The prevention of drug abuse by young people: an argument based on the distribution of drug use
Author: Reginald G. SMART , Paul WHITEHEAD , Lucien LAFOREST
Pages: 11 to 15
Creation Date: 1971/01/01
Paul WHITEHEAD 2
Lucien LAFOREST 3
Addiction Research Foundation
Concern about adolescent drug use has reached epidemic proportions in North America even if drug use itself has not. Documentation of this concern does not need to be extensive and various reviews have been made of the extent of drug use in various populations ( [ 1] , [ 2] , [ 3] ). Although there is a tendency to see all illicit drug use as harmful or dangerous many professionals are chiefly concerned with preventing excessive use or demonstrably harmful abuse. So far, preventive programmes suggested ( [ 4] ) have been based upon education about the harmful effects of drugs or the development of counter cultures ( [ 5] ). Neither of these approaches has been closely derived from scientific evidence about the nature of youthful drug use and neither has been shown to be effective as yet. Clearly, there is a need to develop preventive approaches to drug abuse in North America. The purposes of this paper are to (i) present data concerning the distribution of drug use in a variety of high school populations and (ii) indicate the implications which these distributions have for the prevention of drug abuse.
Most studies of the extent of drug use report data for numbers of users and non-users and, occasionally, data for the frequency of use. The latter is often described in terms of a few categories and the distribution of drug use can rarely be appreciated from such studies. There is a need to know the distribution of drug use among drug users taking all of their drugs together. It is of prime interest to understand the basic distribution of drug use and to know whether different drugs have similar or different distributions. It will be argued here that value of many educational or legal controls on drug consumption will depend on the nature of the underlying distribution and on whether it varies with geographic and cultural contexts.
1.Associate Research Director, Addiction Research Foundation, Toronto.
2. Assistant Professor, Dept. of Sociology, University of Western Ontario.
3.Research Associate, Dept. of Social Medicine, Sherbrooke University.
There are two basically different expectations about the distribution of drug use in young persons. Certain clinical reports and descriptions of heavy drug users suggest that the distribution might be bi-modal, with most persons average or "normal" in drug use and a smaller high point in the heavy or "abusing" category. On the other hand, De Lint and Schmidt ( [ 6] ) and Ledermann ( [ 7] ) have found that the consumption of alcohol in several countries describes a log normal distribution. This is a curve in which there are many infrequent users, fewer moderate users, and even fewer heavy users. The implications arising from a knowledge of the distribution of drug use are considerable. If "normal" and "abusing" drug use can be differentiated then educational and preventive programmes need only focus on preventing drug abuse. However, if there is no clear distinction between users and abusers, then all forms of drug use may have to be modified in order to prevent "abuse ".
This paper reports the distribution of drug use found in five large scale studies of high school students. These studies were made in four different places over a period of three years and all involve large adequately chosen samples. The drug use questions employed were tested against a second method of estimation which gave similar results ( [ 8] ). Considerable confidence can be held in the reported data on drug use.
The details of the conduct of the five studies are described in the original reports ( [ 3] , [ 8] , [ 9] , [ 10] , [ 11] ). Only certain broad similarities and differences will be noted here. All five studies were based on large samples of high school students, chosen at random from the total population. The sample sizes were 6,447 in Toronto (1968), 8,568 in Toronto (1970), 5,900 in the Niagara Counties, 4,501 in Montreal and 1,606 in Halifax. In total, some 27,022 students were involved in these studies. The 1968 Toronto sample was chosen to include 120 students from grades 7, 9, 11 and 13 in 20 per cent of the high school districts in Toronto. A similar practice was used in 1970 except that grade 6 was included. In Montreal, grades 8, 10, 12 and 14 were sampled to
provide proportionate sampling of the two languages (French and English) and the two religious groups (Catholic and Protestant). The sample in Halifax was a 25 per cent random sample of classes from grades 7, 9, 11 and 12 only. Virtually, the same questions about drug use were asked in all five studies and similar tables and statistical analyses have been performed on the data. Four of the studies were done at the same time of the year (April), and the Niagara Counties study was done in February. The first Toronto study was done in 1968. The Halifax and Montreal studies in 1969 and the second Toronto study and the Niagara Counties study in 1970.
It is obvious that the studies have some differences in methodology. The sampling of grades, and of the total populations was not identical, although each sample is representative of its own population. The grades sampled were slightly different, for in Halifax and Montreal, there is no grade 13. In addition, the studies were done during three different years. Nevertheless, it is rare to find surveys in widely spaced areas which are similar in any important characteristics.
Data from five high school drug use studies were employed in determining the fit of the log normal curve to the drug scores of users. The drugs involved were alcohol, tobacco, glue, marihuana, LSD, other hallucinogens, opiates, tranquillizers, stimulants and barbiturates. All five studies enquired about the use of 10 drugs and for each drug these were four categories related to frequency of usage. These were 1-2 times, 3-4 times, 5-6 times, and 7 plus times in the past months. Responses in each category were assigned to the mid-point so that the drug use frequencies were 1.5, 3.5, 5.5 and 7.5 respectively. Each drug user then had a score for each of the drugs used (1.5 to 7.5) and a total score varying from 1.5 to 75.0 comprising the sum of the scores for the 10 individual drugs. Only the total scores were used here. Persons who were not users of any drugs obtained scores of zero and were not included.
The distribution of drug use scores were compared with the expected values derived from the log normal expectancy. The methods used were described by Croxton and Cowden ( [ 12] ). Merely by gross inspection of figures 1-5 it can be seen that these five studies have produced similar distribution of total drug use. Croxton and Cowden also describe a skewness test to statistically determine the goodness of fit. They suggest that if skewness measures are .20 or lower then log normality has been achieved. The skewness measures are .01, .03, .11, .18 and .20 and all within the acceptable limits. All of the distributions are of the same shape, although the means vary from 6.51 to 11.16. It should be noted that although drug use increased substantially from 1968 to 1970 in Toronto, the shape of the curve did not change.
Logarithmic normal curve fitted to drug use scores of Toronto high school drug users, 1968
It has been shown that five scale studies generate drug usage data describing a log normal distribution. These studies were made in widely separated areas and they range over a three-year period. The studies in Toronto in 1968 and 1970 showed that drug usage had increased substantially but the character of the distribution remained the same. These data strongly suggest that, at least for high school students, the basic shape of the drug use distribution will be similar in different places and at different times. Of course, it would be well to have similar data from different cultures and efforts are being made to obtain them. In this connexion it should be remembered that the log normal character of the alcohol consumption data holds for Canada, Finland, the United States and France ( [ 6] ).
Logarithmic normal curve fitted to drug use scores of Toronto high school drug users, 1970
Logarithmic normal curve fitted to drug use scores of Halifax high school drug users, 1969
Logarithmic normal curve fitted to drug use scores of Montreal high school drug users, 1969
Logarithmic normal curve fitted to drug use scores of Niagara Countries high school drug users, 1970
The findings of this study clearly show that the distribution of drug use is not bi-modal but smooth and continuous. This suggests that there is no clear differentiation of drug users into "normal users" and "abusers" on consumption alone. Any definition of drug abuse in terms of extent of use must be arbitrary. It will be necessary then to define points in the distribution above which some physical or psychological pathology occurs. Schmidt and De Lint ( [ 13] ) have shown that about 15 cl. of absolute alcohol is the average amount consumed by alcoholics per day and 10 cl has been shown to be associated with various physical pathologies such as liver cirrhosis ( [ 14] ). Some such damaging level of drug consumption will also have to be determined in order to give meaning to the term "drug abusers ". Studies being made of hippies and young persons who have rejected society and its norms (" drop-outs ") show that almost all of them would fall in the top 10 per cent of the distribution of drug use but these studies have not been completed.
The log normal distribution also provides some interesting predictions about the prevention of "drug abuse" however that comes to be defined. Let us suppose for a moment that the log normal distribution describes drug use in a variety of cultures and at different times. This suggests that the distribution of drug use will retain the same character after successful efforts have been made to increase or decrease consumption. This has already been shown to hold for the increase in Toronto from 1968 to 1970. What this means is that the numbers of drug abusers cannot be reduced unless per capita consumption of drugs falls. There will be no way to cut off the "heavy" or "abusing" part of the distribution. In order to reduce "drug abuse" it will be necessary to lower the whole consumption curve and lower per capita consumption. Any successful effort to reduce drug abuse will mean that most people in the population will have to use fewer drugs such as alcohol, tobacco, and psychoactives. Per capita consumption can only be reduced when many people modify their consumption habits.
Efforts at the prevention of drug abuse by young people probably depend upon convincing or forcing people in general to use fewer mood-modifiers. This process may take several generations to achieve and it may depend upon educational programmes or public policies as yet untried.
Of course, the problems of drug use and abuse are not confined to young persons. A recent study of highschool drug users ( [ 3] ) showed that users of illicit drugs often had parents who were heavy users of psychoactive drugs. Our work suggests that it will be necessary for parents and adults in general to reduce their over-all drug use in order to prevent drug abuse by adolescents of the next generation. This general position predicts that changes in drug laws or public education campaigns against drug abuse will fail unless they reduce drug consumption by the public at large.
Concern about the present drug abuse problem in young people generates a need for preventive measures. Effective prevention will depend upon knowing the distribution of the frequency of drug use in various populations. If normal drug use can be differentiated from "drug abuse ", then preventive programmes need only focus on abuse. However, data from five large scale studies of high school drug use indicate that the log normal expectancy describes the frequency of use. This curve is one in which there are many light users, fewer moderate users, and even fewer heavy users. The curve is continuous and there is no clear differentiation into users and abusers. This means that the numbers of drug abusers cannot be reduced unless per capita consumption of drugs falls. People in general will have to use fewer drugs in order to prevent adolescent drug abuse in the next generation.
Berg, Dorothy, "Extent of illicit drug use: a compilation of studies survey and polls ". Unpublished report. Bureau of Narcotics and Dangerous Drugs, 1969.002
Smart, R. G., and Fejer, Dianne, "The extent of illicit drug use in Canada: a review of current epidemiology ". Addiction Research Foundation Substudy, 2-7 & Jo-69, 1969.003
Smart, R. G., Fejer, Dianne, and Alexander, Eileen, "Drug use among high school students and their parents in Lincoln and Welland Counties ". Addiction Research Foundation, Substudy 1-7 & Jo & AL-70, 1970.004
Smith, Kline, and French, Drug Abuse: Escape to Nowhere - A guide for Educators, National Education Association, 1967.005
Ungerleider, J. T., Information about Project D.A.R.E.; Drug Abuse research and education. Foundation, Los Angeles, 1969.006
De Lint, J. E., and Schmidt, W., Quart. J. Stud. Alc ., 1968, 29, 968-973.007
Ledermann, S., "Alcool, alcoolisme, alcoolisation. Données scientifiques de caractère physiologique, économique et social (Institut national d'études démographiques, Travaux et Documents, Cahier (No. 29), Paris, Presses universitaires de France, 1956.008
Smart, R. G., and Jackson, D., A Preliminary Report on the Attitudes and Behaviour of Toronto Students in Relation to Drugs ", Addiction Research Foundation, Toronto, 1969.000
The Prevention of drug abuse by young people: an argument based on the distribution of drug use 15009
Whitehead, P. C., Drug use among Adolescent Students in Halifax , Youth Agency, Halifax, Nova Scotia, 1969. Support for this study was received from the following sources: the Youth Agency of the Province of Nova Scotia, the Grants Division of the Department of National Health and Welfare, and the Dalhousie University Faculty of Graduate Studies Research Development Fund.010
Laforest, L., "The Incidence of Drug Use among High School and College Students of the Montreal Island Area " , Office de la prévention et du traitement de l'alcoolisme et des autres toxicomanies, Quebec City, 1969.011
Smart, R. G., and White, J., The Extent of Drug use in Metropolitan Toronto Schools: a Study of Changes from 1968 to 1970 . Addiction Research Foundation, Toronto, 1970.012
Croxton, F. R., and Cowden, D. J., Applied General Statistics , Englewood Cliffs, Prentice-Hall, 1955.013
Schmidt, W., and De Lint, J. E., Estimating the Prevalence of Alcoholism from Alcohol Consumption and Mortality Data . In press, Quart. J. Stud. Alc., Sept. 1970.014
Pequignot, G., Bulletin de l'Institut national d'hygiène (Paris), Tome 13 (3), July/Sept., 1958.