The specifics of homicide-suicide in France

Full text

Journal of Forensic and Legal Medicine 100 (2023) 102596
Available online 5 October 2023
1752-928X/© 2023 Elsevier Ltd and Faculty of Forensic and Legal Medicine. All rights reserved.
Research Paper
The specics of homicide-suicide in France
Keltoume Larchet
a
,
*
, Aur´
elien Langlade
a
, Mathieu Lacambre
b
a
Department of Criminological Research, Sub-Directorate of Strategy and Local Streering, National Directorate of the Judicial Police, 101 Rue des Trois Fontanot, 92000,
Nanterre, France
b
Department of Emergency Psychiatry and Acute Care, CHU, Hopital Lapeyronie, INSERM Unit 1061, University of Montpellier, 371 Avenue du Doyen Gaston Giraud,
34090, Montpellier, France
ARTICLE INFO
Handling Editor: Wilma Duijst
Keywords:
Homicide-suicides
Comparative analyses
Couple homicides
Family homicides
ABSTRACT
Background: Homicide and suicide are two causes of violent death. This study focuses on situations in which these
two types of violent death occur at the same time, known as homicide-suicide.
Aim: The aim of this study is to review the circumstances surrounding homicide-suicides as well as the features of
the victims and perpetrators in France at a recent period. The article characterizes homicide-suicides by
comparing them with simple homicides in order to determine whether they exhibit distinctive characteristics and
signicant links.
Methodology: The data analyzed are based on detailed police operational information collected about 1622 ho-
micides that occurred in France in 2019 and 2020. After presenting the characteristics of all homicide-suicides,
this article specically compares those that take place within the family, whether they result in the suicide of the
perpetrator or not, using bivariate tests (chi-square). The tests are based on a distinction between domestic
homicide-suicides within a couple and between other family members.
Results: Analysis shows that 7 % of them were followed by the suicide of the perpetrator (203 cases). Almost all of
them take place within family (91 %), and most often within the couple (60 %). Apart from couple congura-
tions, familial homicide-suicides target the children (21 %) or parents (5 %) of the perpetrators. The perpetrators
are mainly men, while most of the victims are women. Analysis revealed signicant links between certain
variables and suicide (or attempted suicide by the alleged perpetrator) in the context of couple homicide: the
modus operandi, the spatio-temporal setting, the gender and average age of the victim and history of domestic
violence suffered by the victim, as well as all the variables relating to the alleged perpetrator (age, gender, police
and psychiatric history, alcohol consumption, etc.). Among other family members, these variables do not have a
signicant inuence on whether the perpetrator commits suicide following the homicide. Perpetrators of do-
mestic homicide-suicides are less likely to be under the inuence of alcohol (8 %), to be known to the police (19
%), and to have a psychiatric history (11 %) than those who do not commit suicide following the homicide
(respectively 41 %, 54 %, 22 %). There is also less record of domestic violence when the perpetrator commits
suicide (20 %) than when they do not (48 %).
Conclusions: Homicide-suicides are to some extent similar to simple couple homicides. Women are therefore
overrepresented among domestic homicide victims; likewise, they are also overrepresented among the victims of
homicide-suicides, which are mostly committed in family circumstances. The most signicant risk factors are the
presence of rearms in the household and history of domestic violence within the couple.
Recommendations: The results suggest two areas for action: the screening (interpersonal conicts, alcohol,
depression, domestic violence, presence of a rearm) and the prevention of intra-family homicides. Prevention
should focus on screening specic elements: the presence of a rearm, domestic violence, interpersonal conicts,
depression, and alcohol. Prevention could also take place with elderly people at the end of their lives.
* Corresponding author.
E-mail addresses: keltoume.larchet@interieur.gouv.fr (K. Larchet), aurelien.langlade@interieur.gouv.fr (A. Langlade), m-lacambre@chu-montpellier.fr
(M. Lacambre).
Contents lists available at ScienceDirect
Journal of Forensic and Legal Medicine
journal homepage: www.elsevier.com/locate/yjflm
https://doi.org/10.1016/j.jm.2023.102596
Received 20 February 2023; Received in revised form 21 September 2023; Accepted 3 October 2023

Journal of Forensic and Legal Medicine 100 (2023) 102596
2
1. Introduction
Suicide and intentional homicide are both unnatural and violent
deaths that also share the fact that they have been intentionally pro-
voked, either against the perpetrator (self-aggressive behavior) or
against a third-party victim (hetero-aggression). Homicide-suicides are
dened as intentional homicides followed by the suicide or attempted
suicide of the perpetrator within a short period of time, between the
moment the crime was committed and the following 24 h. International
literature provides various denitions of this timeframe, which can
extend to several days
1
or even months.
2
Although the scientic literature on the subject is abundant, it is
mainly based on clinical case studies with a psychopathological orien-
tation or the analysis based on reports of medico-legal autopsies.
3
Epidemiological data on suicide are abundant in both France and
abroad, but homicide-suicide (HS) is much less documented. Several
studies have compared the characteristics of homicide-suicides with
those of suicides, but also with those of simple homicides. These studies
have shown clear differences between these different acts of violence. In
France, as in other European countries, the majority of
homicide-suicides take place within the family and more specically,
within a couple relationship. Given this over-representation of family
contexts in homicide-suicides, it is useful to take a comparative
perspective between these family homicide-suicides and those that are
not followed by suicide.
Firstly, the aim of this study is to ll the lack of data about homicide-
suicides. Secondly, the aim of this study is to improve knowledge and
understanding of homicide-suicides in France by using recent and
detailed national data. To this end, French homicide-suicides will be
described (circumstances, victims, perpetrators) and compared with
homicides not followed by suicide (simple homicides) using bivariate
tests (chi-square). In order to quantify and describe the homicide-suicide
phenomenon in France, the aim is to determine whether there are sig-
nicant differences between domestic homicides followed by suicide
and those that are not. The study also aims to compare these French
homicide-suicides with those committed in other countries. This
approach could improve our knowledge of this doubly lethal phenom-
enon and help to guide preventive measures.
1.1. The concepts of homicide, suicide, and homicide-suicide
The concept of voluntary homicide is not dened in the French Penal
Code. In common usage, it refers to the act of deliberately endangering
the life of another person to the point of killing them. The United Na-
tions Ofce on Drugs and Crime (UNODC) proposes a uniform and in-
ternational denition based on three criteria: an objective element (the
death of a person caused by another person), a legal element (the un-
lawfulness of the act), and a subjective element (the perpetrator’s
intention to kill or seriously injure the victim).
4
Based on this characterization, suicide has similarities with inten-
tional homicide. It also results in an unnatural death although it is not
illegal and is, above all, self-administered. In his seminal study of sui-
cide, ´
Emile Durkheim provides the following denition of suicide: "any
death that results directly or indirectly from a positive or negative act
performed by the victim him/herself and which he/she knew would
produce this result".
5
Homicide-suicide lies at the intersection of the two concepts dened
above. When these two acts, with close ties to death,
6
occur at the same
time, it is then a homicide-suicide.
7
Part of the literature includes in
homicide-suicide the fact that the homicide can be attempted and not
committed. In other words, if the perpetrator attempts to kill one or
more victims but fails to do so and then commits suicide (whether or not
he succeeds), it is also considered homicide-suicide. Based on the
mobilized material, which does not include attempted homicides, this
article only counts committed homicides, but includes attempted sui-
cides and committed suicides. For the sake of simplicity, the expression
"homicide followed by the suicide of its alleged perpetrator" also in-
cludes attempted suicide, although this is not always specied. The
homicide and the suicide or attempted suicide studied in this article
were committed within a very short period of time, most often imme-
diately or at most a few hours after the commission of the intentional
homicide.
1.2. Suicide and homicide mortality in France
On the basis of data collected by the Epidemiology Centre on the
medical causes of death in France (C´
epiDc, Inserm laboratory), the
National Suicide Observatory noted a downward trend in the rate of
death by suicide per 100,000 inhabitants in France since 1980, although
this rate remains high in comparison with other European countries.
This downward trend follows an increase in this rate between the 1960s
and the 1980s.
8
In 2016, the suicide death rate was 14 per 100,000
inhabitants, while suicides accounted for 1.5 % of deaths from all cau-
ses.
9
The level of detail of these statistics does not make it possible to
identify, among these suicides, those that follow a criminal act.
With regard to homicides, there are several sources of administrative
data recording the number of cases (health, police, and judicial sources),
but these have major shortcomings. There is a particularly low level of
detail and a clear overestimation or underestimation of the total number
of homicides.
10–12
Moreover, these data sources are not detailed enough
to identify homicide-suicides among homicides samples or among sui-
cides samples. In short, despite an increasing number of studies on the
topic, putting homicide data into perspective still comes up against a
number of practical obstacles, related to mismatches in denitions,
sources, institutional expertise, and material constraints.
13
2. Methods
Due to the above-mentioned shortcomings, particularly in terms of
detail, the data used in this study come from operational information
collected by a French judicial police service. The data presented in this
article are the rst extensive and detailed sample of homicides-suicide in
France over a recent period. The contribution of the data collected lies in
the fact that it covers a large number of detailed variables on homicides-
suicides: circumstances (place, time, modus operandi, etc.), character-
istics of the victims and the perpetrators (age, gender, police and psy-
chiatric history, substance inuence etc.) These data include detailed
information on 1622 homicide cases committed in 2019 and 2020,
resulting in 1689 victims and 1798 defendants. The homicide rate in
France is thus 1.3 per 100,000 inhabitants. In this article, a perpetrator is
a person who has been questioned by a judicial police ofcer and against
whom there is serious and consistent evidence of guilt attesting to his or
her participation in the commission of the offence. It is important to note
that this is a police operational concept that does not correspond to a
status dened by the French Code of Criminal Procedure, but to a situ-
ation that precedes any criminal decision on whether to prosecute and, a
fortiori, on guilt. They are therefore perpetrators in the police sense and
not in the legal sense, since they have not necessarily been convicted.
The methodological perspective adopted in this article is compara-
tive. In order to determine whether there is a statistical relationship
between the fact that the alleged perpetrator commits suicide or at-
tempts to do so and some of the characteristics of these homicides, we
use, in addition to cross-tabulations, the Chi-square test associated with
Phi and Cramer’s V.
The Chi-square test makes it possible to access the independence
between two qualitative variables, or in other words, to determine
whether there is a signicant relationship of dependence between these
variables. Phi and Cramer’s V, which are used in conjunction with the
Chi-square test, are indicators of strength that measure the impact of the
relationship between two variables. They vary between 0 and 1, and the
closer they are to 1, the stronger the relationship between the two
variables. Analyses were carried out using SPSS software, version 27.
K. Larchet et al.

Journal of Forensic and Legal Medicine 100 (2023) 102596
3
These statistical tests are not carried out between homicide-suicides
and all simple homicides, but with intra-family homicides. This meth-
odological choice is explained by the fact that the vast majority of
homicide-suicides take place within the family. In order to rene the
analysis, it seems appropriate to take this over-representation into
account.
3. Results
During the period studied (2019 and 2020) in France, 7 % cases of
voluntary homicide were followed by the suicide of the perpetrator (126
victims). Perpetrators attempted suicide in 5 % of cases (77 victims).
Homicide-suicides thus represent 12 % of all voluntary homicides in
France, which correspond to a rate of 0.15 per 100,000 inhabitants (203
victims).
3.1. The different types of homicide-suicides
Homicide-suicide can be approached in terms of the relationship
between the victim and the presumed perpetrator. This approach con-
siders homicide-suicides in a dichotomous way, distinguishing between
events involving protagonists who belong to the same family unit and
those who do not (Fig. 1).
Homicide-suicides are mainly committed within the family sphere
(91 %), which implies a close relationship between the victim and the
presumed perpetrator (couple, descendants, siblings, by marriage). The
rate of homicide-suicides within the family is 0.14 per 100,000 in-
habitants. In detail, most homicide-suicides were committed within a
couple (60 %). In accordance with many studies and reports on the
subject, the concept of homicide within a couple concerns both actual
life partners and those who are no longer living together, married or not,
and occasional or not. According to the data presented, most French
couple homicide-suicides (78 %) involved partners at the time of the
incident. Among homicide-suicides committed within a couple, two
main circumstances stand out (Table 1): unaccepted separations (29 %)
and the ageing, or even the illness, of the victim or both members of the
couple (28 %).
In family homicide-suicides committed outside of a couple, several
family relationships can be involved. In most cases (21 %), the homi-
cides were licides, which means that parents kill their children
(whether the child is a minor or an adult). Among these licides, 18 % of
the victims are minors and 3 % are adults (Fig. 1). Filicide-suicides of
minors mainly involve victims between the ages of 1 and 15 years (they
represent 87 % of minor victims killed by parents who commit suicide)
(Table 2). The concept of neonaticide refers to situations in which the
victim is less than 24 h old. No neonaticide were recorded among
licide-suicides. Filicides not followed by suicide mainly involve victims
under the age of one (neonaticides and infanticides), which account for
62 % of simple licides. Filicides of minors are much more likely to be
committed by fathers when they result in suicide (76 %) than when the
perpetrator does not commit suicide (45 %). Women who kill their
children are less likely to commit suicide than fathers who do so. While
16 % of licides committed by mothers are followed by suicide, 43 % of
those committed by fathers are followed by suicide.
By going into more detail of these juvenile licides-suicides, fathers
and mothers do not kill the same victims, whether they commit suicide
or not. Fathers who commit homicide-suicide target children aged 1–14
years (68 %) and 15–18 years (8 %), whereas the children killed by their
mothers who commit homicide-suicide are younger, since 5 % are less
than one (what is known as infanticides) and nearly a quarter (19 %)
aged 1–14 years. In comparison, in licides of minors that do not result
in parental suicide, the victims are younger: no victims over the age of
15 were recorded.
In 5 % of cases, children killed their mother (matricide) or father
(parricide) before committing or attempting suicide (Fig. 1). Matricide
and parricide-suicides are almost exclusively committed by males.
Homicide-suicides against ascendants are not very frequent (5 % of
cases). Both mothers (4 %) and fathers (1 %) are involved. Almost all
these cases of matricide and parricide are committed by men, who have
an average age of 46 years (Table 3). Perpetrators of matricide or
parricide-suicides were slightly more likely to have a psychiatric history
than the perpetrators who killed their parents without subsequently
committing suicide. In facts, people who have killed one of their parents
and then committed suicide had a psychiatric history in 55% of cases.
Those who did not commit suicide following the homicide had a psy-
chiatric history in 43% of cases.
The remaining 5 % of family homicide-suicides outside the couple
involve wider family relationships (2nd, 3rd or 4th degree relatives).
There are few homicide-suicides committed outside the family
sphere (9 %). These facts mainly target victims in the course of their
professional activity (3 %). Other homicide-suicides are the result of
violence or homicide in the couple that led to collateral victims (2 %),
either because they were present at the scene, or because the partner or
ex-partner killed the new partner of his ex-partner, which is called
"adversary" homicide-suicide.
14
The remainder of extra-familial homi-
cide-suicides are neighborhood disputes (2 %). No extra-familial homi-
cide-suicides committed in mass murder contexts were recorded in
France during the period studied.
3.2. The suicide of the presumed perpetrator is signicantly related to the
space-time context when the homicide takes place within the couple
Homicide-suicides committed within a couple are more likely to take
place in rural areas (less than 2000 inhabitants), in metropolitan France
and during the day than those that are not followed by the suicide of the
presumed perpetrator. These three parameters, and mainly the time of
completion, appear to be signicantly related to the fact that the
perpetrator ends his or her life following the couple homicide (Table 4).
The victim’s home appears to be the preferred location for domestic
homicide-suicides since more than three quarters of them take place
there. Whether the crime took place in the victim’s home or was
committed during the week or at the weekend has no bearing on
whether the alleged perpetrator commits suicide. In fact, the differences
between homicides followed by suicide and those not followed by sui-
cide are rather insignicant on these spatio-temporal variables.
Although homicide-suicides between other family members occurred
more often at night, on weekends, and in the victim’s home than those
not followed by the perpetrator’s suicide, bivariate analyses did not
identify a signicant relationship between these factors and the perpe-
trator’s suicide.
Finally, the rst containment related to the Covid-19 pandemic
implemented in France between March and May 2020 does not seem to
have had a signicant impact on the suicide of domestic homicide
perpetrators.
1
3.3. Firearms are the preferred method in domestic homicide-suicides
The modus operandi of homicides followed or not by suicide reveals
signicant differences, especially when they are committed within a
couple. Indeed, in these cases, both the method used and the number of
victims per incident are signicantly related to the suicide of the
perpetrator (Table 5).
Domestic homicides are more likely to be committed with a rearm
when the perpetrator commits suicide than when he does not. More than
half of domestic homicide-suicides are carried out with a rearm (53 %),
1
Bivariate analyses compared homicide-suicides committed in 2020 during
total containment (March 18-May 10, 2020) with those committed during the
same period in 2019. The results suggest that there is no signicant relation-
ship. In other words, containment does not appear to have impacted these types
of homicides.
K. Larchet et al.

Journal of Forensic and Legal Medicine 100 (2023) 102596
4
as opposed to 10 % of other domestic homicides (Table 5). These are
mainly long guns (71 %) and single lethal impact (51 %). The use of
rearms was also more frequent in homicide-suicides committed be-
tween other family members (22%) than in family homicides that did
not involve suicide.
Like simple homicides, most homicide-suicides involve only one
victim. In some situations, a single perpetrator kills several victims,
resulting in multiple homicide-suicide victims committed by a single
perpetrator. Domestic homicide-suicides, whether committed within a
couple or not, are more likely to involve multiple victims than when the
perpetrator does not commit suicide. More than a quarter of non-couple
family homicide-suicides involve multiple victims (26 %), compared
with 4% of non-suicide homicides.
3.4. An overrepresentation of female victims compared to simple
homicides
Women are over-represented among suicide-homicide victims. In
couple suicide-homicides, women represent 92 % of the victims
Fig. 1. Distribution of homicide-suicides according to the relationship between the protagonists.
Source: processing of operational data, French judicial police service.Field: France, 2019–2020.
Table 1
Triggers factors in couple homicides, with and without suicide of the
perpetrator.
Homicides in the couple
Homicides-
suicides
Homicides not followed by a
suicide
Unaccepted Separation 29 % 14 %
Sickness and/or illness 28 % 2 %
Argument 13 % 42 %
Jealousy 11 % 15 %
Depression 6 % 0 %
Other reasons 13 % 27 %
Previous violence
1
in the
couple
20 % 48 %
1 - Physical, sexual or psychological violence.
Table 2
Share of licides, with or without suicide, by sex of perpetrator and age group of victim.
Homicides of minors by their parents (Filicides)
Filicides-suicides Filicides not followed by a suicide
Committed by
men
Committed by
women
Total licides-
suicides
Committed by
men
Committed by
women
Total simple
licides
Victims aged less than 24 h old (Neonaticide) None None None None 16 % 16 %
Victims aged more than 24 h and less than 1 year
old (Infanticide)
None 5 % 5 % 27 % 19 % 46 %
Victims aged between 1 and 14 years old 68 % 19 % 87 % 18 % 20 % 38 %
Victims aged between 15 and 18 years old 8 % None 8 % None None None
Total 76 % 24 % 100 % 45 % 55 % 100 %
Table 3
Characteristics of perpetrators of matricides and parricides, with and without
suicide.
Matricides and parricides
Homicides-
suicides
Homicides not followed by a
suicide
Proportion of male perpetrators 91 % 87 %
Proportion of female
perpetrators
9 % 13 %
Average age of perpetrators 46 years old 38 years old
Psychiatric antecedents of the
perpetrator
55 % 43 %
K. Larchet et al.

Journal of Forensic and Legal Medicine 100 (2023) 102596
5
(compared to 76 % when the alleged perpetrator does not commit sui-
cide or does not attempt to do so) (Table 6). This can be explained by the
fact that women tend to be over-represented among the victims of do-
mestic homicide, and especially in couple homicides not followed by
suicide. It can be noted that women account for 30% of the victims of all
intentional homicides (30 %). A more detailed analysis shows that the
proportion of female victims can vary signicantly depending on the
type of homicide. The proportion of women among victims of homicide-
suicide among other family members is lower (60 %), but still higher
than when the accused does not commit suicide (37 %).
Victims of intimate partner homicide-suicides are older (57 years
old) than victims of simple couple homicides (45 years old). This age
structure must be put into perspective with the signicant proportion of
couple homicide-suicides that take place in a context of ageing partners
(28%).
There is less history of domestic violence in homicide cases involving
the suicide of the perpetrator that in other couple homicides. The same
observation applies to homicide-suicides committed in the context of
ageing protagonists. While almost half (48 %) of the victims of simple
spousal homicides had already been subjected to violence (physical,
sexual and/or psychological) by the perpetrator, 20 % of the victims of
spousal homicide-suicide had.
In the family context, more than half of the homicide-suicides
perpetrated outside the couple involve minor victims (53 %), mainly
killed by a parent. Finally, it should be noted that in cases of couple
homicides, the victim’s prole and the commission of domestic violence
in the past have a signicant inuence on whether or not the perpetrator
attempted suicide.
3.5. Perpetrators of homicide-suicides are older than in simple homicides
The sociodemographic characteristics of presumed homicide perpe-
trators (gender, age, professional activity) as well as their use of psy-
choactive substances (alcohol and their possible police record seem to be
linked to whether or not they commit suicide committing following the
homicide).
According to the data mobilized in this study, men are over-
represented among the perpetrators of simple homicide, and especially
Table 4
Circumstances of domestic homicides, with and without suicide of the perpetrator.
Familial homicides
In the couple Between other members of the family
Homicides-
suicides
Homicides not followed by suicide Chi
2
(Phi) Homicides-
suicides
Homicides not followed by suicide Chi
2
(Phi)
Metropolis 96 % 89 % 4464** (−0,116) 95 % 90 % 0,840 (−0,059)
Rural 27 % 17 % 5064** (−0,123) 18 % 21 % 0,171 (0,027)
Victim’s home 78 % 74 % 0,691 (0,046) 79 % 71 % 0,924 (0,062)
Weekend 25 % 30 % 0,879 (−0,0151) 47 % 33 % 2747 (0,107)
Night 38 % 65 % 19,876** (−0,257) 59 % 45 % 1777 (0,095)
Lockdown
1
6 % 7 % 0,081 (0,016) 13 % 9 % 0,649 (−0,052)
**P <0.05 P <0.001***.
1 - This is the rst national lockdown (total lockdown) that occurred between March 17th and May 11th, 2020 in France.
Table 5
Modus operandi of domestic homicides, with and without suicide of the perpetrator.
Familial homicides
In the couple Between other members of the family
Homicides-
suicides
Homicides not followed by
suicide
Chi
2
(Phi) Homicides-
suicides
Homicides not followed by
suicide
Chi
2
(Phi)
Firearm use 53 % 10 % 71,457***
(0,469)
22 % 12 % 3268 (0,119)
Use of bladed weapon 23 % 42 % 31 % 26 %
Other methods 24 % 48 % 47 % 61 %
Several victims by the same
cause
9 % 2 % 7694*** (0,152) 26 % 4 % 20,827***
(0,295)
**P <0.05 P <0.001***.
Table 6
Characteristics of victims of domestic homicides, with and without suicide of the perpetrator.
Familial homicides
In the couple Between other members of the family
Homicides-
suicides
Homicides not followed by
suicide
Chi
2
(Phi) Homicides-
suicides
Homicides not followed by
suicide
Chi
2
(Phi)
Proportion of female victims 92 % 76 % 13,311***
(−0,200)
60 % 37 % 6729***
(−0,169)
Proportion of minor victims – – – 53 % 39 % 2340 (−0,099)
Average age of victims
1
57 years old 45 years old Eta: 0,307*** 34 years old 36 years old Eta:0,029
History of violence experienced by
the victim
20 % 48 % 24,867***
(−0,281)
– – –
**P <0,05 P <0,001***.
1 - A test of difference of means (Student’s T) was carried out to check whether the difference in the age of the victims is signicant. The Eta coefcient, like Cramer’s
Phi and V coefcients, is used to determine the strength of the relationship.
K. Larchet et al.

Journal of Forensic and Legal Medicine 100 (2023) 102596
6
when they occur within the family unit. This is also the case when the
perpetrator commits suicide: 93 % of the alleged perpetrators of couple
homicide-suicides are men. When domestic homicides occur between
other family members, 74 % of the perpetrators who end their own lives
are male. The proportion is similar when they do not commit suicide (71
%) (Table 7).
Perpetrators of intra-family homicide-suicide, both within the couple
(61 years) and between other family members (47 years), are older than
the other perpetrators (49 and 39 years respectively). There is a
discrepancy between the respective average ages of the victims and the
alleged perpetrators of homicide-suicides: perpetrators tend to be older
than their victims.
With regard to the criminal record of the presumed perpetrators, we
note that in the context of a couple, they are much less likely to be
known to the police when they commit suicide (19 %) than when they
do not (54 %). There is no such difference for homicides between other
family members (46 % vs 48 %).
The association between alcohol consumption and couple homicide-
suicide is signicant and moderate. However, the direction of the rela-
tionship is not intuitive: the proportion of alcoholic perpetrators of
couple homicide-suicide is much lower (8 %) than in simple homicide
(41 %).
Data collected in France show that the proportion of suicides
committed by an intimate partner with a psychiatric history (11 %) is
half that of non-suicides (22 %). These bivariate analyses also show the
psychiatric history of the perpetrator has a signicant inuence on
whether or not he commits suicide after the homicide. When the pro-
tagonists are members of the rest of the family, perpetrators who commit
suicide are much more likely to have a psychiatric history (60 %) than
those who do not commit suicide (39 %). However, according to the chi-
square test, there is no signicant link between suicide and the perpe-
trator’s psychiatric history in these cases. These cases are mainly
matricide, parricide and licide.
In summary, the prole of those accused of homicide-suicide
committed between other family members is more similar to that of
simple homicides than to that of couple homicide-suicides. For domestic
homicides between other family members, only the average age of the
perpetrators has a signicant inuence on whether they commit suicide.
4. Discussion
The results show that the French homicide-suicide rate is 0,15 per
100,000 inhabitants. By way of comparison, epidemiological data show
that homicide-suicide rates vary between 0.01 and 1.33 per 100,000
inhabitants depending on the country and the time period.
15
These data
cover periods from the 1950s to the 2000s, refer to different sources
(local or national data) and are based on heterogeneous materials (case
studies, monographs, health statistics, etc.). Finally, it should be noted
that the incidence of homicide-suicide has remained constant for several
decades in industrialized countries (around 0.2 and 0.3).
16,17
In the light
of this information, the rate of homicide-suicide in France in 2019 and
2020 appears to be low in comparison with other industrialized coun-
tries. Although the incidence of homicide-suicides varies from country
to country, the phenomenon can be described as rare but constant
16,17
or
even decreasing.
18,19
Some researchers postulate that homicide and
homicide-suicide rates move in different directions as the higher the
former, the lower the latter.
15
At the scale of the data mobilized in this
article (years 2019 and 2020), it is not possible to carry out an analysis
of homicide-suicides and homicides over time. However, it is possible to
note that the proportion of homicide-suicides among all homicides was
similar in both year (from 13 % to 11 %). The homicide-suicide rate
decrease slightly in 2020 (from 0.16 to 0.14 per 100,000 inhabitant).
Following the approach taken in this article, it would be interesting in
future studies, to analyze homicide-suicides over a longer period, but
also to compare the evolution of homicide-suicides with that of couple
homicides.
Homicide-suicides are extensively documented in the national and
international literature, but the data come from sources that are so
heterogeneous that comparisons can encounter temporal, geographical
and methodological obstacles. Due to the rarity of the phenomenon data
are scarce and, above all, tend to cover very localized scales and areas,
and are often lacking in detail. As Harper and Voigt
20
pointed out, one of
the methodological difculties of homicide-suicide (as compared to
intentional homicide alone) is that the main protagonists are both dead.
The subject is therefore approached from a variety of materials, ranging
from law enforcement reports, health sources, press articles, interviews
with relatives of the protagonists to specic psychological case analyses.
This means that data relating to the same time period, the same scope or
even the same denition cannot be put into perspective. The material
mobilized in this article is based on operational data and has the
advantage of being exhaustive over a given period and scope.
In France, the only analyses of homicide-suicides based on an
exhaustive quantitative census of a dened area cover the period 1991
to 1996 and were limited to Paris and the inner suburbs. Using data
collected from the Paris Forensic Institute, Lecomte and Fornes char-
acterized homicide-suicides committed in Paris and the Paris region
Table 7
Characteristics of perpetrators of domestic homicides, with and without suicide.
Familial homicides
In the couple Between other members of the family
Homicides-
suicides
Homicides not followed by
suicide
Chi
2
(Phi) Homicides-
suicides
Homicides not followed by
suicide
Chi
2
(Phi)
Male perpetrators 93 % 78 % 13,398***
(0,200)
74 % 71 % 0,144 (0,025)
Average age of perpetrators
1
61 years old 49 years old Eta: 0,401*** 47 years old 39 years old Eta:
0,280****
Unemployed 18 % 42 % 20,076***
(−0,246)
43 % 52 % 0,710
(−0,058)
With a police record 19 % 54 % 39,317***
(−0,344)
46 % 48 % 0,049
(−0,016)
Perpetrator under the inuence of
alcohol
2
8 % 41 % 39,338***
(−0,347)
– – –
Psychiatric antecedents of the
perpetrator
11 % 22 % 5386** (−0,141) 60 % 39 % 2462 (0,149)
Note **P <0,05 P <0,001***.
1 - A test of difference of means (Student’s T) was carried out to check whether the difference in the age of the victims is signicant. The Eta coefcient, like Cramer’s
Phi and V coefcients, is used to determine the strength of the relationship.
2 - Due to small numbers, bivariate analyses could not be performed for family homicide-suicides (excluding couples) committed by an alcoholic perpetrator.
Note for the reader: 93 % of the perpetrators of homicide-suicides in couples are men. 78 % of the perpetrators of simple homicides in couples are men.
K. Larchet et al.

Journal of Forensic and Legal Medicine 100 (2023) 102596
7
between 1991 and 1996.
21
According to this monograph that studied 46
HS cases involving 133 victims, most homicide-suicide share some
specic features: the perpetrators are mostly men, who use a weapon
and act in their own homes, in a family context. International studies
show a similar picture of homicide-suicide. The data presented in this
article point in the same direction: mostly male perpetrators, in a family
context, at home and with the use of a weapon.
Homicide-suicide is not a homogeneous phenomenon.
1
In order to
consider this aspect, it is necessary to analyze it through the prism of the
inter-acquaintance between the victim and the perpetrator, as suggested
by numerous studies, in accordance with the classication established
by Marzuk et al..
17
This basic classication distinguishes three types of
homicide-suicides: couple homicide-suicides (uxoricide-suicides), fam-
ily homicide-suicides outside the couple including one or more children
of the perpetrator (licide-suicides), the mother of the perpetrator
(matricide-suicides), the father of the perpetrator (parricide-suicides),
and extrafamily homicide-suicides.
22–24
This distinction, used by almost
all international research is directly relevant to the data presented in this
article.
The present study shows that most homicide-suicides are domestic,
and more specically take place within the couple. The rate of domestic
homicide-suicide per 100,000 inhabitants in France is 0.14, which is
higher than in some European countries, where it varies from 0.04 (in
Poland and Spain) to 0.16 (in Finland). In comparison, these European
homicide-suicide rates are lower than those found in other countries,
particularly in the United States, where the rate is estimated to be be-
tween 0.2 and 0.3 per 100,000 inhabitants
17
and as high as 0.7 per 100,
000 inhabitants in Florida based on Florida-specic data between 1988
and 1994.
15
According to the data presented in this article, the pro-
portion of couple homicide-suicides among all homicide-suicides is 60 %
in France. As a comparison, this proportion is 60 % in the Netherlands,
68 % in Poland, and 85 % in Spain.
23
The fact that six out of ten homicide-suicides in France take place
within a couple appears to be consistent with the annual census of vi-
olent deaths within couples carried out by the Delegation for Victims
(DAV) attached to the French Ministry of the Interior. The DAV notes
that in France in 2020, 40 % of homicides within couples resulted in a
suicide or attempted suicide by the presumed perpetrators, and that this
proportion is stable each year (D´
el´
egation aux victimes, 2020). The
portrait of homicide-suicides is therefore very similar to that of domestic
homicides (over-representation of female victims, male perpetrators,
older than the rest of the victims and perpetrators, mostly committed in
the victim’s home). The literature also conrms this analysis.
23,25
The
overrepresentation of female victims compared to single homicides is in
direct agreement with EHSS data that three-quarters of family
homicide-suicide victims are female.
23
This overrepresentation of
women among the victims and men among the perpetrators has led
many researchers to adopt a reading of the subject inherited from
feminist criminology, according to which homicide-suicide, like couple
homicide, should be viewed through the prism of gendered relations of
domination.
20
Perpetrators of homicide-suicides in the couple tend to be slightly
older (61 years old) than their victims (57 years old), as established in
the scientic literature.
26
The higher average age of couple
homicide-suicides is related to the proportion of homicide-suicides
induced by the aging of the protagonists (28 %). This appears to be
consistent with the lower proportion of perpetrators known to the police
(19 %), which is also consistent with the literature.
25
The proportion of
French homicide-suicides in couples because of the ageing or the illness
of the victim or both members of the couple is similar in the
United-States (25 %).
27
In these situations of "violent acting out during
the aging of the couple," it is common, but not systematic, that the victim
and alleged perpetrator have agreed to die jointly.
28
These suicide pacts,
dened as "voluntary death after prior mutual agreement of at least two
people",
29
are formalized by a farewell letter. They are also called
"homicide-suicide pacts" to distinguish them from congurations in
which neither of the two protagonists causes the death of the other, but
each kills himself, i.e., it is indeed a double suicide.
30
In these situations
of aging protagonists, the question of the link between
homicide-suicides and simple suicides, and even of the status to be
accorded to the act of suicide, arises more strongly, particularly because
age is considered a risk factor for suicide.
8,31,32
According to the data
collected, there is less previous violence in the couple when the perpe-
trator commits suicide (20%) than when he does not (48%). This is even
more noticeable when the homicide is committed in the context of
illness or advanced age of the protagonists, which has generally not been
preceded by violence in the couple. A similar proportion of victims of
domestic violence (25 %) are found in the literature.
27
This points to a
specicity of couple homicides due to the aging or illness of the pro-
tagonists. Most of these cases are followed by the suicide of the perpe-
trator. However, those couple homicides differ signicantly from those
committed for other reasons (argument, jealousy, unaccepted separation
…). In fact, according to the data presented in the results section, in
cases of ageing or illness, the perpetrators are less likely to be under the
inuence of alcohol, to be known to the police and to have a psychiatric
history when they commit suicide. They are also less likely to have a
history of domestic violence.
The victim’s home appears to be the preferred location for French
domestic homicide-suicides since more than three quarters of them take
place there, in line with numerous studies on the issue.
26
The second most represented type of homicide-suicide outside the
couple in France is licide, which is consistent with the ndings of
numerous studies.
17,18,20,27,33
The scientic literature shows that the
determining parameters for understanding licide-suicides refer to the
gender of the perpetrator
22,34–36
No neonaticide (i.e., with victims under
24 h old) is recorded among licide-suicides, which is consistent with
research that neonaticides do not tend to be followed by the suicide of
the perpetrator.
37
The fact that fathers are more likely to kill older
children than mothers is consistent with other studies.
38
Existing liter-
ature indicate that the determining parameters for understanding
licide-suicides refer to the gender of the perpetrator.
22,34–36
Extra-familial homicide-suicides are relatively rare (9 %) in France,
as conrmed by the scientic literature since the 1960s.
33,39
No
extra-familial homicide-suicides committed in mass murder contexts
were recorded in France during the period studied. However, such
homicide-suicide are well documented in the literature.
33
This type of
homicide-suicide is a hot topic in the United States, given the upward
trend in mass killings in schools and universities since the 1990s. Studies
on the matter make connections to the access to rearms and the issue of
psychopathological disorders and social isolation of perpetrators.
23
Homicide-suicides almost never occur in a context where the victim
and alleged perpetrator do not know each other. In 99 % of homicide-
suicides, the victim and perpetrator knew each other. Homicide-
suicides, like simple homicides, therefore, appear to be close and even
domestic crimes. To rene the analysis, the present study conducted a
disjointed analysis of couple homicide-suicides and those that take place
between other family members.
The data presented in this article show a signicant and moderate
relationship between alcohol consumption and couple homicide-suicide.
Homicide-suicide perpetrators are less likely to be alcohol dependent
than non-suicide perpetrators. This point converges with existing liter-
ature according to which presumed perpetrators of homicide-suicide are
less likely to consume alcohol than the rest of the perpetrators of ho-
micide.
20,35,40
Notice that as with the rest of the international literature,
substance use variables can suffer from underestimation.
25
Homicide-suicide in France is very similar to the representation
given in the international literature, no matter the time period, although
some discrepancy have been found (particularly concerning the oper-
ating mode). Homicide-suicides appear to be mostly committed in a
private place, by a perpetrator close to the victim (these two parameters
give them a domestic dimension), committed with the use of weapons
(mainly rearms). Because of this predominant operating mode, the
K. Larchet et al.

Journal of Forensic and Legal Medicine 100 (2023) 102596
8
issue of access to rearms is sometimes coupled with the analysis of
homicide-suicides.
15,18
The study of the link between rearms and
violence goes beyond the scope of homicide-suicides. For instance, a
1997 study entitled “Violence against women” analyzed the inuence of
the availability of rearms in the home and the prevalence of lethal
violence, especially against women.
41
Some studies put into perspective
the restrictive nature of gun legislation as contributing to lower homi-
cide and suicide rates.
15
This factor encourages preventive actions to
focus on the use of legally-held rearms. In France, this mainly concerns
hunting weapons. The predominance of rearm use in domestic homi-
cides is found in numerous studies.
20,25,34
In Europe, according to Eu-
ropean Health Survey System (EHSS data), 48 % of domestic
homicide-suicides are committed by rearms.
23
According to the same
data, bladed weapons were used in 17 % of intra-familial homi-
cide-suicides,
23
slightly fewer than in France.
In most simple homicides, a perpetrator kills a single victim.
42
The
results presented show that the proportion of facts involving multiple
victims is higher when the perpetrator commits suicide than when he
does not. This proportion is even more important in familial
homicide-suicides committed outside the couple. Multiple-victim ho-
micides are also over-represented among homicide-suicides occurring
outside the family.
34
The literature related to suicide acts, whether they are accompanied
by a prior homicide, often raises the issue of the perpetrators’ mental
health. The presence of psychiatric pathologies, especially undiagnosed
ones, thus appears to be a suicide risk factor.
14
These ndings are
consistent with current published data on homicide-suicide: women are
most often victimized in the context of interpersonal conict involving
the use of a rearm by a male (ex)spouse (Liem, 2012
43
). The com-
parison between homicide followed by suicide and simple homicide
enhances recent descriptions of different subtypes of homicide followed
by suicide.
44
For example, over the study period, there were no
homicide-suicides involving infants (neonaticide), whereas in the group
of simple homicides, there were neonaticides and infanticides. These
simple neonaticides and infanticides were committed three times more
often by women than by men, and twice as often for licides. For simple
homicide, the protagonists are slightly younger (45 years old against 60
years old on average for homicide-suicide) and women are the main
victims in a context of history of domestic violence, psychiatric history,
alcohol consumption and previous convictions of the murderer.
Depression is a trigger for the act only in homicide-suicide situations and
not in simple homicide situations. However, it is more common for the
perpetrator of a homicide to have a psychiatric history when they do not
commit suicide. In our study, the depression item is poorly referenced
probably because it is considered in the data collection as a triggering
factor, whereas it is probably more a situation of acute distress that fa-
vors the act in a context of interpersonal couple conict.
It should be noted that when the victim is not the partner (or former
partner), the differences between homicide and homicide followed by
suicide are blurred. Thus, the couple is the main context of homicide-
suicide, raising questions about a probable pathology of the relation-
ship on the perpetrator’s side, a factor precipitating the swing towards
homicide-suicide rather than simple suicide.
45,46
Moreover, 40 % of
uxoricide perpetrators commit suicide or attempt while committing
partner homicide.
47
In order to improve the detection and prevention of
these morbid processes, further research is necessary, specifying, if
possible, the psychiatric antecedents (self-mutilation, violent suicide
attempts, personality disorders, mood disorders, addictions, treatments,
etc.) of all the protagonists. It should consider the type and length of the
relationship, the age difference, the level of education and nancial
resources, and the possible previous recourse to legal, health and social
services. Thus, the stress-vulnerability model of suicidal behavior
48
would be enriched by this research on murder at the edge of the self.
49
Like suicide, it seems obvious to us from these results that couple ho-
micide and homicide followed by suicide should benet from specic
preventive interventions by social, judicial, and health actors.
5. Limitations of the study
This study could be completed in several ways. First, from a meth-
odological standpoint, the size of the sample could be extended beyond
the two-year period. From a temporal point of view, it is important to
underline that the data studied partly relate to the lockdown linked to
the Covid-19 epidemic, which is a specic and unprecedented period. It
would be relevant to analyze its impact in detail.
Although rich, the data used suffer from the incompleteness of some
determinant variables such as the intoxication of the protagonists, with a
lack of details of the type of substance consumed. Similarly, more
detailed information on the type and length of the relationship between
the victims and the perpetrators would be of considerable added value.
A wider range of variables related to the motive for the homicide-suicide
would certainly have enriched the study. Furthermore, the data could
have been more detailed by providing information on the social back-
ground of the protagonists. With the exception of a few localized
studies,
19
the literature does not ll this gap. Furthermore, although a
variable allows us to determine whether the alleged perpetrators have a
psychiatric history, it does not provide any details on the latter, even
though this dimension is abundantly documented in the literature and
even represents a privileged entry point for understanding the
theme
40,50
.
51
; Finally, in the case of multiple victims (when a perpe-
trator kills several victims before committing suicide or attempting to do
so), it would have been even more interesting to consider the number of
victims per perpetrator in the tests performed.
A widely asked research question concerns the links between all or
part of the triad of homicide, suicide and homicide-suicide.
23,33
Dur-
kheim already set the premises of this question in his analysis of suicide,
devoting a chapter to the articulation between homicide and suicide,
5
as
did Marcel Mauss several years later.
52
Homicide-suicide is both
voluntary homicide and suicide. This raises the question of the link and
overlap between the two. Is the starting point suicidal or homicidal
intent
53
? Is the perpetrator’s intent primarily homicidal or suicidal? A
practical implication of this questioning is to introduce the notions of
primary or secondary victims of homicide-suicide.
54
It is not possible to
clarify this notion of intentionality without mobilizing the author’s
discourse.
The comparative perspective aimed at in this article was not oriented
towards a methodical comparison of suicide and homicide-suicide.
However, it does highlight the fact that homicide-suicides share
certain similarities with simple homicides, and more specically with
some of them.
6. Conclusion
This descriptive study, based on police data, sheds original and
complementary light on an infrequent but stable and dramatic phe-
nomenon: homicide followed by suicide. The comparison between
homicide-suicides and homicides not followed by suicide makes it
possible to specify the characteristics of the victim, the perpetrator, the
circumstances, and the context of this particular act, including the
murder of the perpetrator. The screening (interpersonal conicts,
alcohol, depression, domestic violence, presence of a rearm) and pre-
vention of intra-family homicides followed or not by suicides can be
improved considering these results. These results also encourage to
address the issue of the end of life and to consider ways of supporting
these situations, such as decriminalizing euthanasia or legalizing assis-
ted suicide.
Funding
We did not receive funding for this research.
This research did not receive any specic grant from funding
agencies in the public, commercial, or not-for-prot sectors.
K. Larchet et al.

Journal of Forensic and Legal Medicine 100 (2023) 102596
9
Additional data
No additional data is available.
Data statement
Due to the sensitive nature of the data collected in the study, survey
respondents were assured that the raw data would remain condential
and would not be shared.
Data not available/The data that was used is condential.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgments
We would like to thank Dr Wayne Bodkin, Jean-Baptiste Lacambre
and Coline Tessier for their careful review.
References
1. Rouchy E, Germanaud E, Garcia M, Michel G. Characteristics of homicide-suicide
offenders: a systematic review. Aggress Violent Behav. 2020;55. https://psycnet.apa.
org/doi/10.1016/j.avb.2020.101490.
2. Allen N. Homicide followed by suicide : Los Angeles, 1970-1979. Suicide Life-
Threatening Behav. 1983;13(3):155–165. https://doi.org/10.1111/j.1943-
278X.1983.tb00013.x.
3. Karnecki K, Wrocławski A, Pie´
sniak, et al. Homicide mechanisms, weapon types and
causes of death in the autopsy material of the Department of Forensic Medicine
Medical University of Gda´
nsk Poland in the years 2010–2019. J For Legal Med. 2023;
94. https://doi.org/10.1016/j.jm.2022.102481.
4. United Nations Ofce on Drugs and Crime. Global Study on Homicide. Homicide
trends, patterns and criminal justice response. Booklet. 2019;1.
5. Durkheim ´
E. Le Suicide. Presses universitaires de France; 2007. https://doi.org/
10.4000/lectures.473.
6. Guivarch J, Piercecchi-Marti M, Poinso F. Folie `
a deux and homicide: literature
review and study of a complex clinical case. Int J Law Psychiatr. 2018;61:30–39.
https://doi.org/10.1016/j.ijlp.2018.10.001.
7. Trichet Y, Dupont A-S. Logique des homicides dits altruistes. Clinique de
l’infanticide. Bull Psychol. 2011;514:347–357.
8. Anguis M, Surault P. L’
´
evolution des suicides sur longue p´
eriode : le rˆ
ole des effets d’ˆ
age,
de date et de g´
en´
eration. 2002 (drees, ´
Ed.) ´
Etudes et r´
esultats(185).
9. French National Suicide Observatory. Suicide : quels liens avec le travail et le chˆ
omage ?
Penser la pr´
evention et les syst`
emes d’information. 2020, 4e rapport.
10. Mucchielli L. Les homicides dans la France contemporaine (1970-2007) : ´
evolution,
g´
eographie et protagonistes. In: Mucchielli L, Spierenburg P, eds. Histoire de
l’homicide en Europe. La D´
ecouverte; 2009:133–164.
11. Scherr M, Langlade A. Les caract´
eristiques des homicides commis `
a Paris et en petite
couronne. Grand angle n◦35. ONDRP; 2014.
12. Langlade A, Delbecque V, Soullez C. La mesure de l’homicide en France. Cahiers de la
s´
ecurit´
e et de la justice. 2017;41:2017.
13. Liem M, Kivivuori J, Lehti M, Granath S, Sch¨
onberger H. Les homicides conjugaux en
Europe : r´
esultats provenant du European Homicide Monitor. Cahiers de la s´
ecurit´
e et de
la justice(41); 2017:134–146.
14. Chocard A-S. Approche psychopathologique du passage `
a l’acte homicide-suicide.
Imaginaire Inconscient. 2005;2(16):183–198. https://doi.org/10.3917/
imin.016.0183.
15. Large M, Smith G, Nielsen O. The epidemiology of homicide followed by suicide: a
systematic and quantitative review. Suicide Life-Threatening Behav. 2009;39:
294–306. https://doi.org/10.1521/suli.2009.39.3.294.
16. Coid J. The epidemiology of abnormal homicide and murder followed by suicide.
Psychol Med. 1983;13:855–860. https://doi.org/10.1017/s0033291700051576.
17. Marzuk P, Tardiff K, Hirsch C. The epidemiology of murder-suicide. J Am Med Dir
Assoc. 1992;267(23):3179–3183. https://doi.org/10.1001/
JAMA.1992.03480230071031.
18. Milroy C. Reasons for homicide and suicide in episodes of dyadic death in yorkshire
and humberside. Med Sci Law. 1995;35(3):213–217. https://doi.org/10.1177/
002580249503500307.
19. Kivivuori J, Lehti M. Homicide followed by suicide in Finland: trend and social
locus. J Scand Stud Criminol Crime Prev. 2003;4(2):223–236. https://doi.org/
10.1080/14043850310021972.
20. Harper D, Voigt L. Homicide followed by suicide: an integrated theoretical
perspective. Homicide Stud. 2007;11(4):295–318. https://doi.org/10.1177/
1088767907306993.
21. Lecomte D, Fornes P. Homicide followed by suicide: Paris and its suburbs, 1991-
1996. J Forensic Sci. 1998;43(4). https://doi.org/10.1520/JFS14303J.
22. Dubroca O. L’homicide-suicide : une revue de la litt´
erature. Th`
ese de m´
edecine; 2018.
23. Liem M, Oberwittler D. Homicide followed by suicide in europe. In: Dans M. Liem, &
W. Pridemore, Handbook of European Homicide Research. Springer; 2012:197–215.
https://doi.org/10.1007/978-1-4614-0466-8_12.
24. Liem M, Postulart M, Nieuwbeerta P. Homicide-suicide in The Netherlands: an
epidemiology. Homicide Stud. 2009;13(2):99–123. https://doi.org/10.1177/
1088767908330833.
25. Eliason S. Murder-suicide : a review of the recent literature. J Am Acad Psychiatr
Law. 2009;37(3):371–376.
26. Morali A, Baratta A. L’homicide-suicide. Caract´
eristiques cliniques et
´
epid´
emiologiques. Illustration `
a propos d’un cas clinique. L’ Inf Psychiatr. 2011;87
(4):321–328. https://doi.org/10.3917/inpsy.8704.0321.
27. Malphurs J, Cohen D. A newspaper surveillance study of homicide-suicide in the
United States. Am J Forensic Med Pathol. 2002;23(2):142–148. https://doi.org/
10.1097/00000433-200206000-00006.
28. Charazel-Brunel M. Passages `
a l’acte violents au cours du vieillissement du couple.
In: Dans M. (dir.) Charazel-Brunel, La suicide des personnes ˆ
ag´
ees. ´
Er`
es; 2014:157–194.
https://doi.org/10.3917/eres.chara.2014.01.
29. Lacambre M. Suicides homicides et pactes suicidaires. Dans P. Courtet, Suicide et
tentatives de suicide. 2010:206–210. https://doi.org/10.3917/lav.
court.2010.01.0206.
30. Le Bihan P, B´
en´
ezech M. Pactes suicidaires : une revue de litt´
erature. Ann Med-
Psychol. 2006;164(4):292–303. https://doi.org/10.1016/j.amp.2005.11.018.
31. Baudelot C, Establet R. Durkheim et le suicide. Presses universitaires de France; 2002.
32. Rey G. Donn´
ees ´
epid´
emiologiques sur les d´
ec`
es par suicide en France
m´
etropolitaine. Dans Suicide. Quels liens avec le travail et le chˆ
omage ? penser la
pr´
evention et les syst`
emes d’information. Observatoire national du suicide, 4`
e rapport.
2020;Fiche 2:114–120.
33. Liem M. Homicide followed by suicide : a review. Aggress Violent Behav. 2009;15,
163-151.
34. Chocard A-S, Juan F. Les meurtres-suicides. Revue de litt´
erature. Forensic. 2002;
(12):7–12.
35. Banks L, Crandall C, Sklar D, Bauer M. A comparison of intimate partner homicide to
intimate partner homicide-suicide: one hundred and twenty-four New Mexico cases.
Violence Against Women. 2008;14(9):1065–1078. https://doi.org/10.1177/
1077801208321983.
36. Flynn S, Swinson N, While D, et al. Homicide followed by suicide: a cross-sectional
study. J Forensic Psychiatr Psychol. 2009;20(2):306–321. https://doi.org/10.1080/
14789940802364369.
37. Friedman S, Resnick PJ. Parents Who Kill: Why They Do it. vol. 10. Psychiatric Times;
2009.
38. Hatters Friedman S, Hrouda DR, Holden CE, Noffsinger S, Resnick P. Filicide-
suicide: common factors in parents who kill their children and themselves. J Am
Acad Psychiatr Law. 2005;33(4):496–504.
39. West DJ. Murder Followed by Suicide. Cambridge: Harvard University Press; 1965.
40. Rosenbaum M. The role of depression in couples involved in murder-suicide and
homicide. Am J Psychiatr. 1990;147:1036–1039. https://doi.org/10.1176/
ajp.147.8.1036.
41. Bailey J, Kellermann A, Somes G, Banton J, Rivara F, Rushforth N. Risk factors for
violent death of women in the home. Arch Intern Med. 1997;157:777–782.
42. Liem M, Barber C, Markwalder N, Killias M, Nieuwbeerta P. Homicide-suicide and
other violent deaths: an international comparison. Forensic Sci Int. 2011;207(1-3):
70–76.
43. Oliffe J, Han C, Drummond M, Sta Maria E, Bottorff J, Creighton G. Men,
masculinities, and murder-suicide. Am J Men’s Health. 2015;9(6):473–485. https://
doi.org/10.1177/1557988314551359.
44. Jordan J, McNiel D. Homicide-suicide in the United States. J Clin Psychiatry. 2021;82
(2). https://doi.org/10.4088/jcp.20m13528.
45. Notredame C-´
E, Richard-Devantoy S, Lesage A, S´
eguin M. Peut-on distinguer
homicide-suicide et suicide par leurs facteurs de risque. Criminologie. 2018;51(2):
314–342. https://doi.org/10.7202/1054245ar.
46. Vandevoorde J, Estano N, Painset G. Homicide-suicide : revue clinique et hypoth`
eses
psychologiques. Encephale. 2017;43(4):382–393. https://doi.org/10.1016/j.
encep.2016.04.014.
47. Delegation for victims. Etude nationale sur les morts violentes au sein du couple en
2019. Minist`
ere de l’int´
erieur. https://www.interieur.gouv.fr/Archives/Archives-
des-communiques-de-presse/2020-communiques/Etude-nationale-relative-
aux-morts-violentes-au-sein-du-couple-en-2019; 2020.
48. Mann JJ. A brain-centric model of suicidal behavior. Am J Psychiatr. 2020;177(10):
902–916. https://doi.org/10.1176/appi.ajp.2020.20081224.
49. Lacambre M, Barbaray L. Du meurtre de soi au suicide de l’alter ego, les nouvelles
expressions du suicide. In: Dans R. Coutanceau, Violences aux Personnes. Paris:
Dunod; 2014:213–222. https://doi.org/10.3917/dunod.couta.2011.01.0415.
50. Richard-Devantoy S, Voyer M, Gohier B, Garr´
e J-B. La crise homicidaire pendant la
crise suicidaire ? Particularit´
es chez le sujet schizophr`
ene. Ann Med-Psychol. 2010;
168(1):62–68. hal.archives-ouvertes.fr/hal-00610674/dx.doi.org/doi:10.1016/j.
amp.2009.10.004.
51. Liem M, Koenraadt F. Familicide: a comparison with spousal and child homicide by
mentally disordered perpetrators. Crim Behav Ment Health. 2008;18(5):306–318.
https://doi.org/10.1002/cbm.710.
K. Larchet et al.

Journal of Forensic and Legal Medicine 100 (2023) 102596
10
52. Mauss M. Le suicide et d’homicide. Dans M. Mauss, Les causes du suicide. Presses
universitaires de France; 2002. https://doi.org/10.3917/puf.halbw.2002.01.
53. Douki S, Nacef F. Les attentats-suicides sont-ils des suicides ? Dans P. Courtet, Suicides et
tentatives de suicides. Lavoisier; 2010:210–219. https://doi.org/10.3917/lav.
court.2010.01.
54. Felthous A, Hempel A. Combined homicide-suicides: a review. J Forensic Sci. 1995;
40(5):846–857.
K. Larchet et al.