The distribution of drug use in Mexico: data from a national study


Method Sampling
Questionnaire administration
Results and discussion


Author: Elena CASTRO , Zita CHAO, Reginald G. SMART,
Pages: 49 to 54
Creation Date: 1978/01/01

The distribution of drug use in Mexico: data from a national study

Zita CHAO Centro Mexicano de Estudios en Farmacodependencia, Mexico,
Reginald G. SMART, Addiction Research Foundation, Toronto, Canada.

Evidence has been accumulating to show that the distribution of drug use in several populations is continuous, unimodal and log normal. This distribution is one with many infrequent users, fewer moderate users and even fewer heavy users. If drug use is continuous and unimodal this suggests important implications for prevention, i.e. that (i) drug use and heavy drug use are related and cannot be easily differentiated and (ii) drug use in general may have to be reduced in order for heavy use to be reduced.

Studies of alcohol consumption have shown that the distribution holds for a variety of populations in countries such as the USA, Canada, Norway, Sweden, Italy, Denmark and Holland (Bruun et al., 1975). However, data on the distribution of drug use other than alcohol or including alcohol as one of many drugs have come from only three countries--Canada (Smart and Whitehead, 1972, 1973), Britain (Smart and Whitehead, 1973) and the USA (McDermott and Scheurich, 1977). In all cases the distributions were continuous and roughly log normal. These data, however, did not represent national samples. The purpose of this paper is to report data on the distribution of drug use among a national sample of students in Mexico. This study extends earlier findings since it is the first really national sample studied and the first time in which data from a developing country have been obtained. All earlier data came from the highly industrialized countries of Western Europe and North America.

Method Sampling

In 1976 the Centro Mexicano de Estudios en Farmacodependencia (CEMEF) carried out a national survey to determine the trends and current state of drug dependence among the school population between the ages of 14 and 18 years.

A multi-stage sampling approach was used to get a full view of drug use in this population. The sample included both male and female students in a variety of classes and types of schools. Two types of units were considered for the sampling: the first stage included schools and the second stage, groups of pupils within these schools. Thus, the sampling was by conglomerates, stratified and in two stages. Stratification was made according to the type of schools being considered: university preparatory schools, schools of science, vocational schools, secondary federal system schools and technological and industrial schools.

The Mexican Republic was divided into regions according to the Bassols sociodynamic criteria, 1 Within each region, cities were selected in which the survey would take place. Representatives of each region were chosen on the basis of the student population from 14 to 18 years.

1 Regionalizacion de la Republica Mexicana, Angel Bassols B, 1970, La Division Economico Regional en Mexico. UNAM, Angel Bassols B., 1967.

The sample chosen included 10,091 subjects, obtained from 748,375 students attending preparatory schools, secondary vocational-technical and industrial schools that were operating in 1976. The number of questionnaires actually completed was 9,900 or only 1.9 per cent less than the number chosen. This reduction occurred because some students were absent at the time of testing. Table 1 shows the demographic characteristics of the sample.

Questionnaire administration

The questionnaires were completed in groups and administered by two researchers. The research instrument was a self-report questionnaire with 119 items which included an identification chart, Crandall's intellectual-academic responsibility scale, the Marlow-Crowne social desirability scale, a drug dependence attitude scale and questions on the use of various drugs.

The present paper deals only with the drug use items and not with thedesirability or attitudinal items. The drugs under study were: (1) non-narcotic analgesics; (2) hypnotic non-barbiturate sedatives; (3) hypnotic-sedative barbiturates; (4) anxiolytics; (5) amphetamines; (6) stimulants; (7) non-classified depressors; (8)anti-psychotics; (9) narcotic analgesics; (10) marijuana; (11) other hallucinogens; (12) solvents; (13) anti-depressives; (14) heroin; (15) cocaine; (16) opiates; and (17) alcohol .


Demographic characteristics of the sample




5 584 56.40
4 159 42.01
Without information
157 1.59
9 900 100.00
8 451 85.36
1 265
Without information
184 1.86
9 900 100.00
14 - 15
6 208 62.70
16 - 18
3 539 35.75
Without information
153 1.55
9 900

* The percentage of subjects in the sample that belong to preparatory schools is low compared with the percentage of the secondary schools because only selected preparatory schools were included.

The students were asked to write down the name of the drug they had used on a line at the end of the sheet, in order to control all answers and to classify them. The drug classification employed was that adopted by the National Council for Problems of Drug Dependence in Mexico. Reclassification of individual drugs mentioned by students was made according to the Diccionario de Especialidades Farmacéuticas, 22nd Mexican edition, revised by Dr. Emilio Rosenstein.

When the student could not write the name, for drugs numbered 2, 3 and 4, the answer was recorded as a non-classified depressor. For the remaining questions on drugs of medical use, if the name did not appear, the answer was not recorded.

In this paper data on 16 drugs - excluding alcohol - were analysed. Questions on drug use were stated as: (i) used during the last six months (excluding the last month), and (ii) used during the last month. Answers were scored according to the frequency of drug use, with each student receiving a score according to the number and frequency of drug use:

No use
Less than once a week
Once a week
3-5 times per week or daily

Scores could thus vary between 0 and 48 for the sixteen drugs.

Results and discussion

In the analysis of the score distribution of drug use for the last month, 43.24 per cent of the studied population reported some drug use with scores ranging from 1 to 21. Three groups can be distinguished according to their drug use.

  1. On group - 5.7 per cent of the population - reported "minor-risk' drug use in terms of the number and frequency of drugs consumed. These people having a drug use score of 1, used only one drug and that less than once a week.

  2. A second group with the highest percentage of subjects had drug use scores from 2 to 3. In this group, 31.9 per cent of the subjects reported either using 2 or 3 drugs infrequently, or else using one drug more frequently. Since this group included the highest number of subjects, it was considered representative of the current level of drug use (among users only) in the Mexican student population.

  3. The third group included 5.7 per cent of the population under study: these students attained a drug use score between 4 and 21. These students may be considered the "higher-risk" group within our distribution because of the number and frequency of drugs used.

Table 2 shows the distribution of drug use scores in the student population for the last month, the last six months (excluding the previous month) and combining these, for the last six months. It can be seen that the majority of users obtain drug use scores from 2 to 3 on the questionnaire.

Figure 1 presents the distribution of drug use scores, among users only, during (i) the previous six months, (ii) the previous month, and (iii) the previous six months excluding the last month. These distributions of drug use scores, grouped in intervals of 2, are noticeably unimodal and are log normal, with skewness values of 0.18, 0.02 and 0.15, respectively.

FIGURE 1 - Distribution of drug use scores by intervals of 2

Full size image: 26 kB, FIGURE 1 - Distribution of drug use scores by intervals of 2


Score distribution of drug use in the last six months


Users in past month

User in past six months (excluding previous month)

Users in past six months

Score range







01 564 (5.70) 100 (1.01) 664 (6.70)
02 1 458 (14.73) 271 (2.74) 1 729 (17.46)
03 1 696 (17.13) 304 (3.07) 2 000 (20.20)
04 237 (2.39) 5 (0.05) 242 (2.44)
05 125 (1.26) 2 (0.02) 127 (1.28)
06 122 (1.23) 4 (0.04) 126 (1.27)
07 29 (0.29) 2 (0.02) 31 (0.31)
08 19 (0.19) 0 (0.00) 19 (0.19)
09 12 (0.12) 0 (0.00) 12 (0.12)
10 7 (0.07) 0 (0.00) 7 (0.07)
11 2 (0.02) 0 (0.00) 2 (0.02)
12 4 (0.04) 1 (0.01) 5 (0.05)
13 3 (0.03) 0 (0.00) 3 (0.03)
14 0 (0.00) 1 (0.01) 1 (0.01)
15 0 (0.00) 1 (0.01) 1 (0.01)
16 1
0 (0.00) 1 (0.01)
17 0 (0.00) 1 (0.01) 1 (0.01)
18 1 (0.01) 1 (0.01) 2 (0.02)
19 0 (0.00) 0 (0.00) 0 (0.00)
20 0 (0.00) 0 (0.00) 0 (0.00)
21 1 (0.01) 0 (0.00) 1 (0.01)

The results clearly indicate that the data from Mexico are not very different from those of Canada or Britain. They show that among student populations the distribution of drug use is continuous, unimodal and at least roughly log normal. That is, one has many infrequent users, fewer moderate users and even fewer heavy users. The same distribution has been found among a variety of Canadian populations and one British population. It seems then that drug use distributions in Mexico, a developing country, are not substantially different in form from those in two developed countries. This study adds substantially to the data base on drug use distributions - in all cases so far studied drug use is positively skewed and log normal. This suggests that there is no clear dividing line between "normal" and "heavy" users. Further work on the distribution of drug use ought to be performed in a variety of countries, since the same distribution has been found for alcohol and this work has led to the view that alcohol abuse cannot be reduced unless the extent of alcohol consumption in the general population has also been reduced.


Bruun, K., Edwards, G., Lumio, M., Makela, K., Pan, Lyn, Popham, R.E., Room, R., Schmidt, W., Skog, O., Sulkunen, P. and Osterberg, E. Alcohol Control Policies in Public Health Perspective. Finnish Foundation for Alcohol Studies, Helsinki, 1975.

McDermott, Diane, and Scheurich, J. The logarithmic normal distribution in relation to the epidemiology of drug abuse. Bulletin on Narcotics, XXIX: 1, 13-19, 1977.

Smart, R.G., and Whitehead, P.C. The consumption patterns of illicit drug use and their implications for prevention of abuse. Bulletin on Narcotics, XXIV: 1, 39-47, 1972.

Smart, R.G., and Whitehead, P.C. The prevention of drug abuse by lowering per capita consumption: distribution of consumption in samples of Canadian adults and British university students. Bulletin on Narcotics, XXV: 4, 49-55, 1973.