, which would entitle the carer to a Carer’s Assessment under the Care Act 2014. The Children and Families Act 2014 also includes duties for the assessment of young carers and parent carers of children under 18. ↩

29. The definition “life-limiting illness is a term used to describe an incurable condition that will shorten a person’s life, though they may continue to live active lives for many years. There is a wide range of life- limiting illnesses, including heart failure, lung disease, neurological conditions, such as Parkinson’s and Multiple Sclerosis, and cancer that is no longer responding to treatment intended to cure” from St. Clare Hospice (n.d.) stclarehospice.org.uk/what-does-that-mean/ [Accessed 22nd February 2024]. ↩

30. Answers Yes or No were given for 112 victims, and marked as Don’t Know for 21. ↩

31. Information is not available a second victims in a DHR as the information file does not have the structure to provide it for this part (see Appendix 1). Fourteen DHRs where the form is marked by N/K – don’t know. Ten of the 14 DHRs where answer given as N/K are those who died by suicide. ↩

32. Information on MARACs given by SafeLives (n.d.) Resources for Marac meetings, https://safelives.org.uk/practice-support/resources-marac-meetings [Accessed 22nd February 2024]. ↩

33. Answers were given in the form where the victim had died by suicide. The answers have also been taken as when the victim had been referred to a MARAC, and then was the case heard before the homicide.

    There were three victims where it was indicated that a case was heard before the homicide but they had not been referred to a MARAC. These have not been included. ↩

34. The information here is for 90 victims from the newer forms where financial abuse was an option for an answer. ↩

35. In the forms there is one set of questions on aggravating factors for each review. There are 92 victims in the 90 reviews in the new forms used. ↩

36. As small number shown in category “other” Figure 6. ↩

37. From the reviews information is not used from four DHRs where the perpetrators were either not charged or found not guilty. ↩

38. The question gave the option of five different vulnerabilities (counting “other’ as one). No perpetrator had more than three. ↩

39. Agencies classified as “other” include HM Courts and Tribunal Service, Home Office, housing, prison, schools and Youth Offending Teams. ↩

40. Randomly selected 17 of the 28 DHRs with familial victims. ↩

41. The term “types of individual” is used to show lessons learned which refer to the needs of professionals. ↩

42. The Health and Social Care Act 2022 abolished CCGs and replaced them with 42 Integrated Care Systems (ICSs) frequently referred to as Integrated Care Boards (ICB) in England from 1 July 2022. ↩

43. Local Authorities are responsible for a range of services. The general term is used here as some lessons learned are for local authority generally e.g. “County Council does not have a structure for the oversight of actions from domestic homicide reviews”. ↩

44. Randomly selected 34 of the 63 DHRs where victims had been intimate partners. ↩

45. The use of the term “professionals”. ↩

46. With the remaining lesson learned classified as “information – other”. ↩

47. Randomly selected 18 of the 33 DHRs where victims died by suicide. ↩

48. The term “types of individual” is used to show lessons learned which refer to the needs of professionals. ↩

49. Appendix 2. Selection of lessons to be learned gives more detail. ↩

50. This example also included “a Police staff researcher post”. ↩

51. Detail on identification of commonest words given in Appendix 3. Selection of recommendations ↩

52. https://www.gov.uk/government/publications/revised-statutory-guidance-for-the-conduct-of-domestic- homicide-reviews ↩

53. With similar levels of 59%, 57% and 54% for F, IP and S. ↩

54. https://www.victimsupport.org.uk/ ↩

55. https://aafda.org.uk/ ↩

56. With similar levels of 74%, 80% and 76% for F, IP and S. ↩

57. Home Office (2016), Multi-agency Statutory Guidance for the Conduct of Domestic Homicide Reviews. Available from https://assets.publishing.service.gov.uk/media/5a80be88e5274a2e87dbb923/DHR-Statutory- Guidance-161206.pdf [Accessed 22 February 2024]. ↩

58. NHS Confederation (2021), What were clinical commissioning groups? Available from https://www.nhsconfed.org/articles/what-are-clinical-commissioning-groups [Accessed 22 February 2024]. ↩

59. Beard, J (2021) Unification of probation services https://commonslibrary.parliament.uk/research- briefings/cbp-9252/ [Accessed 22 February 2024]. ↩

60. Through the use of Voyant: https://voyant-tools.org/ ↩

61. Files of the text for recommendations are uploaded into https://voyant-tools.org/ ↩