Murder-suicide is a relatively rare event, but one with a very high emotional and social impact. The two components are theoretically related, as they both include aggression and violence(1). In an epidemiological perspective, approximately 10% of homicides are followed by the suicide of the perpetrator(2). Furthermore, over 90% of psychiatric patients that perpetrated a homicide attempt also made suicide attempts(3). Although intensive research has been done on suicide, and several theoretical models have been proposed over the years, there is still a relative paucity of data on the association of homicide and suicide(1), in spite of the public impact that several such events had in the last few years.
Therefore, the present paper aims to review the latest scientific literature on murder-suicide.
We conducted a PubMed search using “murder suicide [MeSH Terms]” as search phrase. A total of 653 articles were retrieved from this initial search. Articles published between 2015 and 2020, that were written in English, were included in the analysis. Articles focused on violent deaths in general and articles focused on firearm legislation were excluded.
After applying the aforementioned inclusion and exclusion criteria, 38 articles were finally included in the present analysis.
Murder-suicide is currently defined as a “homicide that is closely followed by the suicide of the perpetrator”(2). Other authors also use terms like homicide-suicide, extended suicide, or dyadic death. The latter terms try to include phenomenological aspects in the very term used, but since the phenomenology of this phenomenon is still intensely debated, the term “murder-suicide” is the one that is more frequently used.
The time elapsed between homicide and suicide varies, according to author, from 24 hours to several days or weeks(4).
There are two different forms of murder-suicide described, that are significantly different between them. In most cases, men kill their partners or children(2). The second form is terrorist suicide(6). The two forms differ in that, in the latter, the perpetrator has no personal relationship with the victim, the motivation, goals and expectations related to the act are very different and, finally, the prevalence of mental and behavioral disorders in the two groups of perpetrators is significantly different(6).
Homicide-suicide is a relatively rare phenomenon. It accounts for 20-30 deaths weekly in the USA(5), but the incidence has a large variability according to country(4). It is estimated at 0.02/100,000 inhabitants in Finland, 0.04/100,000 inhabitants in Italy, 0.05/100,000 inhabitants in England and Wales, 0.07/100,000 inhabitants in Australia, 0.2/100,000 inhabitants in France, between 0.27 and 0.38/100,000 inhabitants in the USA and, finally, 0.44/100,000 inhabitants in Japan(4).
The demographic data on the perpetrators show that 90% of them are male, and that their victims are mostly female (77%) or children (13%)(5). They tend to be much older than perpetrators of homicides alone or persons who died by suicide without committing homicide(4). Furthermore, 30-55% of perpetrators are involved in an intimate relationship, and in 60% of the cases the perpetrator and the victim are intimate partners(5,6). Also, 22.5-37.5% of the perpetrators were under the influence of both alcohol and drugs at the moment of the act, with an additional 10% being positive for alcohol alone(5,6).
Homicide-suicide almost always happens relatively shortly after separation, divorce or relational conflicts(4), and the familial ties between the perpetrator and the victim make suicide after homicide more likely(8), with the risk of suicide after homicide increasing proportionally to the level of proximity between perpetrator and victim(4). For example, biological fathers are at a greater risk for suicide after murdering their children than stepfathers or adoptive fathers(4).
One study found that 40% of murder-suicide perpetrators in England and Wales addressed general practitioners or sought counselling within one month before the act and that 62% of the perpetrators were previously diagnosed with a mental and behavioural disorder(9). Major depressive disorder seems to carry the highest risk for murder‑suicide, especially if it is comorbid with substance use disorders(4). Psychosis was rarely diagnosed in perpetrators and personality disorders were virtually never diagnosed in this group(9), although 57-72% of perpetrators had criminal records at the moment of the act(4,10). Studies also show that 41% of perpetrators previously attempted suicide(9), and that early adverse childhood experiences, a significant risk factor for many mental and behavioral disorders, also increase the risk for murder-suicide(4).
The motivational factors behind murder‑suicide are extremely diverse and poorly documented in the literature. However, two of them stand out as being most frequent – i.e., jealousy and delusions.
Jealousy is the most frequent, with possessive/obsessive convictions and ruminations being very frequently encountered in perpetrators(4). From a psychodynamic perspective, the function of such an act could be the object-subject fusion in death(4).
Although psychosis is rarely diagnosed in perpetrators of murder-suicide(9), delusions are thought to be present(4). In this instance, the function of the act could be to spare a loved one from suffering – i.e., especially in filicide-suicide cases, with the perpetrators being dominated by emotions of relief, anger or fear rather than compassion(4).
Discussion and conclusions
Most studies conducted so far were based on forensic data, thus our understanding of the psychological processes behind murder-suicide is extremely limited. Nevertheless, the evidence shows that, as recommended before(11), any patient with suicidal ideation should also be evaluated for homicidal ideation. Furthermore, the data show that homicide, suicide and murder-suicide are very different phenomena; therefore, they should be reported separately.
Some authors go as far as to suggest that murder-suicide should be considered a form of murder, rather than a form of suicide(2), although altruistic homicide-suicide is a well documented pattern, because the rates of murder‑suicide are strongly associated with those of homicide(12), and because the rates of murder‑suicide are not highly correlated with those of suicide(2).