USP researchers analyzed samples from 3,577 victims from different regions of the country
By José Tadeu Arantes
A study carried out at the University of São Paulo found that more than half (53%) of the victims of violent deaths that occurred in four Brazilian capitals had alcohol or drugs in their system in analyzes carried out shortly after the death. 3,577 cases were evaluated in Belém, Recife, Vitória and Curitiba, representing, respectively, the North, Northeast, Southeast and South regions. The results were in the magazine Toxics.
“The objective was to produce standardized and comparable data on the role of psychoactive substances in deaths from external causes in Brazil”, says the biomedical toxicologist, post-doctoral researcher at the Faculty of Pharmaceutical Sciences of the University of São Paulo and first author of the article.
Laboratory analyzes included alcohol, a range of illicit drugs and psychoactive medications, with standardized protocols. The team also adopted operational precautions to reduce losses due to degradation. “Especially in the case of alcohol, if the sample is not stored properly, the substance can degrade and mask the result.”, explains the researcher.
“The association between the substance and violent death in the case of homicide is very complicated, because we are only looking at the victim, we are not looking at the aggressor. Still, it is possible to attribute the high presence of cocaine not only to the acute use of the substance, but to the social and economic context in which the illegal market operates, to the trafficking, sale and purchase environment that characterizes what we call structural violence.”, argued Bombana.
Bombana says that the study was made possible through an agreement signed in 2020 between USP and the (National Secretariat for Drug Policies and Asset Management) to map the relationship between alcohol and drug use and violent deaths. The four capitals were chosen by combining two criteria: magnitude of the problem and strategic relevance.
“These cities were selected based on the mortality rate from external causes and because they are strategic points on the drug trafficking route.“, explains the researcher. The choice also took into account the country’s role as an international circulation corridor: “Often the drug comes from other countries and passes through Brazil to be distributed to the United States, Europe, Africa”.
Collection took place between 2022 and mid-2024. “We assembled and trained teams of four researchers in each of these cities to collect blood samples during necropsies. This material was frozen and sent to our laboratory at USP, where we had a team of five researchers to carry out the analyses.”, explains Bombana.
Men are the majority
The profile of the victims reflects the most common aspect of violent mortality in the country: 90% were men, 56% were aged 30 or over and 67% died by homicide. This last data is especially relevant when compared to the percentages of deaths due to traffic accidents (15%) and suicides (9%). In the North and Northeast, the highest percentage was of individuals characterized as “brown”, according to the nomenclature adopted by the (Brazilian Institute of Geography and Statistics), while in the Southeast and South the majority were “white”.
Among all victims, 53% tested positive for at least one psychoactive substance. The most detected were: cocaine (30%), alcohol (28%), benzodiazepines (7%) and marijuana (2%). “The predominance of cocaine was very significant in homicide cases, while alcohol was the substance most detected in deaths due to traffic accidents. Benzodiazepines prevailed in suicides”, relata Bombana.
The presence of alcohol in traffic deaths is a long-standing problem in the country. “The topic has been discussed for at least 30 years, without a solution being found. The legislation is very robust, but perhaps what is missing is greater control over the sale of alcohol. Some countries have much stricter and more restrictive rules for sales”, he ponders.
Risk signs
Cross-sectional research does not allow establishing a cause and effect relationship. It is a type of study in which data is collected in a single “snip” in time, as if composing a photograph of reality.
In the research, for each victim, the researchers recorded the type of death (homicide, traffic, suicide, etc.) and the result of the post-mortem toxicological analysis (cocaine, alcohol, benzodiazepines, etc.). Then, they compared the two sets of data. This makes it possible to measure prevalence – for example: “53% had some substance in their blood” – and identify associations – for example: cocaine appeared more in homicides; alcohol, in traffic deaths. But the research does not allow us to prove, through cause and effect, that cocaine “caused” the homicide. Likewise, the cross-sectional design, alone, does not “close” the causal chain between alcohol consumption and traffic death. “What can be safely stated is the existence of consistent signs of risk”, highlights Bombana.
By analyzing police records associated with homicide cases, the team found that about 85% of deaths resulted from gunshot wounds. “This occurred at a time when, through decrees and ordinances, the then federal government relaxed rules for purchase and possession, increased limits on weapons and ammunition, expanded authorized categories and reduced control and inspection mechanisms, a context that helps characterize the pattern of lethality observed.”, emphasizes Bombana.
As for suicides, the emphasis on benzodiazepines raises questions about medication use, self-medication and vulnerability. The researcher suggests a plausible hypothesis, without attributing direct causality: “The use of these substances can end up serving as a trigger to move from suicidal ideation to actual suicide.”.
More broadly, this observation summarizes a mechanism common to different types of violent death: the consumption of substances can lead the individual to enter more dangerous environments (in the case of homicides) or act in a more dangerous way (in the case of traffic accidents).
Tailored interventions
The map of fatal occurrences is not uniform. There are differences in the pattern between the four capitals studied: Recife with a prevalence of deaths associated with alcohol (alone or combined); Vitória and Belém with the highest concentration of deaths associated with the use of illegal drugs (without alcohol); and Curitiba with alcohol taking precedence over illegal drugs. “Brazil has continental dimensions and each city has social, cultural, health and security specificities. The pattern of substance use reflects these specificities”, comments Bombana. For the researcher, this heterogeneity should guide tailored interventions, supporting public policies focused on the reality of each city or region.
Although he emphasizes that he is not an expert in public policies, Bombana argues that tackling the problem tends to be more effective when focused on public health and harm reduction rather than repression. “Perhaps the criminalizing policy, the so-called ‘war on drugs’, is not the best option. Portugal decriminalized and saw a decrease in the number of users, petty crimes, homicides and overdoses. The differences between Portugal and Brazil are enormous, of course. Starting with the sizes of territories and populations. Still, the Portuguese example suggests that a harm reduction policy is perhaps the most interesting path.”
The study was conducted by the “Alcohol, Drugs and Violence” group at the USP Faculty of Medicine, coordinated by Bombana and the professor, who also signed the article, and received support from Fapesp through a grant to Bombana.
This text was originally published on May 25, 2026. The content is free for republication, the source is cited, and has been adapted to the standard of Poder360.
