Sm. Sci. Med. Vol. 23. No. 7. pp. 717-719. 1986 Printed in Great Britain. XI1 rights reserved

027:.9536,86 S3.W + 0.00 Copyright % 1986 Pergamon Journals Ltd

PATTERNS OF DRUG USE AMONG THE PAST DECADE

ADOLESCENTS:

REGINALD G. SMART and EDWARD M. ADLAF Addiction Research Foundation, 33 Russell St, Toronto, Ontario. Canada MjS 2S1

Abstract-In recent years investigators have examined patterns of drug use among various populations. None, however, have examined the changes in drug patterns over time. The present study analyzed changes in patterns of drug use among a population of drug-using adolescent students in Ontario (Canada). Three cross-sectional surveys conducted in 1968, 1977 and 1985, which sampled students in grades 7, 9, 1 I and 13 (ranging in age from 12 to 20 years), were employed. Using eight substance types as measures (cannabis, barbiturates, stimulants, tranquillizers, inhalants, LSD, heroin and others), four general patterns were constructed: single-drug psychotherapeutic users; multiple psychotherapeutic users; exclusive cannabis users; and multiple-illicit. The findings indicated the following: (i) a significant decrease in the representation of both exclusive and multiple psychotherapeutic users between 1968 and 1977; (ii) a significant increase in exclusive cannabis users between 1968 and 1977, and following this a decline into 1985; and (iii) a significant increase in multiple illicit users between 1968 and 1977. In all a major shift from a psychotherapeutic-illicit dichotomy to an overwhelming illicit pattern has occurred during the past decade.

Key words-drug use, polydrug use

A common goal in the scientific enquiry of drugs and their use involves the description and character- ization of users. In recent years, the field of drug research has seen a wider application of statistical methods to achieve this goal. These multivariate techniques, in particular cluster and factor analytic models, have the advantage of describing empirically based typologies, that is, the objective description of ‘natural groups’ [l]. Excellent examples of these approaches applied to drug-using groups include the studies by Braucht et al. [2] and Douglass et al. [3]. Although these techniques are important tools in our understanding of drug users and their characteristics, we have virtually no information on how patterns of drug use are changing over time. Those few studies that collect trend data, for example [4-6], have not addressed this issue. This paper reports what is be- lieved to be the first attempt to examine how patterns of drug use among Ontario students have changed since the drug era of the late 1960s.

More than a decade ago, Smart and Whitehead [A characterized adolescent drug users as follows: (i) medicinal users, those who report the exclusive use of a psychotherapeutic substance, specifically stimulant, barbiturate or tranquillizer users only; (ii) mood- modifiers, those who report using two or more psycho- therapeutic substances only; (iii) cautious trippers, those who report using cannabis only; and (iv) incautious trippers, those who report using one or more illicit drugs (with or without psychotherapeutic substances). These typologies or patterns were not empirically derived, but intended to make sense of the licit-illicit continuum. From representative student samples they indicated that the most common types of drug users were incautious trippers, followed

The views expressed in this paper are those of the authors and do not necessarily reflect those of the Addiction Research Foundation.

by medicinal users, cautious trippers and mood- modifiers. Yet we do not know how or if these patterns have changed over time when applied to a very different cohort of today’s students.

The purpose of the present study, then, is to examine the changes in drug-use patterns among cohorts of adolescent drug-using students spanning more than a decade; specifically, the changes between the years 1968, 1977 and 1985. Particular attention will focus on the patterns related to polydrug and cannabis use.

.MEl-HOD

Sample

The data analyzed here were selected from three independent cross-sectional drug use surveys con- ducted in 1968 [8], 1977 [9] and 1985 [6, IO]. Ontario (Canada) students enrolled in grades 7 (aged 12-13 years), 9 (1415 years), 11 (16-17 years) and 13 (18-19 years) were sampled for all surveys; however, these populations differed in that the 1968 sample was restricted to students in Metro Toronto, whereas the 1977 and 1985 samples were both province-wide. Consequently, the present analysis precluded the selection of Metro students only from the province- wide samples designs since the number of cases would be too small to ensure reliable estimates of pattern structure. However, to be sure that the Ontario estimates were reflective of the Metro Toronto sample, we compared patterns between Metro and non-Metro students. No statistically significant differ- ences were found. Consequently, we felt confident that the Ontario and Toronto estimates were similar enough to warrant their combination.

Variables and data

In order to compare patterns across time, drug use measures were constructed for the following eight

717

718 REGKALD G. SMART and EDWARD M. ADLAF

categories: cannabis, including marihuana, hash and hash oils; barbiturates; stimulants; tranquillizers; inhalants; LSD; heroin; and, other illicit substances- PCP, cocaine, speed (i.e. metamphetamines) and other hallucinogens. The use of alcohol and tobacco were ignored, thus, students who reported the exclusive use of alcohol and tobacco were excluded, resulting in sample sizes of 1260 in 1968; 1961 in 1977, and 1381 in 1985.

To ensure comparability to the 1968 sample, no distinction was made between medical (use on medi- cal advice) and non-medical use (use without medical advice). In addition, the 1968 sample measured drug use in the past 6 months prior to the survey; in contrast, the 1977 and 1985 samples measured use during the last 12 months prior to the survey. Our experience, however, suggests that these two measures do not significantly differ with regard to estimates of prevalence.

In order to construct drug use patterns, a type by number of drugs matrix was computed allowing for the easy identification of pattern categories. Though not ideal, the patterns constructed by Smart and Whitehead with the 1968 data are used in order to obtain the longest possible trend comparison. In total six patterns of drug use are examined across time. The first three reflect the exclusive use of psycho- therapeutic substances, namely stimulants only, barbiturates only and tranquillizers only. The sum of these three patterns reflects the larger category of exclusive psychotherapeutic use. The fourth pattern multiple psychotherapeutic use reflects the propor- tion of students reporting the use of two or more psychotherapeutic substances, but no use of any other substance type. The fifth pattern reflects exclusive cannabis use; thus, these students reported using only cannabis (recall that alcohol and tobacco use have been ignored). Finally, the sixth pattern, other illicit use, refers to the percentage of students who reported using one illicit substance (other than cannabis) or more illicit substances. Thus a student who reported use of only one illicit substance type, for instance, LSD would be represented in this pattern. Also, cannabis users who reported using any other substance would also fall into this category.

‘Although strictly speaking medical use and inhalant use are included in this pattern, and thus licit virtually all students represented in this pattern reported use of one or more illicit drugs. Additionally, for the 1977 and 1985 samples the multiple-illicit pattern was further partitioned to indicate: (i) exclusive use of an illicit drug other than cannabis; and (ii) use of two or more illicit drugs. Pairwise r-tests for proportions were employed to determine the levels of significance between percentage estimates.

RESULTS

As seen in Table 1, significant changes, especially between the years 1968 and 1977, occur among all patterns. Beginning with the psychotherapeutic users, we find significant declines for both psycho- therapeutic drug-user patterns. The proportion of drug-using students reporting exclusive use of one of the psychotherapeutic drugs declines from 38.5% in 1968 to 20.3% in 1977 (P < 0.001). This declining

Table 1. Percentage reporting pattern of drug use by year of survey

Year

1968 1977 1985 Pattern of use (.V=l260) (.V=l961) (,Y=1433)

Psychotherapeutic drug use I. Exclusive psychotherapeutic

use: 38.5. 10.3 21.0 (i) Stimulants only 10.5’ 4.4) 7.3

(ii) Barbiturates only 3.2’ 9.2 8.8 (iii) Tranquilizers only 24.a* 6.7t 4.9

2. Multiple psychotherapeutic use (users of two or more psycho- therapeutic drugs only) 8.3. 5.S 4.1

Other drug use 3. Exclusive cannabis use 16.8’ 23.7t 20.4 4. Other illicit use (users of one

or more illicit drugs) 36.4’ 50.4 52.3 (i) Users of two or more

illicit drugs - 40.6t 50.1 (ii) Users of one illicit drug - 9.u 2.2

*I968 estimate is statistically significant from 1977 at least to the P < 0.05 level.

tl977 estimate is statistically significant from 1985 at least to the P < 0.05 level.

trend however, does not continue into 1985, as seen in Table 1. Examining changes in the exclusive use of specific substances we see that exclusive users of stimulants and tranquillizers decrease between 1968 and 1977 (P < 0.001 in both cases). The proportion of students using tranquillizers continues its down- ward trend into 1985, declining from 6.7% in 1977 to 4.9% in 1985 (P c 0.05). However, the proportion of stimulant users increases significantly from 4.4% in 1977 to 7.3% in 1985 (P < 0.001). Exclusive barbiturate use increases from 3.2% in 1968 to 9.2% in 1977 (P < O.OOl), while remaining relatively stable between 1977 and 1985. With respect to multiple- psychotherapeutic use, we find a moderate decline between 1968 and 1977 (8.3% versus 5.5%, respec- tively; P -c O.OOl), and stability between 1977 and 1985.

In contrast to the prominent decrease in most psychotherapeutic patterns, the proportion of illicit drug users increases significantly over the study period. As we see in Table 1, in 1968, 16.8% of the drug users report using cannabis exclusively, in contrast to 23.7% in 1977 (P < 0.001). Between 1977 and 1985, however, this proportion drops to 20.4% (P < 0.01). Increases in the other illicit use are like- wise significant between 1968 and 1977 (36.4% versus 50.4%; P < O.OOl), but not between 1977 and 1985 (50.4% versus 52.3%).

Although we find no substantive change between 1977 and 1985 in the representation of other illicit users, when this category is further partitioned into exclusive-illicit users (only one drug) and multiple- drug users (two or more illicit drugs) significant trends emerge. These comparisons, unfortunately, were obtainable only between the years 1977 and 1985 since the raw data for the 1968 sample were unavailable to allow for this separation. AS we see in Table 1, the proportion of drug users reporting use of two or more illicit substances increases from 40.6% in 1977, to 50.1% in 1985 (P <O.OOl). Correspondingly, the proportion of users reporting exclusive use of an illicit drug decreases from 9.8% in 1977 to 2.2% in 1985 (P < 0.001). Thus, these data

Patterns of drug use among adolescents 719

indicate that over the past several years there has been a substantial increase in the prevalence of reported polydrug use among illicit drug using students.

SUMMARY AND DISCLSSION

We began this paper by noting the lack of empirical data confirming commonly held perceptions of the changing patterns of drug use among drug-using adolescents during the past decade, in particular, that relating to an increase in polydrug use. Consequently, we replicated an a priori drug-use typology earlier suggested by Smart and Whitehead. The main results were as follows: (i) a significant decline in the propor- tion of both exclusive and multiple psychotherapeutic drug users between 1968 and 1977 only; and con- versely, (ii) a significant increase in the proportion of both exclusive cannabis users and other illicit drug users between 1968 and 1977. Between 1977 and 1985, however, the proportion of users reporting exclusive cannabis declined significantly.

It is likely that the growth in the use of cannabis during the late 1970s partially accounts for the decline in the representation of the two psycho- therapeutic-use patterns. Since users of one drug are generally more likely to use another, exclusive and multiple-psychotherapeutic users may have been at the greatest risk to adopt cannabis (and other drugs) into their repertoire at a time when cannabis was becoming more available. Alternatively, in 1968 psychotherapeutic pill use may have been perceived as conventional behaviour, specifically an expression of adult role models, and with time was supplanted by the growing conventionality of cannabis use. In sum, by one means or another, intercohort changes have led to a shift from the largely dichotomous psychotherapeutic-illicit patterns to that based more predominately on illicit ones.

In addition, the growth in cannabis use is reflected in a significant increase since 1968 in the proportion of drug-using students reporting the sole use of cannabis. The increase in this type of cannabis use undoubtedly reflects the growth in its availability and the weakening of negative prescriptions toward its use. Interestingly enough, similar trends emerge among high-school seniors in the United States.

Johnston et al. [4] found that the proportion of seniors reporting cannabis as the only illicit drug used in the prior year was 19% in 1975; 25% in 1977; and 18% in 1984. Our data exhibit a declining trend between 1977 and 1985 similar to that indicated by the American survey.

At present, virtually all users of illicit drugs, other than cannabis users, are polydrug users. Moreover, our expectation that the prevalence of polydrug use has increased during the last few years is confirmed by these data. Between 1977 and 1985 we find a significant increase in the proportion of adolescent drug users reporting illicit multiple drug use (two or more drugs).

I.

2.

3.

4.

5.

6.

7.

8.

9.

10.

REFERENCES

Lorr M. Cluster and typological analysis. In ~aro Analysis Strategies and Designs for Subsrance Abuse Research (Edited by Bentler P. M. and Lettieri D. J.), pp. 103-124. National Institute on Drug Abuse, Rock- ville, 1976. Braucht G. N., Kirby M. W. and Berry G. J. Psycho- social correlates of empirical types of multiple drug abusers. J. con. clin. Psychol. 46, 1463-1475, 1978. Douglass F. M., Khavari K. A. and Farber P. D. Three types of extreme drug users identified by a replicated cluster analysis. J. abnorm. Psycho/. 89, 240-249, 1980. Johnston L. D., Bacham J. G. and O’Malley P. M. Use of Licit and Illicit Drugs bv America’s High School Srudenrs 1975-1984. National Institute on Drug Abuse, Washington, 1985. Blackford L. Summary Report-Surveys of Student Drug Use. San Mateo County, California, 1977. Smart R. G., Adlaf E. M. and Goodstadt M. S. Alcohol and Other Drug Use Among Ontario Srudents in 1985 and Trends Since 197% Addiction Research Founda- tion, Toronto, 1985. Smart R. G. and Whitehead P. C. A typology of high school drug use: Medicinal usage, mood-m&ificat& and tritmina. Int. 1. Addict. 7. 735-738. 1972. Smart k’. 6. and Jackson D. 2 Preliminary Report on the Attitudes and Behaviour of Toronto Sludents in Relufion to Drugs. Addiction Research Foundation, Toronto, 1969. Smart R. G. and Goodstadt M. S. Alcohol and Drug Use Among Ontario Students in 1977: Preliminary Findings. Addiction Research Foundation, Toronto, 1977. The survey was fielded by the Survey Research Centre, Institute for Social Research, York University, Ontario.

S.S.M 21 7-F