Luciana Morita Ishihara Drug Use in College Students: a 13-year Trend

Uso de drogas entre alunos universitários: tendências em 13 anos ABSTRACT OBJECTIVE: To analyze drug use trends among college students in 1996, 2001 and 2009. METHODS: A cross-sectional epidemiological study with a multistage stratifi ed cluster sample with 9,974 college students was conducted in the city of São Paulo, southeastern Brazil. An anonymous self-administered questionnaire was used to collect information on drug use assessed in lifetime, the preceding 12 months and the preceding 30 days. The Bonferroni correction was used for multiple comparisons of drug use rates between surveys. RESULTS: There were changes in the lifetime use of tobacco and some other drugs (hallucinogens [6. Differences in the use of other drugs over the 12 months preceding the survey were also seen: reduced use of inhalants [9.0% to 4.8%] and increased use of amphetamines [2.4% to 4.8%]. There was a reduction in alcohol [72. CONCLUSIONS: Over the 13-year study period, there was an increase in lifetime use of tobacco, hallucinogens, amphetamines, and tranquilizers. There was an increase in amphetamine use and a reduction in alcohol use during the preceding 12 months. There was an increase in amphetamine use during the preceding 30 days. 498 Drug use in college students Wagner GA et al Drug use in young people has heralded social and political changes that have taken place in other social spheres in European countries and the United States. It is a challenge for public policymakers as new uses of old drugs must be identifi ed for planning effective actions. Drug use among the young is greater in developing than developed countries. A Brazilian study including a sample of the general population aged 12 to 65 years in the 108 largest Brazilian cities showed that 22.8% of 7,939 respondents reported use of at least one drug (other than tobacco and alcohol) once in their lifetime. d Considering drug use by age groups, 78.6% of those aged 18 to 24 years had tried alcohol, 39.5% tobacco and 10.8% inhalants. Inhalants seem to be mostly abused by Brazilian young people. College students have deserved attention because they suffer serious consequences resulting from illicit drug use or the use

Drug use in young people has heralded social and political changes that have taken place in other social spheres in European countries and the United States.It is a challenge for public policymakers as new uses of old drugs must be identifi ed for planning effective actions.
Drug use among the young is greater in developing than developed countries.a,b,c A Brazilian study including a sample of the general population aged 12 to 65 years in the 108 largest Brazilian cities showed that 22.8% of 7,939 respondents reported use of at least one drug (other than tobacco and alcohol) once in their lifetime.d Considering drug use by age groups, 78.6% of those aged 18 to 24 years had tried alcohol, 39.5% tobacco and 10.8% inhalants.Inhalants seem to be mostly abused by Brazilian young people.[15] According to epidemiological studies conducted in the United States, about 30% of college students reported tobacco use in the preceding 30 days -a high rate considering the information available on the risks of smoking-, about 20% or less reported using marijuana, and less than 2% reported using cocaine. 6The First Nationwide Survey on the use of alcohol, tobacco and other drugs among college students carried out in 27 Brazilian state capitals e showed that 48.7% of students reported using illicit drugs in their lifetime.Marijuana was the substance most frequently used, followed by amphetamines, tranquilizers, inhalants and hallucinogens, especially regarding recent use (use in the preceding 12 months and in the preceding 30 days).Illicit drug use was greater among students from the Southern and Southeastern regions.
Similarly to the University of Michigan Monitoring the Future Study, 6 the Universidade de São Paulo (USP) carried out a study on drug use among its students aiming to identify trends and developing prevention actions and intervention strategies targeted to this population.e Three cross-sectional surveys were carried out at the USP in 1996, 2001 and 2009 1,13,e by the Interdisciplinary Group for Alcohol and Drug Studies.
The fi rst study (1996) pointed out that alcohol and tobacco were the most widely-consumed drugs, with prevalences of use of 82.5% and 25.6% in the preceding 12 months. 13Drug use was greater among males and students not living with their families.
The study conducted in 2001 revealed some changes.There was an increase in the rate of experimentation with alcohol, tobacco, marijuana and hallucinogens, 1,13 as well as increased use of tobacco, marijuana, amphetamines and inhalants over the preceding 30 days 14 among males.No specifi c increases were observed among females.
A third study was conducted in 2009 to follow up drug use changes at the USP from 1996.The present study aimed to analyze drug use trend in college students between 1996 and 2009.

METHODS
Cross-sectional epidemiological study with 9,974 undergraduate college students from a public university conducted in the state of São Paulo, Southeastern Brazil.In the fi rst two studies (1996 and 2001), the sampling unit was the student selected by means of systematic random sampling stratifi ed according to study area (Arts & Humanities, Biological Sciences, and Mathematical and Physical Sciences).The sample was selected from students enrolled in any undergraduate course through the statistical yearbook.f The primary sampling unit was the student classroom (set of students attending a given class).The sample was stratifi ed according to the study area and student classrooms were defi ned as clusters.Since a class may include students from several different colleges, the number of students per study area was totaled and the percentage corresponding to each area was calculated.The area with the highest proportion of students was defi ned as a study area.
It was assumed that the absolute difference between a given proportion obtained for the sample and the corresponding proportion for the entire population should lie within a 95% confi dence interval, with a 5% margin of error.The sample size would be equivalent to 40%, i.e., the prevalence of use of at least one illicit drug in lifetime, according to the outcomes of the fi rst wave of this series of studies (38.1% [95% CI 35.3; 41.0]).Another 20% of the sample was randomly selected to compensate for potential losses of colleges or questionnaires.
Based on a reference list of undergraduate students, 228 classrooms were selected to participate in the study (76 for each study area).This estimate was based on the planning effect which is dependent on the average cluster size and intraclass correlation coeffi cient. 13All students in each classroom selected were contacted.The total number of students was determined by the classroom selected.The sample comprised 4,759 respondents, 81.1% of the students selected in 2009 (92.6% in 1996 and 79.1% in 2001).The total sample comprised 9,974 students (2,374 participants in 1996 and 2,841 participants in 2001) (Table 1).The distribution of students in 2009 according to study area was 39.6%, 30.7%, and 29.7% from Arts & Humanities, Mathematical and Physical Sciences and Biological Sciences, respectively (Table 2).
The study questionnaire was adapted from a research instrument developed by the World Health Organization. 1,13Drug use was assessed in lifetime (experimental use "at least once in their lifetime"), in the preceding 12 months (i.e., "at least once in the 12 months preceding the survey") and in the preceding month ( i.e., "at least once in the 30 days preceding the survey").It was an anonymous self-administered questionnaire comprising 58 closed-ended questions.Students took on average 30 minutes to complete the questionnaire.
The adapted instrument assessed the profi le of the students: a) sociodemographic information (gender, age, marital status, socioeconomic status, ethnic group and others); b) undergraduate course attended  hallucinogens and ecstasy) through the description of the frequency of use in their lifetime and in the preceding 12 months and preceding 30 days.A fi ctitious drug called Relevin was included in the questionnaire to test the reliability of the responses.g A positive answer on the use of this drug invalidated the entire questionnaire, thus preventing the use of this data in the analysis.Six questionnaires were excluded from the analysis.
The data were collected between March and November 2009.Valid questionnaires were single-entered using SPSS.Quality control was performed by retyping 70% of the questionnaires.When the database was complete, consistency analyses and general checks were performed.
All statistical analyses were performed using SPSS for Windows version 13.1 and SAS version 9.2.A descriptive analysis was conducted with the 2009 data.Data was weighted as they were based on complex sampling plans.A trend analysis could not be performed due to the small number of surveys (three) and a comparison between drug use prevalence was performed using the Bonferroni method,
In the preceding year, 66.5% of the respondents passed all their subjects while 15.4% failed in one subject but not in the entire academic year.
Among sampled students, 81.5% reported the use of drugs in 2009.Among these, 92.5% used alcohol, There were changes in the lifetime use of tobacco and some other drugs (hallucinogens, amphetamines, and tranquilizers) from 1996 to 2009 (Table 3).Differences in the use of other drugs over the 12 months preceding the survey were also seen: reduced use of inhalants and increased use of amphetamines.There was a reduction in alcohol, tobacco and marijuana use and an increase in amphetamine use in the preceding 30 days (Table 4).

DISCUSSION
Over the 13-year study period, an increase in the lifetime use of tobacco, hallucinogens, amphetamines and tranquilizers was seen.There was an increase in amphetamine use and a reduction in alcohol use in the 12 months prior to the study.In the preceding 30 days, there was an increase in amphetamine use.
Female students were found to consume more tobacco, marijuana, anticholinergics and inhalants in the 30 days preceding the survey and were major consumers of tranquilizers and amphetamines in 2001. 14Among American college students, as shown in the University of Michigan Monitoring the Future Study, tranquilizer use followed a cohort behavioral pattern. 6This behavior was particular in the sample studied: there was a rising consumption from 1996 to 2009 with a tendency towards stabilization from 2001 to 2009.However, female students continued to prefer amphetamines and tranquilizers.
The popularity of amphetamines can be attributed to its ease production and ready availability, which facilitates purchase and experimentation.Data from the Brazilian population show an increase in consumption of appetite suppressants from 2001 to 2005, as previously reported among middle and high school students from 1987 to 2004. 4 These students seem to consume alcohol, tobacco, marijuana, hallucinogens and synthetic drugs more often than Brazilian college students in general.h Both experimentation and regular use of alcohol seem to have reached a ceiling, particularly among Biological Sciences and Arts & Humanities students.Alcohol is also the most widely used drug among Brazilian college students in general among both males and females.
Use is initiated at the age of 16. 5 Even though alcoholrelated disorders are more prevalent among adults over 25 years old, high-risk use is more frequent among young drinkers.Such use is associated with several negative consequences and is one of the main causes of morbidity and mortality among college students.i The most prevalent negative consequences of alcohol use are traffi c accidents, violence, sexual abuse, sexual harassment, health problems, declines in academic performance and interpersonal problems. 9,11perimentation with ecstasy has also shown an increase since 2001, the year its analysis was introduced into the research instrument.Paradoxically, there was a reduction in inhalant use for all measurements over the 13-year period, which might be related to an increased use of ecstasy or reduced consumption of inhalants by the students sampled.The change in the study design, from a stratifi ed random sample (1996 and 2001) to a cluster sample (in 2009) made the fi eldwork easier, as less time was required for data collection, analysis of results and identifi cation of drug use trends.However, the introduction of a selection method using clusters increased the variability among the respondents and reduced the accuracy that had previously been achieved in the 1996 and 2001 studies.This diffi culty was offset by an increase in sample size.Weights were assigned to the sampling units in the three studies so that the distribution of respondents in the sample would refl ect the distribution of students in the general college population, without affecting the results or their scope.Although the sample design in the 2009 study was different from the previous years, the comparability was not limited.Kish 7 (1965) stated that comparison between distinct sampling plans does not interfere with or bias the inferential analyses h Nicastri S, Oliveira LG, Wagner GA, Andrade AG.Prevalência e padrão de uso de tabaco e outras drogas (exceto álcool): estimativa de abuso e dependência.In:

Table 1 .
Sample design and total number of college student respondents in the three waves of the study.São Paulo, SoutheasternBrazil, 1996, 2001 and 2009.
(study area, academic year and term); c) academic life events (involvement with college activities and places of socialization); and d) academic performance and satisfaction about the undergraduate course chosen, among other factors.The focus of the study was to investigate undergraduate college student use of alcohol and other drugs (tobacco, marijuana, cocaine hydrochloride, crack cocaine, methamphetamines, anticholinergics, tranquillizers, opiates, barbiturates, anabolic androgenic steroids, inhalants,

Table 2 .
Sociodemographic characteristics of students by area of study.São Paulo, Southeastern Brazil, 2009.

Table 3 .
Proportion of lifetime drug use in three surveys and differences between them.São Paulo, SoutheasternBrazil, 1996, 2001  and 2009.
g Hibell B, Guttormsson U, Ahlström S, Balakireva O, Bjarnason T, Kokkevi A, et al.The 2007 ESPAD Report: Substance Use Among Students in 35 European Countries.Sweden: The Swedish Council for Information on Alcohol and other Drugs; 2009.

Table 4 .
Proportion of drug use in the 12 months preceding the survey in three surveys and differences between them.São Paulo, SoutheasternBrazil, 1996, 2001 and 2009.

Table 5 .
Proportion of drug use in the 30 days preceding the survey in three surveys and differences between them.São Paulo, Southeastern.Brazil, 1996, 2001 and 2009.
It describes a signifi cant fraction of middle and upper-middle class Brazilian youth who have high access to information and education equivalent to fi rst-world countries, maybe because São Paulo has the largest socioeconomics characteristics in Brazil.The degree and extend of psychoactive drug use among undergraduate students attending the university studied is signifi cant.Special attention should be given to alcohol consumption and increased consumption of other substances mainly in the preceding 30 days.Prevention and treatment programs targeting drug use should be expanded in Brazil.About 28.0% of the Brazilian universities have developed programs for drug use prevention and/or guidance/counseling.Of these, seven institutions have their programs reviewed, and it has been identifi ed in a single program the need for implementing curriculum components that address participatory and preventive education with regard to drug use among the students and colleges here studied.e c Andrade AG, Duarte PCAV, Oliveira LG.I Levantamento nacional sobre o uso de álcool, tabaco e outras drogas entre universitários das 27 capitais brasileiras.Brasília: Secretaria Nacional de políticas sobre Drogas; 2010.p.129-82.