This study described the consumption of illicit drugs among university students and assessed consumption rates, as well as their associated factors. Marijuana was the illicit drug most commonly consumed by students, followed by hallucinogens. Regarding associated factors, being male, 23 years old or older, having experimented with drugs before the age of 18 and living alone or with friends were risk factors for lifetime drug use; being single, separated or widowed, having higher income, and having experimented with drugs before the age of 18 were risk factors for consumption in the 30 days prior to the interview.

Rates used in this study described the prevalence of having used no drugs, at least one drug and two or more drugs. Pilatti et al.²⁰ (2013) found a 33.3% lifetime prevalence for using at least one drug (marijuana, cocaine, inhalant or ecstasy) and a 17.4% 30-day prevalence. Although 30-day use prevalence were similar to that found in this study, our lifetime use prevalence was higher.²⁰ As a higher prevalence could be expected, considering the study population was first to fifth year university students,²⁰ one possible justification would be that Pilatti et al.²⁰ (2013) did not evaluate hallucinogens, the second most commonly used drug in this study. Andrade et al.¹² (2012) interviewed university students nationwide from all years and found 48.7% lifetime prevalence for the use of one drug and 25.9% for 30-day use, slightly higher than those observed in our study. This could be due to the fact that these authors included more drug varieties than this study did (as amphetamines, anticholinergics, tranquilizers, opioid analgesics, barbiturates, anabolic androgenic steroids), as well as that students were evaluated in later university years.

Although many studies have reported marijuana as the most commonly used illicit drug by university students^14,16,20,22, others reported inhalants and solvents to be the most common^13,28. A nationwide Brazilian study found a lower prevalence of marijuana use (26.1% lifetime; 9.1% 30-day use), as well as lower lifetime prevalence for all other drugs except inhalants, whose prevalence was substantially higher (20.4%) than that of this study¹¹. Canuto et al.¹³ (2006) analyzed first-year Brazilian university students, finding greater lifetime inhalant consumption (23.0%)¹³. The highest prevalence for this substance was associated with bars/nightclubs consumption and obtaining the substance with friends, suggesting that its use occurs in a recreational context¹³.

In our study, the age of first drug experimentation and the student’s living situation were associated with both outcomes (lifetime and 30-day use rates). First experimentation before 18 years old increased the chance of being in a higher consumption category than first experimentation after 18 years old (p = 0.003 for lifetime and p = 0.019 for 30-day rates). Other studies have found similar results, indicating that the risk of abusing or developing an addiction to illicit drugs is greater for individuals who begin using it during adolescence rather than in adulthood^29,30. For example, a 2012 United States study with adults being treated for drug dependence found that of those who first experimented with marijuana at 14 years old or younger, 13.2% were classified as addicted, six times higher than the rate of adults who first experimented with it at 18 years old or older³¹.

Regarding living situation, individuals living alone or with friends were more likely to be in a higher drug consumption category than those living with relatives or partners (p = 0.020 for lifetime and p = 0.026 for 30-day rates), corroborating the literature¹⁴. Living outside family care is reported as a facilitator of health risk behaviors. Individuals who leave their parents’ home to study and remain far from them, for instance, may have a greater sense of freedom towards new experiences and curiosities, such as the use of licit and illicit drugs³².

Regarding lifetime drug use rates, sex (p = 0.001) and age (p = 0.003) were also associated variables. Men were more likely to be in a higher drug consumption category than women. This finding could be explained by the fact that women perceive drugs as more dangerous³³. It may also reflect a greater social tolerance for men who use drugs and a greater social stigma for women³³. Biologically speaking, drug metabolism is different in men and women, with men generally having higher clearance rates³⁴. A similar result was found by Silva et al.¹⁶ (2006) and McCabe et al.³⁵ (2007), who assessed 12-month illicit drug use. Nevertheless, sex was not associated with 30-day use rates in this study adjusted analysis (p = 0.559).

In our study, age increase is proportional to the increase in chances of being in a higher drug consumption category, corroborating Passos et al.²³ (2006), who evaluated young people, mostly between 17 and 23 years old and found that as age increases, so does illicit drug use²³. Despite the trend found in this study and that of Passos²³ et al., involving very similar age groups, nearly all included individuals were between 18 and 24 years old, age group of greatest drug consumption³⁶. A study on young American adults between 18 and 29 years old found that this group has higher rates for drug use, abuse and dependence than older age groups³⁷. The difference between the results of the present study and these of Johnston et. al³⁷ could be due to the fact that after young adulthood some social responsibilities (adult roles) are assumed, such as having children, and the use of illicit drugs decreases³⁸.

Marital status and socioeconomic level were associated with 30-day use rates. Higher illicit drug use was observed among single, separated and widowed individuals (p = 0.005). A longitudinal study by Duncan et al³⁹. (2006) that assessed the effects of marriage and cohabitation on drug use, reported a decrease in the consumption of marijuana among married couples, presenting a protective effect against drug use³⁹. Thus, it has been observed that socially deviant behavior, such as illicit drug use, are incompatible with traditional adult social roles, such as marriage³⁸, which could justify the finding of the present study that single individuals consume more drugs.

Socioeconomic level was associated with 30-day drug use, in which individuals with higher incomes presented higher consumption (p = 0.001). Other studies found similar results^16,40. This was also consistent with the findings of Humensky (2010)⁴⁰, who evaluated the relationship between adolescent socioeconomic status and 30-day drug use in early adulthood. Among people whose parents had higher incomes, a higher use of marijuana and cocaine was found⁴⁰, implying that illicit substances demand is price sensitive⁴¹, This finding can be firstly explained by the fact that it is easier for students with higher incomes to acquire drugs⁴² and, secondly because parents with higher incomes generally have a greater commitment to work, contributing less to the monitoring of their children’s behavior⁴³. A study by Martins et. al⁴⁴ (2008) supports this argument by indicating that parental monitoring or positive parenting practices reduce the risk of adolescent drug use and abuse⁴⁴. Finally, students of lower economic classes tend to consider university education a path to social mobility⁴⁵, whereas richer students celebrate youth by socializing and engaging in deviant behaviors⁴⁵.

This study has some limitations. Due to the cross-sectional design, the main limitation of the study is a possible reverse causality bias in some associations, such as the marital status variable. Losses to follow-up and refusals were significant, restricting findings extrapolation, with a greater conflict among men, students aged 23 and older, and students of exact sciences and soil/agricultural science programs; which may have led to selection bias. Students not reached could have presented higher drug use, leading to difficulties to attend classes and a greater possibility of abandoning university studies. In addition, as losses were greater among men and students of exact sciences and soil/agricultural sciences, also those with a greater chance of using any substance, the prevalence found could be underestimated. Some students may have preferred not to report illicit drug use, even with guaranteed anonymity. Thus, drug use prevalence should be interpreted as minimum estimates of the actual values, which could be higher. However, using a self-administered confidential questionnaire is a standard procedure for obtaining information about this type of behavior. Another limitation is the generalization of the findings to the entire university population, since only new students were included in the study population. As it was performed in a consortium format, the number of questions in the study was also limited, preventing a more detailed outcome assessment, such as when drug experimentation first occurred and situations and places where drugs were consumed. The long duration of data collection is also one of the limitations of this study. Students who were interviewed at the end of fieldwork are more exposed to the university environment than those who were interviewed at the beginning.

The study has some strengths, as it contributes to the comprehension of the use of illegal drugs and its associated factors among university students.

In conclusion, marijuana was the most frequently illicit drug used by students, and 5.1% of the participants reported using two or more illicit drugs in the 30 days before the survey, representing the group at highest risk of dependence and possible psychological distress. Our results are expected to contribute to a better understanding of the context of substance use in this population.