Volume 32 Number 3
May 1993

Roland J. Lamarine

Adolescent alcohol and drug uses have attracted considerable attention in recent years. For American Indians, the issue of drug and, particularly, alcohol abuse has had serious repercussions, in some cases nearly sounding a death knell for entire tribes (National Clearinghouse, 1989). Morbidity and mortality data indicate that alcohol poses a significant risk to the health of a large number of American Indians (Lamarine, 1988). Recent research confirms the existence of a grave health crisis confronting American Indian/Alaska Native adolescents (Blum, Harmon, Harris, Bergeisen, & Resnick, 1992).

Despite substantial widespread difficulties with alcohol, the patterns of drug and alcohol use are extremely variable among the more than 300 tribal groups living in the U.S. (May, 1986). Since the mid-1970s, regular, systematic surveys have been conducted to assess the extent of drug and alcohol use among large cross sections of American Indian high school students. The most recent national survey data for American Indians in grades 7-12 indicate that between 1975 and 1987, lifetime alcohol use (having used alcohol at least once in their life), increased from 76% to 81 %. During the same period, self-reported marijuana use increased from 41% to 61% (Beauvais, Oetting, Wolf, & Edward, 1989). Compared to non-Indians, American Indian consumption of alcohol was 21% higher and marijuana use was 35% greater during 1985 (Owan, Palmer, & Quintana, 1987).

Among the most important risk factors predictive of adolescent substance use are drug use among peers, drug use among adult role models, and early use of alcohol (Newcomb, Maddahian, & Bentler, 1986). Wallack and Corbett (1987, p. 233) have succinctly summarized the following information:

Alcohol and drug use by peers and siblings has been associated with adolescent use of alcohol, marijuana, and other drugs in a number of studies (Halebsky, 1987; Thorne & DeBlassie, 1985). Parental alcohol and drug use, including tobacco, has also shown a strong association with adolescent use in a variety of studies (Friedman, 1985; Gorsuch & Butler, 1976; Halebsky, 1987; Harburg, Davis, & Kaplan, 1982; Johnson, Shontz, & Locke, 1984; Kandel, Kessler, & Margulies, 1978).

Though drug education alone may not be sufficient in effecting desirable changes in adolescent behaviors relating to drug use, it is a necessary base upon which effective programs may be built. An understanding of the information channels that American Indian adolescents perceive as most effective in their drug education could provide valuable information for the development of successful drug education programs (Bailey, 1985). However, an examination of current literature revealed a lack of research describing where American Indian adolescents obtained information about alcohol and other drugs.

Despite a variety of contradictory findings, studies of adolescent sources of information about alcohol and drugs point to a clear difference in the sources of information between substance abusers and non-abusers. In general, abusers tend to place more emphasis on their own experience and friends, while nonabusers attend more closely to family and teachers as sources of alcohol and drug information. Both groups report mass media sources as influential (Beck & Summons, 1986; Sarvela, Newcomb, & Littlefield, 1988). When questioned about sources of accurate drug information during adolescence, university students clearly perceived school as the most important source of information followed by print media, and peers (Wilson & Lamarine, 1990).

The present pilot study surveyed selected American Indian high school seniors attending school in New Mexico. The purpose of the study was to identify sources of information and to examine their relationship to perceived harmfulness of these substances, parental attitudes toward adolescent use, frequency of use, age of first use, and other intrapersonal, social, and environmental factors pertinent to substance use within the context of field testing the survey instrument.

A nonrandomized sample of 168 American Indian high school seniors participated in this cross-sectional study. Students were selected from three school districts in New Mexico which volunteered to participate in a pilot study examining the efficacy of this line of research. The school districts which participated primarily served the Navajo and Mescalero Apache tribes. Letters explaining the nature of the research were sent to all parents of high school seniors at participating schools. Only those students who returned completed parental permission forms (approximately 50% of the total) were allowed to participate in the survey.

The survey instrument was a questionnaire which contained demographic items regarding gender, age, tribe, and family structure. Respondents were asked the age at which they first tried alcohol, cigarettes, chewing tobacco, marijuana, and other drugs and how old they were when they first learned about the dangers of these substances. Subjects were also asked to select sources of information from which they first learned about these substances. All of these items have been described in more detail elsewhere (Wilson & Lamarine, 1990).

Environmental influences on behavior were explored through an item which asked "Have any of the following people ever had any drug or alcohol problems?" Answer choices included self, father, mother, siblings, grandparents or other relatives, and close friends. Information about personal involvement with a variety of licit and illicit substances was obtained. Participants were asked how often they had used each substance during the last year with answer choices ranging on a five point Likert scale from "never" to "almost every day." Subjects were also asked how often they used more than one drug on the same occasion.

Parental attitude toward adolescent drug use has been identified as an important variable determining adolescent use. Respondents were asked "How do your parents/guardians feel about your use" of the same five substances previously listed. Answer choices ranged on a five point Likert scale from strongly against to strongly in favor. To ascertain student perceptions of the dangers of drug use, a question was included which asked "How harmful do you think regular use of the following substances would be?" Response choices ranged from harmless to extremely harmful on a five point Likert scale.

It was anticipated that some students would have been successful in avoiding involvement with drugs. To determine the frame of reference which may have facilitated their choice to resist using drugs, an item was included which inquired about "A good reason(s) for not using alcohol or other drugs is that you..." Answer choices included a variety of options concerning the possibility of addiction, legal difficulties, peer or parental disapproval, spiritual or health reasons, and not living up to one's own expectations. The survey also contained items relating to involvement in drug education, student GPA, unexcused absences from school, and attitude toward school.

The questionnaire was administered anonymously by classroom teachers at the individual schools following a predetermined protocol. It was pretested for readability and clarity among a sample of American Indian high school seniors in a New Mexico school district that was not included in the survey. Administration time was approximately 20 minutes. The protocol for this study was approved by the Institutional Review Board at the researcher's university.

Respondents were primarily Navajo (93%) and female (54%). Sixty-seven percent described the family in which they grew up as a two parent family, while 26% reported that they lived with only one parent. Students were asked their age "the first time" they tried a variety of drugs and their age when they "learned about the dangers" of those same drugs. Results are presented in Table 1. The time spans between students first receiving information and their first use of a substance are almost all indicative of health information being available before students started to use dangerous substances. The single exception involved chewing tobacco, which many of the respondents began using at an early age.

Table 1

Mean Age of First Experience and First Information

About Alcohol, Tobacco, and Other Drugs

Sources of drug information most relevant for this population were determined by asking "Where did you first learn about the following drugs?" Results are presented in Table 2. For all substances combined, the major sources of information were school, electronic media, and print media.

Table 2

Percentage of Subjects Indicating Sources of Information (SOI)

for Each Variable*

% indicating yes as a source of information (Sol) and (ranking).

To assess one important environmental influence on drug use, students were asked to identify people close to them who had "ever had any drug or alcohol problems." Grandparents and other relatives were selected by 39% of the sample, 34% identified father, 24% siblings, 6% mother, 13% selected themselves, while 56% identified close friends. In response to another question related to environmental influences on drug use, "How do you like school?" a large majority (77%) of the subjects reported they "like it" or "really like it."

Alcohol and drug use patterns were examined by asking subjects how often they had used various drugs during the last year. Heavy alcohol use (daily or weekly) was reported by 19% of the sample, while 58% stated that they rarely or never drank. Regular (daily or weekly) cigarette smoking was indicated by only 14% of those sampled and regular smokeless tobacco use was limited to 13% of those reporting. Marijuana use (6%) on a regular basis (daily or weekly) and other drug use (4%) were limited to a small percentage of the survey group. Polydrug use (two or more drugs on the same occasion) was reported as a regular practice (daily or weekly) by 10% of the sample, while 85% stated that they rarely or never participated in this practice.

Significant gender differences in the frequencies of drug use were found for alcohol (X=10.74, p=.03, df=4), cigarettes (X=11.25, p=.02, df=4), chewing tobacco (X=22.18, p=<.001, df=4), and marijuana (X=9.55, p=.049, df=4) with males exhibiting heavier use in each category (see Table 3).

Table 3

Gender Differences in the Frequency of Drug Use.

Numbers Represent Percent Reporting Weekly or Daily Use.

Since parental attitudes and role modeling have been identified as important determinants of adolescent behavior, respondents were asked "How do your parents/guardians feel about your use" of several types of drugs. The percentage of parents who were perceived as somewhat or strongly against their children's use of drugs included chewing tobacco (82%), cigarettes (84%), alcohol (95%), marijuana (96%), and other drugs (96%).

To examine the hypothesis that parents' attitudes toward drug use were related to the frequency of adolescent drug use, correlation coefficients were calculated for these two variables. The results suggested a moderate, positive correlation for alcohol (r=.22) which was the weakest relationship noted, while chewing tobacco (r=.43) produced the strongest association. Other correlations included cigarettes (.42), other drugs (.39), and polydrug use (.36).

Perceived risks are sometimes more important than actual risks in determining adolescent risk-taking behaviors. Student perceptions of the dangers of various drugs were evaluated by asking "how harmful" they thought it would be to use various drugs on a daily basis. Sizeable majorities identified each of the drugs listed on the survey as extremely or mainly harmful. The percentage responses in each drug category included marijuana (84%), other drugs (79%), alcohol (65%), cigarettes (61%), and chewing tobacco (54%).

Correlation coefficients relating frequency of drug use with student perceptions of drug harmfulness indicated low but positive associations. The strongest correlations involved marijuana (r=.29) and other drugs (r=.21). Cigarettes (.14), alcohol (.09), and smokeless tobacco (.07) use were all weakly correlated with student perceived drug harmfulness.

Often drug surveys focus only on negative behaviors. To assist in understanding why some adolescents are better able to resist pressures to engage in undesirable behaviors, respondents were asked to select "a good reason(s) for not using alcohol or other drugs." In decreasing order of importance students selected "would disappoint your parents" (86%), "would damage your health" (84%), "might become an addict" (81%), "could get into trouble with the police or school" (77%), "might lose close friends who do not approve" (65%), "would disappoint yourself" (51%), and "would not be following your spiritual beliefs" (46%).

Drug education curriculum implications are evident in an examination of survey results comparing age at first experience with various drugs with age of first information about those drugs. It would be advisable to present students with accurate information about the risks involved with drug use at least 1 year before they are likely to use those drugs. The data from Table I suggest that respondents to this survey have been informed in a timely manner about all of the substances examined with the exception of smokeless tobacco. Survey participants reported using smokeless tobacco at a mean age of 12.7 years but they did not receive their first health information about this product until they were approximately 12.8 years of age. The advent of smokeless tobacco use in this population was at an early age, perhaps necessitating an earlier intervention program focusing on smokeless tobacco somewhere around fifth grade.

American Indian high school seniors in this survey were similar to university students examined in an earlier study, who also reported that school was their primary source of information about alcohol, tobacco, and other drugs (Wilson & Lamarine, 1990). Unlike the university students, American Indian adolescents chose electronic media (television, movies, and radio) and print media as their second and third choices. Besides the obvious conclusion that schools are instrumental in educating American Indian adolescents about drugs, it should be noted that American Indian adolescents appear particularly receptive to the influence of electronic media information channels. This latter result may reflect the current proliferation of access to television, even on the most remote sections of many reservations.

An examination of individuals perceived by the students as having alcohol and drug problems produced an interesting dilemma. While 56% of the respondents stated that close friends had such problems, only 13% identified themselves as suffering from drug or alcohol problems. It is possible that similar behaviors were perceived as "problems" when they occurred in friends but were not perceived as "problems" when they affected the respondents suggesting some degree of denial.

Frequency of substance use data indicated that only 19% of those sampled reported using alcohol more than once a month, while the percentages for cigarettes (14%), smokeless tobacco (13%), marijuana (6%), other drugs (4%), and polydrug use (10%) were all equally low. Gender differences in the frequency of drug use are shown in Table 3. Significant gender differences in the frequency of use indicating heavier male usage are consistent with previous research (Beauvais, Oetting, Wolf, & Edward, 1989).

Comparisons of the frequency of substance use reported in this study were made with national survey data for American Indians and are presented in Table 4. These data indicate that the current study group seems to be more successfully avoiding the use of marijuana and tobacco products. Care should be exercised in the interpretation of these results, since the non-random nature of the sampling procedure may have led to a self-selection process favoring relatively low-risk students who were more likely to return parental consent forms.

Table 4

Comparisons Between a National Sample of Indian 7-12th Graders Reporting

Lifetime Use of Drugs (Ever Tried), 1986-87 and Drug Use During Last Year

(1989-90). Numbers Represent Percent in Each Category.

* Beauvais, Oetting, Wolf, & Edwards, 1989.

The comparatively low rates of substance use among this survey population may be partially attributable to the large proportion reporting parental disapproval of substance use ranging from 82% for smokeless tobacco to 96% for marijuana and other drugs. Perceived harmfulness of drug use may also be an important determinant for moderating drug related involvement. When questioned about the dangers of marijuana, the largest majority, 84% of those responding, viewed it (albeit incorrectly) as extremely or mainly harmful, while at the low end of the scale, only 54% felt that smokeless tobacco was extremely or mainly harmful. An additional consideration is related to the large proportion (77%) of students reporting that they liked school, a finding that is consistent with the schools' graduation rate of approximately 93% for 1991-92. This corresponds well with Mason's research which offered support for the position that school can provide a stable norm reference group for students who want to avoid involvement with drugs or alcohol (Owan, Palmer, & Quintana, 1987).

A majority (61 %) of students reported having a drug or alcohol class during the most recent school year; administrators at the schools involved indicated a major emphasis on alcohol and drug education programs had been initiated at their institutions. In retrospect, this turned out to be a good sample to ask about reasons for not using alcohol and other drugs. Eighty-six percent selected "would disappoint your parents" as a good reason not to use drugs and alcohol. Damage to one's health (84%) may be a less surprising selection in view of what students may have learned from school and mass media drug education efforts. The questions on perceived harmfulness of substances indicated that sizeable majorities of students perceived these substances as dangerous to their health.

Correlations between parents' views on adolescent drug and alcohol use and self-reported frequencies of substance use by adolescents support the hypothesis that parents' views do make a difference in adolescent behaviors related to substance use. For example, students in this survey reported relatively low levels of drug use, while their parents expressed high levels of disapproval for using drugs; correlation coefficients for these two variables were moderately high. Furthermore, students selected parental disappointment as their main reason for not using alcohol and other drugs.

Limitations on the interpretation of the data collected in this pilot study involve precautions associated with self-reported drug use patterns. The collection of surveys in sealed envelopes, assured anonymity of responses, and administration of surveys by familiar, trusted, classroom teachers were methods used to insure the highest levels of trust and honesty feasible with such surveys.

However, unless respondents are convinced of their anonymity, there may be a tendency to inaccurately report levels of personal use. As a pilot study, a limited number of school districts were sampled. Results from this study population lack external validity due to the non-randomized sample. Conclusions drawn from comparisons with national survey data are offered for illustrative purposes only. Institutional Review Board protocol for such research necessitated that parental consent forms be obtained for 0 participants. It is reasonable to assume that some of the most "at risk" students would be less likely to return parental consent forms, thus contributing to an under representation of drug abusers and heavy users.

It is suggested that future surveys require passive consent, the return of parental consent forms only in the case of parents wishing to exclude their underage children from participation in the study, otherwise, the most "at risk" youngsters may continue to be omitted from the survey. It would also be instructive to examine drug use patterns and attitudes in school districts where less emphasis has been placed on drug education programs. The results of this pilot study may also indicate that young people with serious drug and alcohol dependencies are underrepresented in school-based surveys because such individuals may be less likely to return parental consent forms or may have higher levels of absenteeism than students who do not abuse substances.

Results of this pilot study support the efficacy of a more extensive survey of American Indian adolescent drug use patterns. A comprehensive survey would be relatively low in cost and in risks to participating students. It would yield a wide variety of data useful to drug education curriculum planners. Among the more salient outcomes of such a project would be useful information regarding the frequency and nature of substance use among students. Differences in the age of first experience and the age of first information could provide an essential framework for appropriate curriculum sequencing.

An understanding of student sources of information about tobacco, drugs, and alcohol would offer educators an opportunity to target information channels which are most influential for their students. Source of information research has been used successfully by community health education programs which have been structured around the identification of information channels that provide the most direct link to target populations.

Often schools are selected as the vehicles for addressing social problems whose etiology is intrinsically woven into patterns of family and community dysfunction. Information about the effects of family relationships on drug use among adolescents could provide insight into the need for intensive community as well as school-based intervention efforts. In addition, comparisons between students who have participated in drug education programs with those who have not been involved in such programs could provide general information about program effectiveness.

Perhaps the most promising avenue of exploration opened by this pilot project concerns the examination of students' reasons for not using drugs. A more extensive study of American Indian adolescents who have been "successful" in avoiding drug and alcohol problems could yield data useful in developing improved prevention programs and policies.