Murder followed by suicide: Norwegian

data and international literature

KAREN GALTA, SIRI LERSTØL OLSEN, GUSTAV WIK

Galta K, Olsen SL, Wik G. Murder followed by suicide: Norwegian data and international literature.
Nord J Psychiatry 2010;64:397–401.

Background: In Scandinavia, the knowledge base for murder–suicides is uneven, and there has
been no statistics in criminal records. Method: We collected data from criminal registers in Norway
from 1990 to 2007, and seek to compare this with international literature from the last 50 years.
Nord J Psychiatry Downloaded from informahealthcare.com by EBSCO on 12/01/11

Results: Over 90% of murder–suicide perpetrators are males and 80% of their victims are females.
A vast majority of perpetrators are, or have been, intimate partners with their victim. The woman
has often indicated an intention to break up from the relationship before the homicide. Compared
with isolated homicides, the ages of both the perpetrator and victim are higher; most perpetrators
use firearms, and tend to be less socially marginalized. Psychiatric instability is often a background
factor, but should be seen in view of longstanding personal conflicts and threat of or loss of family,
employment or social reputation. Norwegian statistics show that one in four murderers who have
killed a near acquaintance also will commit suicide. Conclusions: Increased knowledge is essential
to prevent further tragedies of murder–suicide. Scrutinizing the taboo surrounding the topic will
probably lead to improved awareness. We suggest that a connection should be made between
For personal use only.

the homicide(s) and the suicide in criminal registers.

• Domestic violence, Dyadic death, Extended suicide, Murderer, Murder–suicide.

Karen Galta, M.D., Department of Clinical Medicine, University of Bergen, NO-5020 Bergen,
Norway, E-mail: kgalta@hotmail.com; Accepted 7 March 2010.

L ethal violence in close relationships has become an

increasing problem in Scandinavia and other Western
societies. A particular subgroup of such violence is the
murder–suicide, where the perpetrator commits suicide
after the killing of one or several family members. As
surrounded by taboo, this subgroup of homicide has been
little investigated and silently related to as “family trag-
edies”, in which surviving family members should be
sheltered. At the same time, contemporary media reveals
headlines or captions such as “Shoot his family—and
then himself ” (Kongsvinger, Norway) (1); “Four people
dead in a domestic tragedy” (Sveio, Norway) (2);
“Daddy came with a bomb for birthday present” (Odense,
Denmark) (3).
Felthous & Hempel wrote that murder–suicide reminds
us of both our limitations and challenges us to extend
our efforts to understand and to heal (4). The fact that
the perpetrator, and often his closest relatives as well,
are dead and unavailable for assessment makes it more
difficult to perform research concerning the perpetrator’s
mental health. Although increased knowledge is neces-
sary to prevent further tragedies, the knowledge base is
uneven, at least in Scandinavia. For example, there have
been no statistics on murder–suicide in criminal registers.
The homicide(s) has been registered in one category and
the suicide in another, without any notice of the connection
between the two incidents. Detailed information has been
held back, even for researchers, because of the altruistic
motive: to spare surviving family members.
Method
In this study, we compare literature on murder–suicide in
close relationships from the last 50 years, and data from
criminal registers in Norway from 1990 until 2007. We dis-
cuss characteristics of murder–suicides concerning the per-
petrator, the victims and the circumstances, as well as theories
related to causes and motives.
What here will be characterized as “murder–suicide” is
the situation where one or several homicides are followed
by perpetrator’s suicide. This act has been given different
names in the literature as: “homicide–suicide,” “extended
suicide,” “and dyadic death”. The definitions of these des-
ignations are all vague, however, and we have chosen the
term “murder–suicide”, as it is the most widely used and
neutral to perpetrator’s possible motives. We also define
murder–suicide in close relationships as a situation where
one person kills an “intimate other” before committing

© 2010 Informa Healthcare DOI: 10.3109/08039481003759201

 K GALTA ET AL.
suicide shortly thereafter. No unambiguous strict time limit
for this is given in the literature and we have therefore
included cases where the homicide(s) and the suicide are
likely parts of the same action. The term “close relations”
includes family members, present or former spouse, or
cohabiter.
The data was accumulated from different Norwegian
criminal registers and, as these were incomplete, from
reports in the press. As Norway is a country with
few inhabitants, and consequently a low number of
murder–suicides, no statistical conclusions are drawn.
Results
Characteristics on murder–suicide based on
international literature
SEX, AGE AND RELATIONS
Nord J Psychiatry Downloaded from informahealthcare.com by EBSCO on 12/01/11
Some characteristics seem to be in common for murder–
suicides in the Western hemisphere. For example is the per-
petrator most often a man, over 90% in a US sample, with the
same study showing that victims are female in 82% (4). This
is in contrast to isolated homicides in which 75–90% of the
victims are male (USA) (4). In Sweden, about 25% of perpe-
trators with intimate female victims commit suicide after the
homicide. This is about twice the rate compared with suicides
For personal use only.
after lethal violence under other circumstances (12%) (5).
The perpetrator and his victim are usually intimate
partners (6, 7). The victim has often conveyed an inten-
tion to end the relationship or marriage before the homi-
cide (7). A typical situation is thus that the man has been
rejected by a woman who has expelled him from the
household, initiated separation or divorce, and sometimes
turned her affections to another man (4). Most murder–
suicides take place at the homes of the perpetrator or the
victim (4, 8). For Australia, this figure is 70% (9). The
ages of both perpetrator and victim are slightly higher in
murder–suicides than in isolated homicides (4). The vic-
tim tends to be younger (mean age ⫾standard deviation
34.7⫾11.5) than the perpetrator (age 42.3⫾14.2) (4). This
age difference is confirmed by Swedish data indicating
that perpetrators who commit suicide after the homicide(s)
are about 10 years older than those who do not (5).
When children are victims of murder–suicide, the perpetra-
tor is most likely the biological parent (4, 10). In homi-
cides when children are victims, a significant percentage will
be followed by the perpetrator’s suicide (4, 10). In Sweden,
53% of child-killers committed suicide after the act (11).
As a rule, mothers only kill their children, whereas fathers
kill their spouse as well (10). Males who murder their entire
family (and sometimes the family pet as well) have been
called “family annihilators” (10).
ETHNICITY AND SOCIAL SETTING
In Western countries, most murder–suicides involve Cauca-
sians. In Australia, Carcach & Grabosky found that 87% of the
398
perpetrators and 86% of the victims were Caucasians (9). In the
USA, ethnic minorities committed a majority of isolated homi-
cides, whereas Caucasians committed 58.8% of murder–
suicides (4). The same is valid for Sweden, where 90% of
murder–suicide perpetrators were born in Sweden, while
about 40% of perpetrators of other spousal homicides were
immigrants (8).
Murder–suicides also differ again from isolated homi-
cides when it comes to perpetrator’s social situation. As
example, the chances are higher that the murderer has
employment if he commits suicide after the incident (9).
In Sweden, 80% of spousal murder–suicide perpetrators
were employed in contrast to 51% of those who did not
commit suicide after the homicide (8). Both isolated sui-
cides and murder–suicides occur more often in the middle
and higher social classes (4). In Australia, male perpetra-
tors of isolated homicides more often have a criminal
record (36%) than men who commit murder–suicide
(18%) (9). In Sweden, 42% of perpetrators of isolated
spousal homicides had a history of violence or threats
towards the victim, whereas this figure was 10% for mur-
der–suicide perpetrators (8). Thus, isolated homicides tend
to be committed in marginalized social settings, by per-
petrators who are unemployed, have a criminal record,
and often abuse alcohol or drugs (12), whereas murder–
suicide perpetrators are less socially marginalized (8).
LETHAL MEANS USED
Several studies conclude that firearms are more commonly
used in murder–suicides compared with isolated homicides
(4). In California, one study found the chance of the perpe-
trator committing suicide after a firearm homicide was 13
times higher than when other lethal means were used (13).
In Australia, firearms were used in 59.7% of murder–
suicides and 18.6% of isolated homicides (9). Furthermore,
male perpetrators tend to use more violent means commit-
ting murder–suicide than women. Byard et al. pointed out
that Australian males shoot, strangulate and stab, whereas
females tend to use carbon monoxide gas or poisoning in
the murder–suicide-act (10).
Firearms are more available in USA than in Scandina-
via, which is probably reflected in the fact that shooting
is the most common method in American isolated homi-
cides; however, in murder–suicides, the use of firearms is
still higher, with rates between 88% and 95% in later
studies (4). In Sweden, stabbing is the most common
lethal means in general lethal spousal violence (40% stab-
bing, 20% shooting). When it comes to murder–suicides,
however, firearms are used in 70% of cases (5, 8).
Theories on causal relationships based on
international literature
There are some studies concerned with searching causal
relationships between the perpetrators’ psychopathological
NORD J PSYCHIATRY·VOL 64·NO 6·2010
 MURDER FOLLOWED BY SUICIDE

condition and the act of murder–suicide. Such studies are
mostly based on surviving perpetrators or those who left
suicide notes.
DEPRESSION
In the majority of contemporary studies, depression is
suggested to be the commonest mental disorder behind
murder–suicide (4). Also, 50 years ago, West found that the
most prevalent disorder in his material was depression (14).
However, at the time of West’s study, a majority of perpe-
trators were females, with 41% being mothers, who almost
exclusively killed their children before committing suicide.
It has been suggested that mothers who kill their own chil-
dren prior to their suicide often have an altruistic motive,
i.e. in a depressive delusion, they wish to save their children
from what is thought to be a horrible fate (10). The mother
Nord J Psychiatry Downloaded from informahealthcare.com by EBSCO on 12/01/11
testing (4). This seems frequent among female perpetrators
who kill their children and male perpetrators who erase their
entire families, the family-annihilators, in comparison with
perpetrators who kill their spouse only (4). Okumura &
Kraus described a psychotic woman dominated by a perse-
cution complex in which a voice told her to kill her son and
herself. The suicide attempt failed and the mother could be
examined some time afterwards (16).
JEALOUSY
Jealous paranoia appears to be an important causal factor
for murder–suicides with 50–75% of all murder–suicides
being of this type (17). The jealousy paranoia type of
murder–suicide has been thought to represent an “extended
suicide” motivated by the perpetrator’s fear of loosing his
or her partner trough rejection or illness. The perpetrator

commits the crime in the belief that “my child will not make
it being left in the world without me” (14). Women have
also been suggested to consider their children as their own
organic continuation, which may make the act more natural
to these mothers (14).
Severe mental depression may also be a cause for
male perpetrators, and an article from USA discusses a
father, who killed his wife and their three children but
survived his own suicide attempt (15). The father was
For personal use only.
thus considers his spouse as a part of him, and takes her
with him in his suicide (18).
Murder–suicide in Norway 1990–2007
During the period between 1990 and 2007, there were a total
of 675 victims of lethal violence in Norway (19), 41%
(280 victims) of which had a close relation to the perpe-
trator. During the same period (1990–2007), there were

778 perpetrators of lethal violence (19), 34% (263 perpe-

found to be severely depressed and suicidal, most proba-
bly constituting major factors leading towards the homi-
cide and attempted suicide.
PSYCHOSIS
A significant minority of murder–suicide perpetrators might
be psychotic with thought disturbance and loss of reality
trators) of which were in close relation to their victim(s).
Of those perpetrators, 23% (60) committed suicide after
the act. See Table 1, which shows murder–suicides in close
relationships committed in Norway 1990–2007.
Which relations between the perpetrator and the victim(s)
are most common in murder–suicide in close relations?
The victim is most often a present or former partner (spouse

Table 1. Murder–suicide in close relations in Norway 1990–2007.

Total number
Cohabiter Former cohabiter Divorced/separated Married Parents (victims) Children (victims) Other relatives of perpetrators

1990 3 1 (1) 3 (1) 3 (1) 3 13 (3)

1991 4 (1) 2 4 (2) 8 (1) 3 (2) 2 20 (6)
1992 2 1 (1) 3 4 (1) 3 3 (2) 1 17 (4)
1993 3 1 6 (4) 4 (1) 4 (3) 1 17 (6∗)
1994 6 (2) 1 1 1 2 11 (2)
1995 3 1 1 3 (1) 8 (1)
1996 4 3 (1) 2 (1) 2 1 4 (1) 18 (3)
1997 1 1 (1) 2 3 2 4 (1) 2 (1) 15 (3)
1998 3 1 2 4 1 10
1999 1 1 (1) 2 (1) 2 4 (2) 6 (1) 15 (4∗)
2000 3 3 (2) 6 (4) 5 1 (1) 18 (7)
2001 2 (1) 3 2 (1) 4 (1) 4 (3) 2 14 (3∗)
2002 1 (1) 3 4 (3) 1 4 (1) 2 (1) 15 (5∗)
2003 4 (1) 1 1 4 (1) 5 (3) 11 (3∗)
2004 3 2 (1) 1 2 2 4 14 (1)
2005 4 1 4 (1) 3 1 2 14 (1)
2006 1 (1) 2 1 4 (2) 2 (1) 4 (4) 3 13 (6∗)
2007 2 6 (1) 4 5 (4) 3 20 (2∗)

*Actual number of perpetrators who committed suicide that given year.

NORD J PSYCHIATRY·VOL 64·NO 6·2010 399

 K GALTA ET AL.

Other our statistical calculations more difficult. For instance, in

relatives Cohabiter Fig. 2, one cannot include the categories for parents who
5% 10% Former
cohabiter
kill their children, children who kill their parents or other
7% relatives. This is because we cannot be sure about the
actual numbers of perpetrators in these particular catego-
Divorced/ ries. If for instance one father kills two children, this
separated
5% will be stated as two, but the actual number of perpetra-
Children tors is one. In the categories for present or former
(victims)
37%
partners, one can assume that there is only one perpetra-
tor and one victim. In this, it should be mentioned
Married that international studies have concluded that offenders
32% who kill their own children are likely to commit suicide
afterwards (the odds ratio for the offender to commit
Parents suicide after killing his child is 9.7 with P ⬍ 0.01 in
(victims) Australia (9)).
4%
Nord J Psychiatry Downloaded from informahealthcare.com by EBSCO on 12/01/11

Fig. 1. Relationship between victim and perpetrator in murder–

suicide in close relationships.
or cohabiter), and these categories constitute a total of 54%.
Perpetrators who kill their children constitute 32% (Fig. 1).
The perpetrator is most likely to commit suicide (38%)
when the victim is his spouse, while 18% of perpetrators
who killed their divorced/separated spouse, 26% of per-
For personal use only.
Discussion
In summary, the typical murder–suicide perpetrator is an
outwardly well-functioning male in his 40s. His victim is
most likely his female partner, who often has indicated
an intention to break up the relationship. The perpetrator
typically comes from the middle or upper social strata
and seldom has a criminal record. Firearms are the most
commonly used lethal means. In cases with female per-

petrators who killed their former cohabiter and 15% who
killed their present cohabiter committed suicide after the
homicide. In total, 26% of perpetrators who killed their
present or former spouse or cohabiter committed suicide.
See Fig. 2, where percentage of murder–suicide and lethal
violence are given for different relationships.
The statistics from Norway include only murder–
suicides where there was a close relationship between
perpetrator and victim, which keep a few murder–sui-
cides committed outside such a relationship out of the
material. Table 1 shows the number of victims killed in
a close relationship. The numbers enclosed within paren-
theses shows how many perpetrators committed suicide.
The numbers enclosed within parentheses and marked
with an asterisk show the actual number of perpetrators
who committed suicide that given year. An offender may
have victims in two different categories, which makes
Total 26%
Married
Divorced/ separated 18%
Former cohabiter 26%
Cohabiter 15%
0% 10% 20% 30%
Percentage of perpetrators who committed suicide
Fig. 2. In which close relation does the perpetrator most often commit
suicide after the crime?
petrators, the victim is nearly always her own child. In
Norway, one out of four perpetrators of lethal violence in
close relationships will commit suicide after the murder.
Psychic imbalance is per definition essential in the
complex picture that leads a person to commit murder–
suicide, and psychiatric conditions such as depression,
psychosis and jealous paranoia have been pointed out as
central. However, those factors have seldom been recog-
nized beforehand and they should thus probably be seen
in a complete setting, where the murder–suicide most
likely is the end result of an interaction between indi-
vidual psychopathology and social factors. Such factors
could be loss of family or partner, employment, or social
reputation. The fear of loss of honour or face may thus
forego and dominate over specific psychopathological
motives in some murder–suicides.
Increased knowledge is necessary to prevent further
murder–suicide tragedies. However, the exploration of these
incidents has been insufficient. One reason for this is that
both perpetrators and victims are dead and unavailable for
assessment. Another important factor is without doubt the
38%
general silence surrounding the phenomenon as such. Scru-
tinizing that taboo will probably lead to improved aware-
ness and knowledge. It is probably also the best way to
establish adequate preventive ventures for the future.
Several international authors have pointed out the fact
40%
that registration of murder–suicides is deficient. Some also
recommend the authorities to establish a category in
national statistics exclusively for murder–suicides (20). We
suggest that a first step to elucidate the phenomenon of

400 NORD J PSYCHIATRY·VOL 64·NO 6·2010


murder–suicide is to register a connection between the
homicides and the suicide, and let these data be accessible
to science.
Declaration of interest: The authors report no conflicts of
interest. The authors alone are responsible for the content
and writing of the paper.
References
1. Heading, “Skjøt familien—og så seg selv”. VG 30 September
2002.
2. Heading, “Fire døde i familietragedie”. Aftenposten 21 January 2003.
3. Heading “Pappa kom med bombe i bursdagspresang”. VG 08
August 2005.
4. Felthous AR, Hempel A. Combined homicide–suicides: A review.
J Forensic Sci 1995;40:846–57.
5. Rying M. Dödligt våld mot kvinnor i nära relationer. BRÅ-
Nord J Psychiatry Downloaded from informahealthcare.com by EBSCO on 12/01/11
rapport 2001:11. Stockholm: Brottsförebyggande rådet; 2001.
6. Hanzlick R, Koponen M. Murder–suicide in Fulton County,
Georgia, 1988–1991. Am J Forensic Med Pathol 1994;15:168–73.
7. Aderibigbe YA. Violence in America: A survey of suicide linked
to homicides. J Forensic Sci 1997;42:662–65.
8. Belfrage H, Rying M. Characteristics of spousal homicide perpe-
trators. Criminal Behav Mental Health 2004;14:121–33.
9. Carcach C, Grabosky PN. Murder–suicide in Australia. Trends &
Issues in Crime and Criminal Justice. Canberra: Australian Insti-
tute of Criminology; 1998.
10. Byard RW, Knight D, James RA, Gilbert J. Murder–suicides
For personal use only.
involving children: A 29-year study. Am J Forensic Med Pathol
1999;20:323–7.
NORD J PSYCHIATRY·VOL 64·NO 6·2010
MURDER FOLLOWED BY SUICIDE
11. Somander LH, Rammer LM. Intra- and extra familial child
homicide in Sweden, 1971–1980. Child Abuse Negl 1991;15:
45–55.
12. Petersson L, Rying M. “Dödligt våld” Brottsutvecklingen i Sverige
2001–2003. BRÅ-rapport 2004:3. Stockholm: Brottsförebyggande
rådet 2004:23–42.
13. Lund LE, Smorodinsky S. Violent death among intimate partners:
A comparison of homicide and homicide followed by suicide in
California. Suicide Life Threat Behav 2001;31:451–9.
14. West DJ. Murder followed by suicide, 1st edition. London:
Heinemann; 1965.
15. Schlesinger LB. Familicide, depression, and catathymic process.
J Forensic Sci 2000;45:200–203.
16. Okumura Y, Kraus A. Zwolf patientinnen mit erweiterter selbstto-
tung—psychologie, persönlichkeit, motiavation, vorgeschichte und
psychosoziale konfiktsituation. Fortschrifte der Neurolgie. Psychi-
atrie und ihrer Grenzegebiete 1996;64:184–91.
17. Marzuk PM, Tardiff K, Hirsch CS. The epidemiology of murder–
suicide. JAMA 1992;267:3179–83.
18. Palermo GB. Murder–suicide—An extended suicide. Int J Offender
Ther Comp Criminol 1994;38:205–16.
19. Kripos criminal register, 2007: www.politi.no/pls/idesk/docs/
f1537358626/drapsstatistikk2007.pdf
20. Palermo GB, Smith MB, Jentzen JM, Henry TE, Konicek PJ,
Peterson GF, et al. Murder–suicide of jealous paranoia type—
A multisenter statistical pilot study. Am J Forensic Med Pathol
1997;18:374–83.
Karen Galta, M.D., Department of Clinical Medicine, University of
Bergen, NO-5020 Bergen, Norway.
Siri Lerstøl Olsen, M.D., Department of Clinical Medicine, University
of Bergen, NO-5020 Bergen, Norway.
Gustav Wik, M.D. Ph.D., Department of Clinical Medicine, University
of Bergen, NO-5020 Bergen, Norway.
401