Marijuana use among adolescents: a cross-sectional study with Brazilian students from a remote area of the Amazon region

Marijuana is the most used illicit drug worldwide. The frequent use of marijuana in adolescence is very worrying, as it is associated with health problems. In Brazil, surveys on the use of psychotropic drugs with adolescent students were conducted in large cities. This study determined the prevalence and factors associated with marijuana use among adolescent students in the Brazilian city of Soure, Pará, in the Amazon region. In this cross-sectional study, 736 high school adolescent students provided information through a structured form at the three schools in Soure in November 2019. Marijuana use, in life and frequent (last 30 days), was calculated. Logistic regression models identified factors associated with different marijuana use. The rates of marijuana use in life and frequent were 17.7% and 7.1%, respectively. The behavioral pattern of the adolescents' coexistence nucleus (parents, relatives, and friends) was fundamental for marijuana use in life. Also, the lack of interference or reduced participation of the family in the adolescent's life, the inadequate family structure, and the continuous example of friends were associated with the frequent use of marijuana. Measures to control and prevent marijuana use are necessary and should be available to adolescents, their families, and friends in this remote area of the Amazon region.


Introduction
Adolescence is the stage of life between childhood and adulthood, from 10 to 19 years old, and is marked by significant changes in the physical, cognitive, and psychosocial fields. These changes affect how adolescents feel, think, make decisions, and interact with others (World Health Organization [WHO], 2020a). Around 1.2 billion people, or 1 in 6 of the world's population, are adolescents. Most of them are healthy, but there are many health problems such as depression, violence, infectious diseases, unplanned pregnancy, and the use of psychotropic drugs (United Nations Office on Drugs and Crime [UNODC], 2018; WHO, 2020b).
In the world, marijuana is the most widely used illicit drug (Peacock et al., 2018). Annually, about 2.5% of the world's population (approximately 147 million people) use marijuana and consumption has increased more rapidly than cocaine and opioids (WHO, 2016). The frequent use of marijuana during adolescence is a concern, as it has been associated with decreased school performance, lower level of education and school dropout, risk of dependence, early onset of psychosis, neuropsychological decline, and use with other illicit drugs (Degenhardt et al., 2016;Gobbi et al., 2019;Silins et al., 2014;Volkow et al., 2014). Parenting styles, substance use by parents and colleagues, risk perception have been described as relevant psychosocial risk factors (Chadi et al., 2018;Libuy et al., 2020).
In 2017, the world record for most marijuana seized was recorded in South America. In recent decades, Brazil has emerged as a major illicit drug market, such as marijuana, crack-cocaine, and cocaine (UNODC, 2019). It is estimated that 4.9 million Brazilians use some illicit drug per year. This rate is higher among men, youth, and adolescents, compared to women (Fundação Oswaldo Cruz [FIOCRUZ], 2017). Major epidemiological surveys on the use of licit and illicit drugs have been conducted in the past decades with students in major Brazilian cities, who point to alcohol and marijuana as the most consumed licit and illicit drugs, respectively (Centro Brasileiro de Informações sobre Drogas Psicotrópicas [CEBRID], 2004;CEBRID, 2010;FIOCRUZ, 2017;Soldera et al., 2004). The 6th national survey on the use of psychotropic drugs among elementary and high school students from the public and private education networks of the 27 Brazilian capitals showed that 25.5% of Brazilian students reported having already used illicit drugs (marijuana, cocaine paste, crack-cocaine, amphetamines, solvents, and others) in their lives, of which 10.6% used in the last year and 5.5% in the last month (CEBRID, 2010).
In the Brazilian Amazon (northern Brazil), there are few epidemiological studies on the use of illicit drugs among adolescents, and most of the information comes from surveys with students from public and private schools located in state capitals. In the cities of Belém (n = 2,067) and Manaus (n = 2,389), the two main cities in this Brazilian region, the prevalence of marijuana use in life among adolescent students was 4.2% and 5.3%, respectively. Being that the frequent use of marijuana (use more than six times in 30 days) among adolescent students in both cities was equal to 0.1%, mainly concentrated in the age group of 16 to 19 years old (CEBRID, 2010). To date, only two studies have been conducted with adolescent students in distant cities (Breves and Capanema) in metropolitan areas in the Brazilian Amazon, which indicated a very high prevalence of lifetime use (8.8% to 23.4%) and frequent use (4.0% to 12.5%) of marijuana, with records also of the use of cocaine paste and crack-cocaine by adolescents. In these studies, the factors associated with the use of marijuana were school failure, reduced participation of parents in the school life of adolescents, parents who do not talk to their children about the use of psychotropic drugs, use of licit drugs by adolescents, involvement in conflicts, and parents/friends who consume licit and/or illicit drugs (Alcantara et al., 2017;Furtado et al., 2017).
The Brazilian Amazon is an underdeveloped rural and socioeconomic region, with high levels of poverty, low level of education, limited transport infrastructure, and inadequate health services . In this context, poverty, malnutrition, domestic violence and neglect of children and adolescents, sex work and illicit drug use, and trafficking are common in this region. To increase vulnerability, this region is an important route for illicit drug trafficking and people due to unsafe borders with the largest cocaine producers (Peru, Colombia, and Bolivia) in the world . The assessment of epidemiological status and the quantification of possible risk factors and their trends over time can help considerably in the direction of strategies to prevent the use of marijuana and other psychotropic drugs, especially in a remote, unknown, and low human development area. Thus, this study determined the prevalence and factors associated with the use of marijuana among adolescent students in the city of Soure, Brazilian state of Pará, in the Amazon region.

Methodology
Soure is a city located in the Marajó Archipelago, one of the areas with the lowest human development index in Brazil, the Amazon region ( Figure 1). This city has about 25 thousand inhabitants and is one of the main tourist areas of the Research, Society and Development, v. 10, n. 14, e499101422387, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i14.22387 This cross-sectional study consisted of quantitative and qualitative information provided by adolescents attending high school in the city of Soure, state of Pará (Campana, 1999;Medronho et al., 2011). In this city, there are only three high schools, and all were accessed. Initially, parents and guardians were consulted about the participation of adolescents in this study. During meetings, parents and guardians of the adolescents were informed of the objectives and conduct of the study, as well as received written invitations to be signed and delivered by their children on the scheduled visit date of the research team at the school. All data were collected during class time through intervention to explain the research objectives and invite all students present to participate in the study by filling out the form. The participation authorization provided by the parents or guardians was fixed to the form filled out by the adolescent and then deposited together in an urn. Each high school class was visited only once and all the students present, included in the age group for adolescents (10 to 19 years old) who filled out and delivered the forms, were included in this study. All students over the age of 19, who provided completed forms without parental or guardian authorization, or who were absent from the classroom at the time of the visit, were excluded from this study.
Data collection was conducted from 4 to 29 November 2019. The instrument used for data collection was a selfcompleted form, without personal identification, covering issues related to school routine, social, economic, and demographic characteristics and use of psychotropic drugs. This form was used in other epidemiological studies (Alcantara et al., 2017;Araujo et al., 2017;Furtado et al., 2017), and contained questions regarding age, sex, school performance, work activity in parallel with the studies, marital status of parents/guardians, participation of parents/guardians in school life, educational level of parents/guardians, family income of parents/guardians, family guidance on the use of psychotropic drugs (licit and/or illicit), involvement in conflicts with the family or coexistence core, use of marijuana in life, frequent use of marijuana in the last 30 days, age who used marijuana for the first time in life, parents used psychotropic drugs (licit and/or illicit), friends used psychotropic drugs (licit and/or illicit), and how many friends used psychotropic drugs (licit and/or illicit). A question related to the use of a fictitious drug ("kripton tea") was used as a marker of information reliability. All information from adolescents who reported using this fictitious drug was excluded. In this study, the use of marijuana a few times in life and the permanent abandonment of such use was considered to use in life. On the other hand, the use of marijuana at least twice a week, over the last 30 days, was considered as frequent use.
All data were entered into an Excel database (Microsoft Corp., USA) and converted to SPSS. Confidence intervals (CI) were constructed to estimate the prevalence of psychotropic drug use. Marijuana use was considered the outcome. On the form, the question "Did you use marijuana during your life?" (Yes/No) was essential to identify the dependent variable and, consequently, the construction of the model of factors associated with the lifetime use of marijuana. Similarly, the question "Have you used marijuana frequently in the past 30 days?" (Yes/No) was essential to identify the dependent variable and, consequently, the construction of the model of factors associated with frequent use of marijuana. A descriptive analysis was conducted to investigate the bivariate relationships between outcome and epidemiological covariates drawn from the epidemiological form data, using Chi-square test. All the potential factors with probabilities of p ≤ 0.2 were examined and included in the final models for distinct use of marijuana, using backward stepwise multiple logistic regression. Multiple logistic regressions were then run to determine the association of each factor with lifetime use, and frequent use of marijuana.
Various possible types of interactions were evaluated to determine how they might improve the final models. The fit of the final model was assessed using the Hosmer-Lemeshow (HL) goodness-of-fit test. A p-value < 0.05 was considered significant in all analyzes. Finally, statistical analyzes for the construction of epidemiological models for adolescents who marijuana use was conducted using the SPSS 23.0 software (IBM, USA).
All participants were included after the informed and written consent of the parents/guardians. This study was approved by the Human Research Ethics Committee of the Instituto de Ciências da Saúde of the Universidade Federal do Pará, Brazil (CAAE 0103.0.073.000-10).

Results
In 2019, 1082 students were enrolled in high schools in the city of Soure. In total, 838 filled out the data collection forms. However, the information provided by one hundred and two students was excluded from this study: ninety-four for being over 19 years of age and eight students for indicating the use of fictitious drugs. Also, another two hundred and fortyfour students did not attend the school on the scheduled date for data collection or simply refused to participate in the study.
Thus, this study consisted of data provided by 736 adolescent students (68.0% of total students).
Most adolescents were female (53.7%), aged 14 to 16 years (48.5%), studied in the morning shift (75.7%), belonged to the family with low monthly income (up to 1 Brazilian minimum wage = R$ 998.00, approximately U$ 190) (71.9%), and some of them reported the need to study and work in informal activities (25.5%). Some adolescents also reported the participation of their parents or guardians in school life (always = 31.1%), and they talked about the use of psychotropic drugs (always = 44.3%). Also, the use of psychotropic drugs (licit and/or illicit) by parents or guardians was registered, as well as the use of these substances by friends of adolescents (69.2%). More specifically, many adolescents rated the family structure as excellent (61.7%) and that they were not involved in conflicts (68.8%). None of the participants claimed to have participated in school activities to discuss and prevent the use of psychotropic drugs. All characteristics of the participants in this study can be seen in table 1.
In total, 130 (17.7%) adolescents have used marijuana at least once in their lives, and another 52 (7.1%) have used marijuana frequently in the last 30 days. In the first group, the average age of marijuana use was 16.3 years ( 2.5). On the Research, Society andDevelopment, v. 10, n. 14, e499101422387, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i14.22387 6 other hand, the average age of frequent use of marijuana was 15.6 years ( 2.1). Among adolescents who frequently used marijuana, the use of alcohol (34.6%), tobacco (17.3%), and crack-cocaine (7.7%) were also reported. These rates related to marijuana use by adolescents are also shown in Table 2.
The bivariate analysis identified four factors associated with marijuana use in life. Multivariate analysis identified the association of the same factors with marijuana use in life. The values of the association of marijuana use in life with "over 16 years", "parents use psychotropic drugs (licit and/or illicit)", "friends use psychotropic drugs (licit and/or illicit)", and "everyone + most friends use psychotropic drugs (licit and/or illicit)" are shown in table 3. The HL test indicated that the final model (HL 2 = 1.9; p = 0.6) had a good fit. Regarding the frequent use of marijuana, bivariate and multivariate analysis identified six factors. The values of the association of frequent use of marijuana with "more than 16 years", "parents who do not participate in school life (never + sometimes)", "parents who do not talk about psychotropic drug use (never + sometimes)", "friends use some psychotropic drugs (licit and/or illicit)", "everyone + most friends use psychotropic drugs (licit and/or illicit)", and "family structure (self-classified) bad" are shown in table 4. The HL test indicated that the final model (HL 2 = 5.2; p = 0.4) had a good fit.

Discussion
Despite the conduct of several surveys on the use of psychotropic drugs by students in large Brazilian cities, there are still gaps to be filled, especially in areas of difficult access and low human development index, such as the Amazon region.
This study identified relevant information on marijuana use among adolescent students in the city of Soure, located in the Marajó Archipelago, one of the areas that have the lowest human development rates in Brazil. In this scenario, the use of marijuana among adolescents was higher than that registered in large Brazilian cities, with a clear influence of behavioral patterns of the coexistence nucleus, such as parents, relatives, and friends.
In Brazil, the average rates of marijuana use in life and the last 30 days among adolescent students were 5.7% and 0.3%, respectively. Approximate values also recorded in higher cities in northern Brazil: Belém (4.2% and 0.1%), Boa Vista (3.8% and 0.1%), Macapá (2.7% and 0.0%), Manaus (5.3% and 0.1%), Palmas (3.9% and 0.3%), Porto Velho (3.8% and 0.2%) and Rio Branco (3.6% and 0.1%) (CEBRID, 2010). Also, marijuana use is less than 5% among adolescent students in some     In this sense, the social environment must be considered because it is a potential epidemiological factor related to the use of psychotropic drugs, i.e., it can prevent or encourage the continued use (Chadi et al., 2018;Libuy et al., 2020;Schenker, & Minayo, 2003). Families structured, which transmit safety and behavioral examples for a healthy life, have been associated with a lower prevalence of psychotropic drug use among adolescents (Soldera et al., 2004;Tavares et al., 2004;Terzic et al., 2013;Wang et al., 2009). On the other hand, the rates of use of psychotropic drugs are higher in a coexistence environment (with parents, relatives, and friends) in which health risk behaviors are commonly performed, such as the frequent use of alcohol, tobacco, marijuana and/or cocaine by friends and family of adolescents (Alcantara et al., 2017;Bernardy, & de Oliveira, 2010;Costa et al., 2020;Elicker et al., 2015;Ferreira, & Sousa Filho, 2007;Furtado et al., 2017;Silva, 2016).
Adolescents of parents who use marijuana are 2.5 to 4.4 times more likely to use marijuana, and 1.8 to 2.75 times more likely to use alcohol, compared to adolescents of parents who do not use marijuana. In addition, adolescents of parents who use marijuana are more likely to use tobacco, report more favorable opinions about marijuana, and have reduced school performance (lower grades) (Epstein et al., 2020).
In this study, the behavioral pattern of the adolescents' coexistence nucleus (parents, family, and friends) was critical for the use of marijuana in life. Also, the lack of interference or reduced participation of the family in the adolescent's life, the inadequate family structure, and the continuous example of friends were associated with the frequent use of marijuana. The family and the school are essential institutions for strengthening self-esteem and promoting the health of adolescents, specifically related to the perception of limits and health risks, such as frequent use of marijuana and other psychotropic drugs.
With the support of these two institutions, adolescents can assess and address the common challenges of their age, relieve tension and curiosity, and make safe choices for your health autonomously (Costa et al., 2020;Schenker, & Minayo, 2003;Silva, 2016).
These findings point to the urgent need for key actions to prevent and control the use of marijuana and other psychotropic drugs by adolescents. A school is an initial and favorable place for the development of regular actions to promote the health of a range of people, with different characteristics. Thus, prevention of the use of psychotropic drugs can be encouraged through actions within the classroom that allow reflection and adoption of healthier choices. This will allow establishing clear parameters for the health promotion of adolescents and their families, as well as reducing the field of uncertainties related to the use of psychotropic drugs. Another very important factor related to this approach is to allow space for adolescents to express themselves, get involved in new proposals, share problems, and look for solutions. A school that includes, brings together, contributes to the development of self-esteem, autonomy, and the perception of limits and risks.
These characteristics are important and must be considered for preventing the use of psychotropic drugs. An example of this intervention was done with adolescents in a public school in the Brazilian municipality of Muriaé (Souza et al., 2015).
Through regular workshops based on the integral promotion of health and the creation of unique modes of care, the prevention of the use of alcohol and other drugs was discussed by adolescent students and specialized professionals (teacher, psychologist, nurse, and social worker). The harm reduction strategy was adopted aiming to reduce losses resulting from the use of alcohol and other psychotropic drugs, without having abstinence as the only way out. In this sense, the adolescents' autonomy was valued through activities that considered their experiences, taking active, conscious, and political positions during the discussions (Souza et al., 2015).
Another key intervention to be considered is the need to care for the families of adolescents who regularly use marijuana and other psychotropic drugs. It is very important to strengthen families, through the establishment of social policies and health promotion, with consistent actions and programs that meet their integral needs, in a perspective of totality, regardless of the way they are configured, to face the inherent issues frequent and abusive use of psychotropic drugs. Public policies and strategies related to mental health, especially aimed at assisting people who use psychotropic drugs, have been implemented in Brazil in the last decades (Machado, & Boarini, 2013). This has enabled the articulation and integration of interventions for treatment, recovery, harm reduction, social, and occupational reintegration with the national health system (Sistema Único de Saúde (SUS)) and the national social assistance system (Sistema Único de Assistência Social (SUAS)) for users and their families in large Brazilian cities. The absence of structures and services for people who use psychotropic drugs and their families in the Amazon region, especially in remote areas such as the cities of the Marajó Archipelago, have hampered access, treatment, and recovery for people who use psychotropic drugs, and all those impacted by chemical dependence in this Brazilian region (Andrade et al., 2017;Oliveira-Filho et al., 2020;Piauiense et al., 2020). This study has limitations and should be considered. The number of adolescents who frequently use marijuana and other psychotropic drugs may be underestimated. About 20% of adolescents did not participate in this study (n = 244) either because they missed classes on data collection days or simply because they did not feel comfortable providing information about the use of marijuana and other drugs. Despite the authors' efforts to ensure safe information, the data obtained on marijuana use were self-reported and may contain response bias.

Final Considerations
This study identified high rates of marijuana use among adolescent students in the municipality of Soure, located in a remote area of the Amazon region, and with a very low human development index. The use of marijuana was associated with several factors related to the adolescents' coexistence nucleus (parents and friends), the lack of interference or reduced participation of the family in the adolescent's life, and the inadequate family structure. Measures to control and prevent the use of marijuana and other psychotropic drugs are necessary and should be available to adolescents, their friends, and family in this Brazilian municipality. In the future, epidemiological studies using other psychotropic drugs as outcomes should be conducted and disseminated, as well as small actions to promote health and prevent the use of psychotropic drugs in schools. The neglect for the epidemiological scenario presented in this study can contribute to the reduction of school performance and neuropsychological decline of adolescents, and the increased risk of chemical dependence on marijuana and other psychotropic drugs. The provision of social justice and equity is extremely necessary. These high rates of marijuana use can be reduced through human, integral and solidary development, respecting local identity and culture, offering knowledge, dignity, autonomy, and social well-being.