Drug Abuse

CORRELATES OF OCCASIONAL CIGARETTE AND MARIJUANA USE: ARE TEENS HARM REDUCING?
KEN RESNICOW,* MATT SMITH,* LANA HARRISON,+ and ERNEST DRUCKER++
*Emory University +University of Delaware; and++Albert Einstein College of Medicine
Addictive Behaviors. Vol. 24, No. 2, pp. 251-266,1999

Abstract-This paper examines characteristics that distinguish heavy and occasional cigarette and marijuana use among U.S. high school seniors. High school seniors who completed the 1994 Monitoring the Future survey (N - 15,929) were classified as nonusers, occasional users, and heavy users of cigarettes and marijuana. Level of use was examined with regard to degree of perceived risk of regular use, perceived risk of occasional use as well as several psychosocial factors previously shown to be associated with use (e.g., peer use, dose friends' approval of use, and self-esteem). Involvement with other drugs and high-risk behaviors across levels of use was also examined. Heavy users of cigarettes or marijuana reported significantly lower perceive risk of regular use of each substance compared to occasional users. Heavy and occasional users did not generally differ with regard to perceived risk of occasional use. Compared to heavy users, occasional cigarette or marijuana users reported significantly lower rates of illicit drug use, heavy alcohol use, and high-risk driving behaviors as well as fewer problem behaviors and higher grades. Some adolescents may moderate their cigarette and marijuana use to minimize harmful effects they associate with heavier use (i.e., they may be practicing harm reduction). Additional research is needed to better understand controlled substance use, and incorporate this information into prevention and cessation programs.

Despite considerable educational efforts, drug use among American teens has increased during the past few years. Among high school seniors, for example, between 1992 and 1995, 30,day prevalence of cigarette use increased from 27.8% to 33.5%, while marijuana use jumped from 11.9% to 21.2% (Cigarette smoking among Ameri can teens rises again in 1995, 1995a; Drug use rues again in 1995 among American teens, 1995b).

Though the ultimate national public health objective may be to eliminate cigarette and marijuana use entirely, some researchers and policymakers contend that this goal is neither attainable nor perhaps even desirable (Des Jarlais, Friedman, & Ward, 1993; Drucker, 1995; Dutican, Nicholson, Clifford, Hawkins, & Petosa, 1994). Instead, it is contended that public health efforts should focus on minimizing the potential physical, psychologic, and social costs associated with substance use, rather than eliminating use per se (Des Jarlais et al., 1993; Drucker, 1995; Duncan et al., 1994). Examples of this so-called Harm Reduction approach include methadone maintenance to mitigate the addictive aspects of heroin use and needle exchange and bleach distribution to reduce the transmission of HIV and other blood-borne illnesses (Des Jarlais et al., 1993; Drucker,1995). Initiatives to reduce driving under the influence of alcohol represent another application of these principles (Duncan et al., 1994).

Harm Reduction principles have also been applied to marijuana and cigarette use. In the Netherlands, for example, sale of cannabis is tolerated in selected "Koffeeshops"; however, sales to minors and purchases exceeding 30 g per customer are prohibited (Drucker, 1995). With regard to cigarette use, Harm Reduction strategies include use of lower tar and nicotine cigarettes, nicotine replacement systems, and controlled smoking programs.

Another potential means for reducing the harmful effects of substance use is to lower heavy use and abuse. Perhaps excluding alcohol (Duffy,1995; Fuchs et al., 1995; Poikolainen, 1995), any substance use carries increased physical risk relative to nonuse. Nonetheless, there may be public health, social, and economic benefits to encourage occasional use as an alternative to heavy use or abuse. For example, there is a dose-response relationship between cigarette and marijuana intake and adverse health effects (Kondo et al., 1996; Menhiratta, Wig, & Verma, 1978; West, 1992; Zang & Wynder, 1992). Additionally, occasional users may be less likely to progress to using "harder" substances (Kandel, Yamaguchi, & Chen, 1992; Kimlicka, 1978, Klonoff & Clark, 1976; Kouri, Pope, Yurgelun-Todd, & Gruber, 1995; Torabi, Bailey, & MajdJabbari, 1993; Willard & Schoenborn, 1995), may have less difficulty quitting (McWhorter. Boyd, & Mattson, 1990; Owen, Kent, Wakefield, & Roberts, 1995; Shiffman,1989; Wilson, Wakefield, Owen, & Roberts, 1992) and may exhibit lower rates of school failure and other illegal/antisocial behaviors (i.e., social morbidity; Shedler & Block, 1990; Willard & Schoenborn,1995).

Several studies, conducted mostly in adults and college students, have examined the correlates and consequences of heavy versus occasional cigarette and marijuana use (Budd, Bleiker, & Spencer, 1983, Hajek, West, & Wilson, 1995; Klonoff & Clark, 1976; Kouri et al., 1995; Oven et al., 1995; Primavera & Pascal, 1986; Shiffman,1989; Wilson et al., 1992). Two studies found that regular marijuana users had lower levels of perceived risk than did occasional users (Budd et al., 1983; Kimlicka,1978), though one study found no such differences (Primavera & Pascal, 1986). Perceived risk is also associated with frequency and intensity of use among high school youth (Bachman, Johnston, & O'Malley, 1990; Bachman, Johnston, O'Malley, & Humphrey, 1988). Compared to abstainers and heavy users, experimenters and occasional substance users appear to have more favorable psychologic profiles (Clifford, Edmundson, Koch, & Dodd. 1991; Shedler & 1$lock,1990). Occasional cigarette smokers also have higher educational attainment than do heavy smokers, and they are more likely to be black and female (Hajek et al., 1995; Resnicow, Kabat, & Wynder,1991; Serxner, Catalano, Dooley, & Mishra, 1992; Wilson et al., 1992).

One issue of interest is the extent to which occasional cigarette users and marijuana users intentionally moderate their intake to reduce deleterious physical and social effects. Understanding why teens moderate their level of substance use can inform public health policy as well as the design of prevention and cessation programs. Using data from the 1994 Monitoring the Future study, this analysis will examine characteristics that may distinguish occasional and heavy cigarette use and marijuana use among U.S. high school seniors.

 

METHODS

 

Study design

Monitoring the Future is a continuing study of lifestyles and values of American youth. conducted annually since 1975 by the University of Michigan's Institute for Social Research, with core funding from the National Institute on Drug Abuse (Bachman, Johnston, & O'Malley, 1996; Johnston, O'Malley, & Bachman,1995).

A multistage sampling framework is used to select participating schools, beginning with enumeration of 72 primary sampling areas throughout the contiguous United States. Within each area one or more high schools are selected. Probability of being selected is proportional to the size of the senior class, with larger schools having a greater chance of being chosen. Within each school, up to 400 seniors may be included in data collection. In schools with more than 400 students, a subset of students is randomly selected. For administrative efficiency, schools participate in the data collection for 2 years. For estimating prevalence rates, data are weighted to ensure samples are representative of the U.S. high school senior population.

Surveys are administered in the spring, usually during normal classroom hours, by a local project representative following a standardized protocol. Teachers remain in the room, but they are asked to avoid walking around so that students can respond without feeling observed. Students' names and addresses are requested to allow follow-up communication; however, students are informed that responses are entirely confidential. Identifying information is written on a separate form, which is removed from the questionnaire and handed in separately. Given the comprehensive nature of the study, it is not presented to participants as a "Drug Study."

Because of the breadth of content addressed, study questions are divided across six forms, distributed to ensure approximately equal samples for each form. About onethird of the items on each form are "core" items, primarily addressing substance-use behaviors and demographic characteristics. Additional information regarding sampling procedures, response rates, questionnaire validity, and other aspects of the study design have been reported elsewhere (Bachman et al., 1996).

 

Sample

 

A total of 15,929 students completed the 1994 survey, comprising approximately 2.600 respondents for each of the six forms. Participants were drawn from 119 public and 20 private schools. The response rate, derived by dividing the attained sample by the total target sample based on school rosters, was 84%.

 

Measures

 

Drug use and other behavioral variables. Cigarette use(all forms). This was classified based on response to a single item: "How frequently have you smoked cigarettes during the past 30 days?" Response categories ranged from not at all to more than two packs per day. Occasional smokerswere defined as those who reported smoking between one cigarette and half a pack per day in the past 30 days. Those reporting use of more than half a pack per day in the past 30 days were classified as heavysmokers.

There is little consensus regarding the definition of "occasional" or "heavy" substance use (Hajek et al., 1995; Klonoff & Clark, 1976; Kouri et al., 1995; Owen et al., 1995; Primavera & Pascal, 1986; Shiffman, 1989; Wilson et al., 1992), and some researchers have considered one-half pack or more per day as heavy smoking (Owen et al., 1995; Shiffman, 1989). Analyses coding half-a-pack-per-day smokers as heavy smokers, however, yielded results virtually identical to those reported with this group classified as occasional users.

 

Marijuana use (all forms). Thiswas classified based on response to the item: "On how many occasions have you used marijuana/hashish during the last 30 days?" Response categories ranged from 0 occasions to 40 or more. Occasional marijuana users were defined as those who reported use on 1 to 5 occasions during the past 30 days. Heavy marijuana users were those reporting use on more than five occasions during the past 30 days.

Hard drug use (all forms). Individuals consuming any of the nine listed substances once or more in the past 30 days were classified as hard drug users.The nine sub stances, each assessed by separate items, were LSD, other psychedelics, cocaine, am phetamines, barbiturates. tranquilizers, heroin, and other narcotics, and inhalants.

Alcohol use (all forms). Individuals consuming alcohol on six or more occasions or consuming five or more drinks in a row on more than two occasions during the past 30 days were classified as heavy alcohol users. Moderate drinkingwas defined as con sumption of 1 to 5 drinks in the past 30 days.

High-risk driving behavior (all forms). Individuals who in the past 12 months re ported receiving a traffic ticket or warning following the use of alcohol, marijuana/ hashish, or other illegal drugs or were in a motor vehicle accident following the use of alcohol, marijuana/hashish, or other illegal drugs, were classified as high-risk drivers.

Problem behaviors (Forms 2 and 6).An index of problem behaviors was computed as the mean response to 17 items, each beginning with an identical stem: "During the last 12 months how often have you... ?" Problem behaviors assessed included argued or had a fight with either of one's parents, used a knife or gun, taken something not belonging to you, gone into some house where you weren't supposed to be, and damaged school property.

Psychosoeial variables

Perceived harm from substance use (Forms 1, 2, 3, 5, 6). Respondents were asked to rate how much they think people risk harming themselves physically or in other ways by using various substances including cigarettes, marijuana, cocaine, and alcohol. Re sponses ranged from no risk to great risk, with higher values indicating greater levels of perceived risk. For each substance. excluding cigarettes, separate questions as sessed perceived harm of occasional use (the composite of items assessing use once or twice and occasional use) and regular use. For cigarettes, only perceived harm of regular use (i.e., one or more packs per day) was assessed. Two indices were computed, one based on the mean response to occasional-use items, the second based on the mean response to regular-use items. For selected analyses, individual items pertaining to perceived risk of cigarette (regular use only) or marijuana use (regular and occa sional use) were examined separately, and the associated general perceived risk scales were recomputed excluding these substance-specific items. Across-forms, internal consistency (alpha) for the occasional-use risk scale ranged from.84 to.90, and for the regular-use risk scale from .80 to .89.

Personal approval of substance use (Forms 1, 2, 3, 6).This was based on response to multiple items, each beginning with the same stem: "Do you disapprove of people (who are 18 or older) doing each of the following. . . T' Specific items included: smoking one or more packs of cigarettes per day, marijuana once or twice, smoking marijuana regularly, trying cocaine once or twice, and taking cocaine regularly. Separate scales were computed for approval of regular and occasional use (combining use once or twice with occasional use). Across-forms alpha for the occasional-use approval scale ranged from .88 to .91, and for the regular-use approval scale from .86 to .92.

Close friends' approval of substance use (Form 6). Thiswas based on response to multiple items each beginning with the stem: "How do you think your close friends feel about you doing each of the following things.. . ?" Specific substances listed were similar to those contained in the Perceived Risk and Personal Approval scales. Separate scales were computed for close friends' approval of regular use (« = .85) and oc casional use (« = .88).

Estimated peer cigarette and marijuana use (Forms 1, 3, 6).These measures were each assessed with a single item: "How many of your friends would you estimate take... ?" Responses ranged from none to all.

Self-Esteem (Forms 2, 5, 6). Scores on self-esteem were based on response to multiple items derived from the Rosenberg Self-Esteem Scale (O'Malley & Bachman,1983; Rosenberg,1979). Items, answered on an agree/disagree continuum, included: I take a positive attitude toward myself, I feel I do not have much to be proud of (reverse coded), and I am able to do things as well as most people. Across-forms alpha ranged from .86 to .89.

Demographics (all forms)

Expendable incomewas computed from two items assessing money earned per week from work or other sources (e.g., allowance). Parent educationwas classified as high, if either parent had completed college, low if not. Income and parent education (as a proxy for household income) were included as covariates as it is possible that level of use is related to how much money is available for purchasing substances. Race/ ethnicitywas assessed for only two groups, White and Black. Other racial/ethnic groups were coded as missing. School gradeswere assessed with a single item, with nine categories ranging from D to A, and higher values indicating better grades. Grades were includes as a marker of school achievement, a possible correlate of sub stance use (Table 1).

Study hypothesis

Analyses focus on the extent to which occasional and heavy users of cigarettes or marijuana differ with regard to their perceived risk of regular and occasional use. Specifically, two hypotheses were examined:

1. Compared to heavy users, occasional users of marijuana or cigarettes will report greater levels of perceived risk associated with regular (i.e., heavy) use of these substances; and

2. Compared to heavy users, occasional users of marijuana or cigarettes will report similar levels of perceived risk associated with occasional use of these substances.

 

Statistical methods

Analyses began by examining levels of perceived risk of use among occasional and heavy cigarette users and marijuana users, represented as the odds (with corresponding 95% confidence intervals) of heavy use compared tooccasional use for each perceived-risk variable. In this stage, analyses are adjusted for gender, ethnicity, expendable income, and parent education. In the next stage, several psychosocial variables that have previously been associated with substance use, and therefore potential mediators of the relationship between perceived risk and level of use, were included in the model (Botvin et al.. 1993; Botvin & Dusenbury, 1992; Botvin, Baker, Goldberg, Dusenbury, & Botvin. 1992; Donaldson, Graham, & Hansen, 1994; McCaul & Glasgow, 1985; Vega, Zimmerman. Warheit, Apospori, & Gil, 1993). They are: personal approval of regular use, personal approval of occasional use, close friends' approval of regular use, close friends' approval of occasional use, estimates of peer use, and selfesteem. Sample size for these latter analyses are substantially reduced as the respective variables appear on only a single form. For illustrative purposes, the same psychosocial and demographic variables are compared among both occasional and nonusers of cigarettes or marijuana.

Analyses conclude with examination of hard drug use. high-risk driving behaviors, heavy alcohol use, problem behaviors. and school grades across levels of use. The former three variables were dichotomous and were analyzed by chi-square, while the latter two were interval data and were analyzed using ANOVA. Because this report focuses on the association among variables rather than establishing prevalence rates, all analyses use unweighted data.

RESULTS

As shown in Table 1, among respondents who reported any use of cigarettes in the past 30 days. 81 % (i.e., 25 %/31 %) were occasional users, while 19% (i.e.. 6%/31 % ) were heavy users. Of those who reported any marijuana use in the past 30 days, 55% were occasional users and 45% were heavy users.

Heavy vs. occasional cigarette use

As shown in Table 2, heavy cigarette users reported significantly lower (OR = .54; CI .47, .63) perceived risk of regular cigarette use than did occasional users. This perception was specific to regular cigarette use, as the two groups did not significantly differ with regard to perceived risk of occasional substance use or risk of regular substance use, other than cigarettes. Blacks were significantly less likely than Whites to be heavy cigarette users as were individuals with high parental education.

Heavy vs. occasional marijuana use

As shown in Table 3, heavy marijuana users reported significantly lower (OR = .46; CI .40, .53) perceived risk of regular marijuana use than did occasional users. The two groups did not significantly differ with regard to perceived risk of occasional marijuana use or risk of regular substance use, other than marijuana. Heavy users reported significantly lower perceived risk of occasional use of substances other than marijuana (OR = .77; CI .65..92). Females were significantly less likely to report heavy use, and individuals with more expendable income were significantly more likely to report heavy use..

Occasional vs. nonuse of cigarettes and marijuana

Occasional users and nonusers of cigarettes differed with regard to perceived risk of regular and occasional use of cigarettes as well as other substances (Table 2). Being fdmale and White and having more expendable income were also significant predictors of being an occasional user as opposed to a nonuser. Occasional users and nonusers of marijuana differed with regard to perceived risk of regular and occasional marijuana use as well as risk of occasional use of other substances (Table 3). Being White and having more expendable income were also significant predictors of occasional use versus nonuse.

Analyses including psychosocial mediators

Analyses adjusting for potential psychosocial mediators are presented in Tables 4 and 5. Again. perceived risk of regular use, but none of the other perceived risk variables, was significantly lower among heavy cigarette (Table 4) and marijuana users (Table 5) than among occasional users. For both substances, individuals reporting higher peer-use estimates were also significantly more likely to report heavy use. Occasional cigarette users reported significantly lower perceived risk of regular cigarette use than did nonusers. Occasional marijuana users reported significantly lower risk of occasional marijuana use and higher perceived risk of other drug use than did nonusers, but reported no differences in the level of perceived risk of regular use.

As shown in Figures 1 and 2, compared to heavy users, occasional users of cigarettes or marijuana reported lower rates of illicit drug use, heavy alcohol use, and high-risk driving behaviors as well as fewer problem behaviors and higher grades. Conversely, occasional users showed less favorable levels of each variable relative to nonusers. All groups in Figures 1 and 2 differ significantly (p < .001). Nonusers of cigarettes and marijuana were significantly less likely to be moderate alcohol drinkers than were heavy and occasional users, whereas heavy and occasional users did not differ with regard to moderate alcohol use (data not shown).

D I S C U S S 1 O N

Consistent with Hypothesis 1, high school seniors who were occasional users of cigarettes or marijuana possessed stronger beliefs about the potential harm of regular use than did heavy users. Consistent with Hypothesis 2, this difference was specific to perceived risk of regular use, as occasional users and heavy users did not generally report different beliefs regarding the risk of occasional use. These findings were still evident after adjustment for several potential mediating variables (i.e., confounds) including self-esteem, estimates of peer use, and close friends' approval of drug use.

One conclusion that can be drawn from these findings is that adolescents who occasionally use cigarettes and marijuana may moderate their level of use to minimize potential harmful effects associated with heavier use. Additionally, occasional users appear to exhibit more favorable levels of other substance use as well as high-risk driving and problem behaviors. In other words, occasional users may be practicing harm reduction. The findings are bolstered by the fact that both occasional and nonusers differed with regard to both perceived risk of regular and occasional use, whereas occasional and heavy users generally differed only with regard to perceived risk of regular use (i.e., heavy use). Additionally, while heavy and occasional cigarette and marijuana users differed with regard to heavy alcohol use, they did not differ with regard to moderate use.

Study limitations and alternative interpretations

Prior to discussing potential intervention and policy implications of these findings, we first consider alternative interpretations and weaknesses of the study. Psychosoeial and behavioral differences between occasional cigarette and marijuana users may be less meaningful if occasional users were recent initiators, and therefore their current level of use may only be a transient stage preceding heavier use. To address this issue, we used data from Forms 1, 3, and 6 regarding age of initiation, and classified as occasional users only those who reported initiating use prior to Grade 11. That is, we attempted to distinguish between long-term occasional users and recent initiators. Approximately 62% of occasional smokers and 52% of occasional marijuana users reported initiation prior to Grade 11. With occasional use coded this way, the pattern of results for both cigarette and marijuana use was similar to that observed with the original coding scheme-that is, occasional and regular users differed only with regard to perceived risk of regular use, but not risk of occasional use (data not shown).

Causal inferences regarding the association of perceived risk and level of substance use must be tempered by the cross-sectional nature of the study. Attitude may result from rather than cause behavior change. And levels of perceived risk regarding regular use may not predict progression to heavy use. Results from correlational and intervention studies addressing cigarette and alcohol use, however, suggest that attitude is antecedent to change in drug use rather than a consequence of it. (Botvin et al,1992; Downs, 1987). Another alternate interpretation is that perceived risk may be only a marker for personality characteristics such as risk taking, law-abidance, self-esteem, or future orientation (Bentler,1987), ¢haracteristics that may be difficult to alter. If so, modifying levels of perceived risk, without altering the associated personality traits, may have little impact on substance-use behaviors. Several observations argue against this conclusion.

analyses adjusting for self-esteem (reported in Tables 4 and 5) the association between perceived risk of regular substance-use and level of use remained analyses (not shown) that included problem behaviors, high risk-driving, and hard drug use (as surrogate measures of risk taking) as covariates, the pattern of findings was similar.'

Second: most other studies have found no differences in personality characteristics between occasional and heavy users of cigarettes and marijuana (Hajek et al., 1995; Kimlicka, 1978; Kouri et al., 1995).2

Third, perceived harm appears to be a reason for discontinuing drug use among college youth (Martin, Duncan, & Zunich, 1983) and reduction in substance use following exposure to an educational program has been shown to be mediated by changes in perceived harm (Botvin & Dusenbury, 1992).

Research and program implications

That perceived risk was associated with substance-use behaviors is consistent with several models of health behavior, most notably the Health Belief Model and Theory of Reasoned Action. which postulate perceived risk or similar constructs as a central (and causal) factor in health-behavior decisions (Becker, 1977; Budd et al., 1983;

 

Perry, Baranowski & Parcel, 1990). Which dimensions of perceived risk regarding substance use are salient to youth cannot be answered by this study as questionnaire items queried only general physical and other risks, but did not differentiate between longterm and short-term risk, nor risk of spec physical, social, or legal consequences (e.g., heart disease, cancer, cognitive impairment, or arrest). Previous studies suggest that effects regarding short-term negative health consequences may be more salient to adolescents than effects regarding long-term health consequences, though the data are not entirely consistent (Botvin & Dusenbury, 1992; McCaul & Glasgow, 1985). Qualitative studies (e.g., focus groups) conducted with occasional users may help elucidate the cognitive processes adolescents employ regarding their involvement in substance use as well as other risk behaviors. Controlled intervention studies that focus on modification of perceived risk as a target outcome, and that examine the, association between change in perceived risk and substance use, are needed to better elucidate the causal role of perceived harm in adolescents' substance-use decisions.

That occasional cigarette and marijuana users generally showed more favorable risk-behavior profiles (e.g., less engagement in other substance use and problem behaviors) relative to heavy users suggests that discouraging heavy use may yield public health benefits. Discouraging heavy use may have a place in primary prevention programs, as well as a secondary prevention strategy for discouraging progression to heavy use among current users.

Conversely, because occasional users showed less favorable behavioral profiles relative to nonusers and most drug use carries increased physical and social risk (relative to abstinence), the possibility exists that encouraging occasional use may have undesired effects among individuals who might otherwise not initiate use. Designing programs and messages that selectively discourage heavy use among individuals who are occasional users or are prone to occasional use, without encouraging initiation among nonusers or those unlikely to initiate use, poses a major challenge for intervention developers. Developing such selective intervention programs will require considerable understanding of how and why some teens choose to control their level of substance use, as well a$ how and why occasional users avoid involvement in other problem and high-risk behaviors. One of the concerns regarding sexual education programs based on harm reduction, as opposed to abstinence-based principles voiced by some, is that such programs (may) increase sexual experimentation, this despite considerable evidence to the contrary (Jacobs & Wolf, 1995; Kirby et al., 1994). Similar concerns are likely to be raised regarding harm-reduction approaches to substance-use prevention, which may impede investigators' ability to conduct this line of research.

In addition o discouraging heavy use, other elements of drug education programs based on harm-reduction principles might include discouraging progression from using one drug or less dangerous drugs to using multiple drugs or more dangerous combinations of drugs. Just as the Dutch have demonstrated that a separation of marijuana markets from those for harder drugs can be achieved through policy changes, it may also be possible for educational programs to "separate" use from abuse (Drucker, 1995). Our findings suggest that educational efforts should focus on discouraging more harmful levels and patterns of use as well as involvement in undesired behaviors associated with use and intoxication (not only with regard to alcohol, but all mood-altering substances) such as driving under the influence and violence (Duncan et al., 1994).

Understanding controlled drug use, and ultimately reducing more dangerous patterns of use and the undesired sequellae, does not necessitate legalization of illicit drugs, nor the promotion of "recreational" use. Equating harm reduction with "drug legalization" or "pro-drug" policies may marginalize harm reduction and limit the ability of researchers to examine these important public health questions. Harm reduction is based on the observation that mood-altering substances have been, and will likely continue to be, part of the human experience; and thus there are benefits to acknowledging and understanding this phenomenon as opposed to condemning it.

Policy implications

This article raises the possibility that decreasing heavy cigarette and marijuana use, as well as use of other drugs, among U.S. teens may be achievable through programs that specifically increase perceived risk of heavy use. According to Problem Behavior Theory, one reason why behaviors such as smoking, illicit drug use, school dropout, and unsafe sexual practices cluster within individuals is that society views each of these behaviors as unacceptable, or in the case of illicit drugs, deviant or antisocial (Jessor, 1987; Jessor & Jessor, 1977; Resnicow, Ross-Gaddy, & Vaughan, 1995). Accordingly, those adolescents who occasionally use drugs may feel ostracized, that they have crossed a boundary of decent and acceptable conduct, and have become part of a "deviant" subculture, where high-risk behaviors are both more prevalent and acceptable (Duncan et al., 1994; Klonoff & Clark, 1976). This suggests that if adolescents who are occasional cigarette or marijuana users felt less disapproval or marginalization for their behavior, they may be less likely to progress to higher levels of drug use or other antisocial behaviors.3 In other words, contrary to most current substance-use prevention programs, it may be beneficial to communicate to youth that not all drug use is equally undesirable nor all drug users undesirables (Duncan et al., 1994; Resnicow & Botvin,1993). As noted by Newcomb and Bentler (1989):

Not all drug use [author's italics] is bad and will fry one's brain. Such claims, as reflected in the national 'hysteria and depicted in media advertisements for treatment programs, repeat the failed scare tactics of the past. All drug abuse (author's italics] is destructive and can have devastating consequences for individuals, their families, and society. The difference or distinction lies in the use versus abuse of drugs. (p. 247)

Zero-tolerance campaigns and fear messages that address only the negative, consequences of substance use may be less credible to adolescents than programs acknowledging both the positive and negative aspects as well as the fact that most people who use substances, particularly cigarettes, alcohol, and marijuana, are "mainstream" citizens, not criminals or derelicts (Duncan et al., 1994). Another issue that warrants investigation is the extent to which occasional use, particularly of cigarettes, can be sustained. There is some evidence that individuals, so-called chippers, can maintain relatively low rates of cigarette use over many years (Hajek et al., 1995; Owen et al., 1995, Shiffman, 1989, Shiffman, Kassel, Paty, Gnys, & Zettler-Segal, 1994). However, additional research is needed to determine whether "heavy use" prevention programs can promote long-term occasional use, and do so without increasing the number of occasional users. Despite the moral and political objections often levied against harm-reduction programs (Des Jarlais, Paone. Friedman, Peyser, & Newman,1995; Drucker, 1995; Duncan et al., 1994; Nadelman, 1989; Neutens,1992; Stout & Kirby, 1983), further research examining the phenomena of occasional (if not controlled or responsible) drug use, and incorporating these findings into prevention and cessation programs, appears warranted.

REFERENCES

Bachman, J. G., Johnston, L. D., & O'Malley, P. (1990). Explaining the recent decline in cocaine use among young adults: Further evidence that perceive risks and disapproval lead to reduced drug use. Journal of Health & Social Behavior, 31(2), 173-184.

Bachman, J. G., Johnston, L D., O'Malley, P., & Humphrey, R. H. (1988). Explaining the recent decline in marijuana fuse: Differentiating the effects of perceived risks, disapproval, and general lifestyle factors. Journal of Health & Social Behavior, 29(1), 92-112.

Bachman, J. G:, Johnston, L D., & O'Malley, P. M. (1996). Monitoring the Future. A continuing study of the lifestyles and values of youth, 1994 (Computer file distributed through the Inter-university Consortium for Political and Social Research). Ann Arbor, Ml: University of Michigan, Survey Research Center.

Becker, M. (197'7). Selected psychosocial models and correlates of individual health-related behaviors. Med ical Care, $,27-47.

Bentler, P. M. ;(1987). Drug use and personality in adolescence and young adulthood: Structural models with nonnormal variables. Child Development, 58(1), 65-79.

Botvin, G., Baker, E., Botvin, E., Dusenbury, L., Cardwell, J., & Diaz, T. (1993). Factors promoting cigarette smoking among black youth: A causal modeling approach. Addictive Behaviors, 1& 397-405.

Botvin, G., &   Dusenbury, L (1992). Smoking prevention among urban minority youth: Assessing effects on outcome and mediating variables. Health Psychology, 11 290-299.

Botvin, G. J., Baker, E., Goldberg, C. J, Dusenbury, L, & Botvin, E. M. (1992). Correlates and predictors of smoking among black adolescents. Addictive Behaviors, 17, 97-103.

Budd, R., Bled   r, S., & Spencer, C. (1983). Exploring the use and non-use of marijuana as reasoned actions: An appli tion of Fishbein and Ajzen's methodology. Drug & Alcohol Dependence, 11, 217-224.

Cigarette smoking among American teens rises again in 1995 (Press Release 12). (1995a). Ann Arbor. University of Michigan.

Clifford, P. R,   mundson, E. W., Koch, W. R., & Dodd, B. G. (1991). Drug use and life satisfaction among college students. International Journal of the Addictions, 26,45-53.

Des Jarlais. D., Friedman, S., & Ward, T (1993). Harm reduction: A public health response to the AIDS epidemic among injecting drug users. Annual Review of Public Health, 14, 413-450.

Des Jarlais, D., Paone, D., Friedman, S., Peyser, N., & Newman, R. (1995). Regulating controversial programs for unpopular people: Methadone maintenance and syringe exchange programs, American Jour nal of Public Health, 85,1577-1584.

Donaldson, S. L, Graham, J. W., & Hansen, W. B. (1994). Testing the generalizability of intervening mechanism theories: Understanding the effects of adolescent drug use prevention interventions. Journal of Behavioral Medicine, 17, 195-216.

Downs, W. R. (1987). A panel study of normative structure, adolescent alcohol use and peer alcohol use. Journal of Studies on Alcohol, 48,167-175.

Drucker, E. (1995). Harm reduction: A public health strategy. Current Issues in Public Health, 1, 64-70.

Drug use rises again in 1995 among American teens (Press Release 13). (1995b). Ann Arbor. University of Michigan.

Duffy, J. C. (1995). Alcohol consumption and all-cause mortality. International Journal of Epidemiology, 24, 100-105.

Duncan, D. F., Nicholson, T., Clifford, P., Hawkins, W., & Petosa, R (1994). Harm reduction: An emerging new paradigm for drug education. Journal of Drug Education, 24,281-290.

Fuchs, C. S., Stampfer, M. J., Colditz, G. A., Giovannuod, E., Manson, J. E., Kawachi, I., Hunter, D. J., Hankinsorn S. E., Hennekens, C. H., Rosner, B., Speizer, F. E., & Willet, W. C. (1995). Alcohol consumption and mortality among women. New England Journal of Medicine, 332,1245-1250.

Hajek, P., West R., & Wilson, J. (1995). Regular smokers, lifetime very tight smokers, and reduced smokers: Comparisop of psychosocial and smoking characteristics in women. Health Psychology, 14,195-201.

Jacobs, C. D., do Wolf, E. M. (1995). School sexuality education and adolescent risk-taking behavior. Jour nal of School Health, 65, 91-95.

Jessor, R. (1910. Problem-behavior theory, psychosocial development, and adolescent problem drinking. British Journal of Addiction, 8Z, 331-3'42.

Jessor. R., & Jqssor, S. L (1977). Problem behavior and psychosocial development. A longitudinal study of youth. New York: Academic Press.

Johnston, L, O'Malley, P., & Bachman, J. (1995). National survey results on drug use from the Monitoring the Future Study, 1975-1994 (Vol. 1). Rockville, MD: Department of Health and Human Services.

Kandel, D., Yaonaguchi, K., & Chen, K. (1992). Stages of progression in drug involvement from adolescence to adulthood: Further evidence for the gateway theory. Journal of Studies on Alcohol, 53,447-457.

Kimlicka, T. M. (1978). A comparison of chronic versus casual marijuana users on personal values and behavioral orientations. International Journal of the Addictions, 13,1145-1156.

Kirby, D., Short, L., Collins, J., Rugg, D., Kolbe. L, Howard, M., Miller, B., Sonenstein, F., & Zabin, L (1994). School-based programs to reduce sexual risk behaviors: A review of effectiveness. Public Health Reports, 109, 339,360.

Klonoff, H., & Clark, C. (1976). Drug patterns in the chronic marijuana user. international Journal of the Addictions, 11, 71-80.

Kondo, K., Tsuzuki, H., Sasa, M.. Sumitomo, M., Uyama, T., & Monden, Y. (1996). A dose-response relationship between the frequency of p53 mutations and tobacco consumption in lung cancer patients. Journal of Surgical Oncology, 61, 20-26.

Kouri, E., Pope, H. G., Jr.. Yurgelun-Todd, D., & Gruber, S. (1995). Attributes of heavy vs. occasional marijuana smokers in a college population. Biological Psychiatry, 38,475-481.

Martin, C. E., Duncan, D. F., & Zunich, E. M. (1983). Students' motives for discontinuing illicit drug use. Health Values, 7, 8-11.

McCaul, K., & Glasgow, R. ('1985). Preventing adolescent smoking: What have we learned about treatment construct validity? Health Psychology, 14, 361-387.

McWhorter, W. P., Boyd, G. , M., & Mattson. M. E. (1990). Predictors of quitting smoking. Journal of Clini cal Epidemiology, 43,1399-1405.

Menhiratta, S., Wig, N., & Verma, S. (1978). Some psychological correlates of long-term heavy cannabis users. British Journal of Psychiatry, 132, 482-486.

Nadelman, E. (1989). Drug prohibition in the United States: Costs, consequences, and alternatives. Science, 245,939-947.

Neutens, J. J. (1992). Sexuality education in comprehensive school health programs: Surviving the "moral smog". Journal of School Health, 62, 74-75.

Newcomb, M., & Bender, P. (1989). Substance use and abuse among children and teenagers. American Psy chologist, 4, 242-248.

O'Malley, P., & Bachman. J. (1983). Self-esteem: Change and stability between ages 13 and 23. Develop mental Psychology, 19, 257-268.

Owen, N., Kent, P., Wakefield, M., & Roberts, L (1995). Low-rate smokers. Preventive Medicine, 24,80-84.

Perry, C., Baranowski, T., & Parcel, G. (1990). How individuals, environments and health behavior interact: Social leaming theory. Its K. Glanz, F. Lewis, & B. Rimer (Eds.), Health behavior and health education: Theory, research, and practice (pp. 161-186). New York: Jossey-Bass.

Poikolainen, K. (1995). Alcohol and mortality - A review. Journal of Clinical Epidemiology, 48, 455-465.

Primavera, L H., & Pascal,   . (1986). A comparison of male users and nonusers of marijuana on the perceived harmfulness of drugs. American Journal of Drug & Alcohol Abuse,12(1-2), 71-77.

Resnicow, K., & Botvin, G. (1993). On the effects of school health education programs: Why do they decay? Preventive Medicine, 22.484-490.

Resnicow, K., Kabat, G., & Wynder, E. L. (1991). Progress in decreasing cigarette smoking. In V. T. Devita, S. Hellman. & S. Rosenberg (Eds.), Cancer: Important advances in oncology 1991 (pp. 205-213). Philadelphia: Lippincott.

Resnicow, K., Ross-Gaddy, D., & Vaughan, R. D. (1995). Structure of problem and positive behaviors in African American youth$. Journal of Consulting & Clinical Psychology, 63,594-W3.

Rosenberg. M. (1979). Conceiving the self. New York: Basic Books.

Serxner, S., Catalano, R., Dooley, D.. & Mishra, S. (1992). Influences on cigarette smoking quantity. Selection, stress, or culture? Journal of Occupational Medicine, 34,934-939.

Shedler, J., & Block, J. (1990), Adolescent drug use and psychological health: A longitudinal inquiry. Amer ican Psychologist, 45.612L-630.

Shiffman, S. (1989). Tobacco "chippers"-individual differences in tobacco dependence. Psychopharmacology, 97,539-547.

Shiffman, S., Kassel, J. D., Paiy, J., Gnys, M., & Zettler-Segal, M. (1994). Smoking typology profiles of chippers and regular smokers;. Journal of Substance Abuse, 6,21-35.

Stout, J., & Kirby, D. (1983). The effects of sexuality education on adolescent sexual activity. Pediatric Annals, 22, 120-126.

Torabi, M., Bailey, W., & Majd-Jabbari, M. (1993). Cigarette smoking as a predictor of alcohol and other drug use by children and adolescents: Evidence of the "gateway drug effect." Journal of School Health, 63,302-306.

Vega, W., Zimmerman. R., Warbeit. G., Apospori, E., & Gil, A. (1993). Risk factors for early adolescent drug use in four ethnic and racial groups. American Journal of Public Health, 83,185-189.

West. R. (1992). Smoking: its influence on survival and cause of death. Journal of Royal College of Physi cians of London, 26, 357-366.

Willard. J., & Schoenborn. C. (1995). Relationship between cigarette smoking and other unhealthy behaviors among our nation's youth: United States 1992. Advance Data, 263, 1-11.

Wilson, D., Wakefield, M., Owen, N., & Roberts. L (1992). Characteristics of heavy smokers. Preventive Medicine, 21,311-319.

Zang, E., & Wynder, E. (1992). Cumulative tar exposure. A new index for estimating lung cancer risk among cigaret smokers. Cancer, 70.69-76.

 

1. In this analysis for cigarette use, the odds ratio (OR) for perceived risk of regular use was similar in magnitude (OR = .71) to the OR reported in Table 4; however. the confidence interval (Cl.38,1.4) included 1.00.

2. ln one study that found significant personality differences between frequent and occasional users, the algorithm used to classify level of use differed from that used here (Shedler & Block, 1990).

3. Negative social norms regarding substance use may have a protective effect by suppressing heavy use among occasional users. However. that perceptions regarding friends' approval of use did not significantly differ between heavy and occasional cigarette or marijuana users (see Tables 4 and 5) would seem to argue against this interpretation.