ABSTRACT
Introduction
Operational definitions
Sampling
Research instruments used
Method of data collection
Data processing for socio-demographic and personality variables
Results
Test-retest reliability
Discussion
Author: D. MOHAN,, P. K. RUSTAGI,, K. R. SUNDARAM, G.G. PRABHU
Pages: 1 to 8
Creation Date: 1981/01/01
This study was carried out on a sample of 254 male senior high school students in New Delhi. The authors found that variables contributing significantly to drug use were age, heterosexual dating, drug abuse among family members and drug abuse among friends. However. drug use was not found to be significantly associated with family income. Father's occupation, family structure and place of residence. Since the sample studied was relatively small. additional confirmation is needed..
This study introduces the concept of "relative risk" to the field of drug abuse. Relative risk attempts to identify those variables which may contribute towards drug-using behaviour. This measure of strength of association of the variable under study with drug-using behaviour is relatively easy to compute and comprehend. In this pilot project, male students of a senior high school were studied by means of a questionnaire [ [1]] . Socio-demographic, interpersonal and intra-psychic variables were assessed. The first two of these are discussed here. The findings on intra-psychic variables will be reported in part II.
Drug abuse: the use of any of the following substances without medical prescription: alcohol, tobacco, cannabis, hallucinogens, cocaine, amphetamines, barbiturates, methaqualone, minor tranquillizers and opiates [ 2] ;
Non-user:one who had never used any of the above substances;
Current user: one who had used any of the above substances during the past year;
Life-time user: one who had ever used any of the above-mentioned substances (includes current users);
Overall relative risk: result obtained after controlling relevant variables using the "subclassification approach" [ 3] .
Male students in the ninth, tenth and eleventh classes of two English-medium schools situated in New Delhi, one for boys and the other co-educational, constituted the basis of this study. A sample of 254 students was obtained by stratified cluster sampling, taking each section of 30-40 students as a cluster.
A self-administered questionnaire designed to ascertain socio-demographic data and drug use;
Multiphasic personality questionnaire;
Eysenck personality inventory:
An orientation questionnaire about drug use.
The last three items will be reported in part II of this study.
The questionnaires were administered to small groups of students during school hours, in the absence of school personnel. Complete confidentiality was assured to the respondents. Students were instructed to write their roll numbers on the questionnaires.
Test-retest reliability was calculated on one randomly selected sample of 25 students by administering the same questionnaires three months later.
Two comparisons were made for each variable by X2 calculation. One was between non-users and current users and the other between non-users and life- time users. A significant difference in one or both of these comparisons was the basis of selecting a variable for relative risk calculation. During relative risk calculation for each one of the variables so selected, other variables were controlled by the "subclassification approach" adapted from Mantel and Haenszel [ 3] . Significance level in all calculations was p 0.05.
Table 1 shows that alcohol and tobacco were the most commonly used drugs; 78.6 per cent of current users or 21.8 per cent of the whole sample and 84,2 per cent of life-time users or 33.5 per cent of the whole sample restricted their drug abuse to these substances. Other drugs used were tranquillizers, cannabis, sedatives, narcotics and hallucinogens.
Current users |
Life-time users |
|||
---|---|---|---|---|
Drugs used |
Number |
Percentage |
Number |
Percentage |
Tobacco only
|
7 | 2.8 | 16 | 6.3 |
Alcohol only
|
27 | 10.6 | 41 | 16.1 |
Tobacco and alcohol
|
21 | 8.3 | 28 | 11.0 |
Other drugs with or without alcohol and tobacco
|
15 | 5.9 | 16 | 6.3 |
Total
|
70 | 27.6 | 101 | 39.8 |
aNumber of non-users = 153 (60.2 per cent),
The current drug abuse rate was significantly higher in older age groups but no significant association was observed between life-time use and age (see table 2).
Non-users |
Current users |
Life-time users* |
|||||
---|---|---|---|---|---|---|---|
Age |
Number |
Percentage |
Number |
Percentage |
Number |
Percentage |
Total |
12-13
|
19 | 70.4 | 1 | 3.7 | 8 | 29.6 | 27 |
14-15
|
88 | 61.5 | 39 | 27.3 | 55 | 38.5 | 143 |
16-18
|
43 | 57.3 | 26 | 34.7 | 32 | 42.7 | 75 |
Number of non-respondents
|
3 | 4 | 6 | 9 |
Non-users vs. current users
|
Non-users vs. life-time users
|
x
2 = 7.81
|
x
2 = 1.44
|
DF = 2
|
DF=2
|
p
0.05
|
NS
|
*Includes current users.
Both current use and life-time use were significantly associated with heterosexual dating (see table 3).
Non-users |
Current users |
Life-time users* |
|||||
---|---|---|---|---|---|---|---|
Heterosexual dating |
Number |
Percentage (Horizontal) |
Number |
Percentage (Horizontal) |
Number |
Percentage (Horizontal) |
Total |
Present
|
25 | 37.3 | 30 | 44.8 | 42 | 62.7 | 67 |
Absent
|
125 | 68.7 | 38 | 20.9 | 57 | 31.3 | 182 |
Number of non-respondents
|
3 | 2 | 2 | 5 |
Non-users vs. current users
|
Non-users vs. life-time users
|
X
2 = 17.26
|
X
2 = 18.83
|
DF = I
|
DF = I
|
p 0.001
|
p 0.001
|
*Includes current users.
Drug abuse by family members was reported significantly more frequently by current users and life time users than by non-users (see table 4)
Non-users |
Current users |
Life-time users* |
||||||||
---|---|---|---|---|---|---|---|---|---|---|
Percentage |
Percentage |
Percentage |
||||||||
Drug abuse among family members |
Number |
Horizontal |
Vertical |
Number |
Horizontal |
Vertical |
Number |
Horizontal |
Vertical |
Total |
Present
|
81 | 50.6 | 31.9 | 57 | 35.6 | 81.4 | 79 | 49.4 | 78.2 | 160 |
Absent
|
72 | 76.6 | 28.4 | 13 | 13.8 | 18.6 | 22 | 23.4 | 21.8 | 94 |
Non-users vs. current users
|
Non-users vs. lifetime users
|
x
2 = 15.34
|
x
2 = 15.60
|
DF = I
|
DF = I
|
p 0.001
|
p 0.001
|
*Includes current users.
Table 5 shows that drug abuse among friends was reported significantly more frequently by current users and life-time users than by non users.
Non-users |
Current users |
Life-time users* |
||||||||
---|---|---|---|---|---|---|---|---|---|---|
Percentage |
Percentage |
Percentage |
||||||||
Drug abuse among friends |
Number |
Horizontal |
Vertical |
Number |
Horizontal |
Vertical |
Number |
Horizontal |
Vertical |
Total |
Present
|
56 | 44.8 | 36.6 | 53 | 42.4 | 75.7 | 69 | 55.2 | 68.3 | 125 |
Absent
|
97 | 75.2 | 63.4 | 17 | 13.2 | 24.3 | 32 | 24.8 | 31.7 | 129 |
Non-users vs. current users
|
Non-users vs. lifetime users
|
X
2 = 27.85
|
X
2 = 23.28
|
DF = I
|
DF= I
|
p 0.01
|
p 0.001
|
*Includes current users.
Non-users, current users and life-time users were subclassified with respect to the above four variables, i.e., age, dating, friends' use and family use of drugs.
Overall relative risk values for each variable was computed after controlling the other three by the subclassification approach. Variables in descending order of magnitude of relative risk of current use were friends' use (2.49), followed by family use (1.97) and dating ( 1.86) (see table 6). In the case of all these three variables the relative risk of life-time use was less than that of current use. The relative risk of both current and life-time use for older age groups was not significant.
Current use |
Life-time use |
|||||
---|---|---|---|---|---|---|
Variable |
Relative Risk |
X 2 |
p |
Relative Risk |
X 2 |
p |
Older age
|
1.1 | 0.07 |
NS
|
1.18 | 0.76 |
NS
|
Dating
|
1.86 | 9.01 | 0.01 | 1.75 | 11.86 | 0.001 |
Family use
|
1.97 | 6.29 | 0.05 | 1.82 | 9.04 | 0.01 |
Friends' use
|
2.49 | 11.81 | 0.001 | 1.47 | 11.51 | 0.001 |
aFor calculation of relative risk with respect to age, only two sub-divisions were made, i.e. 12-15 years and 16-18 years.
Neither current use nor life-time use was significantly associated with per capita income of the famiIy, father's education, father's occupation (professional/government office/business/other), family structure (nuclear/joint) and place of residence (home/hostel).
Test-retest concordance at three months' interval was 100 per cent for drug abuse, 100 per cent for age, 96 per cent for heterosexual dating, 92 percent for family use, 96 per cent for friends' use, 75 per cent for per capita income, 100 per cent for father's occupation, 96 per cent for father's education, 100 per cent for family structure and 100 per cent for place of residence.
In the present study, the most commonly used drugs were found to be alcohol and tobacco, which some people believe may lead to the abuse of harder drugs [ 4] . Many other investigators [5-11] have reported increases in drug abuse prevalence with increasing age and academic levels. In this study, when considered alone, older age was found to be significantly associated with drug abuse but on calculation of relative risk after controlling family use, friends' use and dating, the association was not found to be significant. The absence of any significant association in this study between drug abuse and socio-economic variables, i.e., per capita income, father's education and father's occupation, was an important negative finding. This was contrary to the findings reported in several studies from India [12-14] and elsewhere [9. 15-18]. One of the reasons for this observation could be the relatively narrow range of variation in the sample, most of the respondents having a relatively homogeneous background and socio-economic status and hence it needs confirmation on a larger and more representative sample. Nor was family structure (nuclear/joint) found to be significantly associated with drug abuse. This finding was not consistent with two other studies on the Indian student population [13, 14]. However, it is an important variable in the lndian setting and needs further exploration.
Whether a student stayed at home or in a hostel did not appear to be related to drug abuse. This finding was similar to one study [ 19] but not in line with some others [ 10, 14, 20].
Drug abuse was found to be more common in those who reported heterosexual dating behaviour. This supported the finding of a school study [ 13] and college study [ 14] conducted in Delhi. Relative risk of drug abuse associated with dating was l.86 for current use and l .75 for life-time use. Dating is a departure from traditional Indian social norms and reflects an imitation of Western countries. Higher risk of drug abuse associated with dating could be explained in terms of a higher level of extroversion or the existence of a "youth sub-culture" for which no evidence has yet been found. The psychological and sociological factors might be working synergistically in this sphere.
Like many studies conducted in lndia [8 , 10, 12, 21] and elsewhere [9, 17, 18, 22-24], the present survey also highlighted the importance of interpersonal variables associated with drug abuse. Relative risk of current use, associated with family use, was l .97, while that of life-time use was l.82. User family members in general and parents in particular may serve as models for adolescents, and not only reinforcers of behaviour [ 25] . Relative risk of current use associated with friends' use was 2.49 while that of life-time use was l.47. Though a cause and effect relationship can conclusively be established only in a prospective study, friends may influence a person to start using drugs by making the drugs available, providing an example or defining the nature of physiological experience. In conforming to peer pressure, many teenagers do things which they claim they actually disapprove of. Relative risk of current use was highest in association with friends' use, followed by family use and then dating. In the case of all these three variables, relative risk of life-time use was less than that of current use, which may indicate the role these factors play not only in initiating drug abuse but also in maintaining it. The difference between relative risk values was the greatest for friends' use, thus highlighting the importance of the peer group in continuing drug use.
High test-retest concordance (92 to 100 per cent) was observed for all socio-demographic and interpersonal variables except per capita income (75 percent), which in a sense was expected, as not all children would know their parents' exact income.
Associations found in this study could be used as leads for larger and more representative retrospective and prospective studies, wherein one might be able to establish cause and effect relationships and associations due to common causes. The usefulness of these associations would lie in their predictive and etiological implications, helping preventive planning.
R. G. Smart, "Recent studies of the validity and reliability of self-reported drug use 1970-1974", Canadian Journal of Criminology and Corrections. vol. 17. 1975, p. 326.
002D. Mohan. ed.. Research Issues and Priorities in Alcohol and Drug Abuse. Proceedings of a seminar held at the All India Institute of Medical Sciences (New Delhi, All India Institute of Medical Sciences, 1978).
003N. Mantel and W. Haenszel. "Statistical aspects of the analysis of data from retrospective studies of disease", Journal of the National Cancer Institute. vol. 22. 1959. p. 719.
004D. Kandel, "Stages in adolescent involvement in drug use", Science. vol. 190, 1975. p. 912. .
005R. H. Blum, Students and drugs (San Francisco. Jossey-Bass, 1969).
006R. Brook. J. Kaplum and P. C. Whitehead, "Personality characteristics of adolescents as measured by MMPI", British Journal of Addiction. vol. 69. 1974. p. 61.
007W. Buikuishan and H. Timmerman, "The development of drug taking among secondary school children in Netherlands", Bulletin an Narcotics. vol. 24, No. 3(1972). p. 7.
008M. J. Joshi and A. K. Niyogi, "Smoking habit of young men and its prevention". Indian Journal of Public Health. vol. 8, 1964, p. 17.
009D. Kandel, "Some comments on the relationship of selected criteria variables to adolescent illicit drug use". in Predicting Adolescent Drug Abuse: A Review of Issues. Methods and Correlates. D. J. Lettieri. ed. (Rockville. Maryland. National Institute of Drug Abuse. 1975).
010J. S. Mathur and N. R. Sarna, "Drug addiction amongst male medical student". Indian Journal of Preventive and Social Medicine. vol. 6, 1975, p. 18.
011v. N. Sharma, L. K. Kothari and R. M. Kasliwal, "Ecology of medical students" . Indian Journal of Medical Education. vol. 4, 1965, p. 234.
012A. K. Gupta and B. B. Sethi, "Psychosexual aspects of cigarette smoking", Indian Journal of Psychiatry. vol. 19, 1977, p. 11.
013M. G. Thomas, "Prevalence and pattern of drug abuse amongst high school students". M.D. thesis submitted to All India Institute of Medical Sciences. New Delhi, 1975.
014D. Mohan and others, "Factors associated with the prevalence of drug abuse among Delhi University students", Indian Journal of Psychiatry. vol. 20, 1978, p. 332.
015C. P. Eichberg and P. M. Bentler, "Current issues in the epidemiology of drug abuse as related to psychosocial studies of adolescent drug use", in Predicting Adolescent Drug Abuse: A Review of Issues. Methods and Correlates. D. J. Lettieri, ed. (Rockville, Maryland, National Institute of Drug Abuse,1975).
016E. M. Harris. "A measurement of alienation in college student marijuana users and non users", Journal of School Health . vol. 41. 1971. p. 130.
017J. J. Prendergast, Jr.. "Family characteristics associated with marijuana use among adolescents". International Journal of Addictions . vol. 9. 1974, p. 827.
018R. G. Smart and D. Fejer, "Drug use among adolescents and their parents: closing the generation gap in mood modification". Journal of Abnormal Psychology , vol. 79, 1972. p. 153.
019R. N. Banerjee, "Prevalence of habit forming drugs and smoking among college students-a survey", Indian Medical Journal . vol. 57. 1963, p. 193.
020B. B. Sethi and R. Manchanda, "Pattern of drug abuse among male students", Indian Journal of Psychiatry , vol. 20. 1978. p. 66.
021S. Chitnis. Drugs on College Campus (Bombay. Tata Institute of Social Sciences, 1974).
022D. Kandel, "Interpersonal influences on adolescent drug abuse", in Drug Use-Epidemiological and Sociological Approaches . E. Josephson and E. Carroll. eds. (Washington, D.C., Hemisphere. 1974).
023M. A. Lavenhar and others, "A survey of drug abuse in six suburban New Jersey high schools: II-Characteristics of drug users and non users", in Student Drug Surveys , S. Einstein and S. Allen. eds. (Farmingdale. New York, Baywood, 1972).
024R. A. Staffenhagen, C. P. McAcree and L. S. Zheutlin. "Social and academic factors associated with drug use on the University of Vermont campus", International Journal of Social Psychiatry , vol. 15, 1969, p. 92.
025A. Bandura and F. J. McDonald, "Influence of social reinforcement and behaviour of models". as quoted in B. Kuppuswamy. Elements of Social Psychology (New Delhi, Vikas, 1973).