Violent Deaths Amongst Cannabis Users

At the Department of forensic Medicine in Stockholm various studies on drug-related mortality are carried out. One important object is to create a register of deaths related to illicit drug use in the Stockholm area. This register includes users of heavy drugs chiefly amphetamine and heroin.

However, in recent years a number of cases were found with tetrahydrocannabinol but no other illicit drugs, in blood or cannabinoids in urine.
At closer investigation, it appeared that manners of death among the cannabis users were unexpectedly violent compared to the ways of death among other drug addicts. The proportion of suicides, 10 out of 24, was particularly high. This finding was remarkable, since cannabis use is not generally associated with violence.
The literature contains several descriptions of delusional states, paranoid symptoms and personality disturbances in cannabis users . These abnormal mental states are of comparatively short duration, usually lasting for just a few hours. Even first-time users may be affected.

These so-called cannabis psychoses are not very frequent and are often induced by long periods of intensive cannabis use. It is uncertain whether or not these psychoses can be separated from schizophrenia or other known paranoid states . The entity distinguishing cannabis psychosis from other mental disorders is the significant subsidence of the symptoms after some days or a few weeks. In most cases, the patient recovers totally but remains at risk for relapse in connection with repeated cannabis use. In case studies of cannabis psychoses, it is not uncommon to see descriptions of violent and destructive outbursts, suicide attempts and assaults on other persons, often with major violence. In a Swedish follow-up study of conscripts with a history of cannabis use, an excess mortality was noted 15 years after conscription. The excess mortality from suicides was particularly high, and the share of suicides increased in proportion to the magnitude of the cannabis abuse.

During these four years, a total of 13,417 medico-legal autopsies were made on males and females of all ages (about 26% after violent deaths). A total of 1.9 million people resided in the service area in 1987; of these, about 1.6 million lived in Stockholm county and 668,810 in the city of Stockholm. Information about deceased persons was obtained from police reports and was, when possible, supplemented with information from clinical journals, families and friends as well as from social workers. Complete autopsies were carried out, with few exceptions, when the blood tests analyses concerning HIV-infection were positive The autopsies were supplemented with histological investigations. Toxicological analyses were made, with the purpose of disclosing the presence of alcohol, barbiturates, tranquilizers, opiates, central stimulants and cannabis, as suggested by autopsy findings, past history, police records, or circumstances of death. Toxicological analyses were also carried out when insufficient information about the circumstances of death was at hand, for instance in cases of violent death, or when the postmortem findings were inconclusive, and also when information about unusual behaviour or symptoms was available.

During the four-year period of 1987-1990, cannabis was found to be the only narcotic drug in postmortem samples in 24 cases; in 8 of these, cannabis was the only finding, while alcohol, alcohol and medicinal drugs, or only medicinal drugs was demonstrated together with cannabis in 10, 5 and 1 instance, respectively. The number of deaths among these 24 cannabis users during the study period is shown in Figure 1.

There were 23 males and 1 female, 20-43 years of age (mean age 29.6 years). About two thirds (15) of the persons in this series were, at the time of death, under the influence of alcohol.

Figure 1. Numbers of cannabis user deaths in Stockholm 1987-1990

Figure 2. Manners of death among 24 cannabis users in comparison to the same number randomly selected alcohol users, amphetamine and heroin users’ deaths. N = Natural deaths. A = Accidental deaths. S = Suicides. H = Homicides. U = Undetermined deaths.

Among the 24 cannabis users, only one died from a non-violent cause, while the remaining 23 (96%) died as a result of violence, compared to a total of 26% of the entire number of 13,417 investigated postmortems during the same period (Table I). No stigmata which could be indicative of cannabis use were noted at the postmortems. None of the cannabis users had blood tests that were positive for HIV.


Cause and manner of death Total n=24
Non-violent deaths, total 1

Complications of alcohol abuse (303)

Violent deaths, total 23
Accidental deaths, total 8

Traffic accidents (E 89)
Alcohol intoxication (E 859)
Drowning (E 910)

Suicide 10

Tablet intoxication (E 950)
Carbon monoxide intoxication (E 951, 952)
Jumping from high place (E 957)

Homicide, total 5
Stabbing (E 966) 5

The manners of death among 24 cannabis users were compared with those in the same number of users of alcohol, amphetamine and heroin (Figure 2); a similar distribution for alcohol users and amphetamine users was noted. There were obvious differences between cannabis users and heroin users, with suicides dominating for cannabis users, while deaths of undetermined origin, following collapse in connection with intravenous drug administration, predominated among heroin users.

Accidental deaths
Eight of the 24 cannabis users died in accidents, 7 males and one female, ranging in age from 22 to 43 years (mean 32.5 years). Two users died of alcohol intoxication. At the time of death, the concentrations of THC in the blood were only 0.8 and 2.0 ng/g. One male with 0.5 ng THC/g blood and 2.7 g ethanol/l blood died from drowning after suddenly loosing his balance and falling into the water. Five cannabis users died in traffic accidents, 3 as drivers of motor vehicles, and 2 as passengers with cannabis-influenced drivers. One of the deceased car drivers had 30 ng THC/g blood (and no other drug), while the remaining 4 traffic accident victims had 1-4 ng THC/g blood in combination with alcohol (0.3, 0.9, 2.8 and 3.2 g/1), in two cases also with low concentrations of chlormezanon and paracetamol. In one of the automobile accidents, high speed ( 3 times exceeding the speed limit through central Stockholm) preceded collision with another car; in another, the cannabis influenced driver passed another car, at twice the speed limit, and just continued to drive on the left side of the road until he collided with a car coming from the opposite direction. In the two others, the drivers were not able to keep their cars on the road in a minor curve; and in yet another, the car was simply driven into a ditch.

Cannabis Associated Deaths in Medico-Legal Postmortem Studies Preliminary Report
Rajs, Fugelstad and Jonsson. Dept.of Forensic Medicine. Karolinska Institute, Stockholm
Dept. of Psychiatry, St.Goran’s Hospital, Stockholm and National Laboratory of Forensic Chemistry, Linkoping, Sweden. Reported in the procedings of the Second Int.Symposium. Paris. April 1992

Back to top of page

Powered by WordPress