Smoking, Drinking, and Drug Use in Young Adulthood

The Impacts of New Freedoms and New Responsibilities

Jerald G. Bachman, Katherine N. Wadsworth, Patrick M. O'Malley, Lloyd D. Johnston, John E. Schulenberg

• Publisher: Psychology Press
• Year: 2013
• ISBN:
  9781134803538
• Edition: 1
• Topic: Psychology
• Subtopic: History & Theory in Psychology
• Index: Psychology

Why do some young adults substantially change their patterns of smoking, drinking, or illicit drug use after graduating from high school? In this book, the authors show that leaving high school and leaving home create new freedoms that are linked to increases in the use of cigarettes, alcohol, marijuana, and cocaine. They also show that marriage, pregnancy, and parenthood create new responsibilities that are linked to decreases in drug use. The research is based on more than 33, 000 young people followed from high school through young adulthood by the nationwide Monitoring the Future project. Every two years, participants reported on their drug use, as well as their schooling, employment, military service, living arrangements, marriages, pregnancies, parenthood, and even their divorces. The unique qualities of this research--large nationally representative samples, follow-ups extending up to 14 years beyond high school, and multiple approaches to analysis and data presentation--allowed the examination of several important influences simultaneously, while retaining much of the rich detail encountered in the real world. On the whole, the results are encouraging, suggesting that the potentials for change and improvement during the transition to adulthood are as important as the detrimental effects of problem behavior in adolescence. This research is a "must" read for anyone concerned with how new freedoms and responsibilities impact adolescents, young adults, and the use of licit and illicit drugs.

1

Introduction and Overview

Since the 1960s, few issues have captured more intense and enduring interest among so many different segments of our society than drug use among young people. From a public health perspective, anyone’s use of licit or illicit drugs can be risky. However, when such use occurs among the nation’s youth—the nation’s future—concerns about potential health risks are joined with concerns about wasted potential, about significantly and negatively altered lives, about lifelong negative consequences, and even about lifelong drug abuse.

The ravages of youthful drug use have received much attention, as they should. It is clear that for some young people, illicit drug use does significantly and negatively alter their lives and those of their families and communities. Two facts remain, however. First, most young people who use illicit drugs do not use them to excess. Second, most who use illicit drugs eventually “mature out” of their drug use as they move into adulthood. Stating these facts is not an endorsement of a laissez-faire policy toward adolescent drug use—rather, it helps place adolescent drug use in a developmental context and underscores the need for continued scientific study about the natural history of drug use from adolescence through adulthood.

CHANGE AND STABILITY IN DRUG USE

Many aspects of personality and patterns of behavior are well developed by late adolescence. By the time young people graduate from high school and enter into young adult roles, many abilities and interests have been formed, along with patterns of interpersonal skills and relationships. These profoundly affect individuals throughout their adulthood.

Something else happens to most young people in the United States before they complete high school: they develop patterns of drug-using behavior. Among the nationally representative samples of high-school seniors we studied during the past two decades, (a) two-thirds or more had tried cigarettes, and one in five was a daily smoker; (b) four-fifths or more had tried alcohol, and more than half had been drunk at least once during the past year; (c) half or more had tried at least one illicit drug, including one-quarter to one-half (depending on the year) who had used marijuana during the past year.

What happened to their drug-using behaviors after graduation? Did the patterns of drug use continue to be much the same as they were during the end of high school? Did the new freedoms that many experienced lead to increased substance use? Did their new responsibilities in marriage and parenthood lead to decreases in use? Each of these three questions implies a different pattern of change or stability; nevertheless, the answer to each is clearly “Yes."

First and foremost, drug use remains fairly stable in individuals over time. If we want to know whether an individual will use a given drug during young adulthood, by far the best predictor is the individual’s use of that drug during the high-school years. In particular, cigarette use shows a great deal of consistency across time; most young adult users were regular daily users before they left high school. Although use of alcohol and use of illicit drugs does not reach the point of chemical dependency among the large majority of high-school users, these behaviors also show considerable constancy. This stability is hardly surprising; throughout high school young people develop and express appetites, attitudes about taking risks, patterns of recreational behavior, and friendships, all of which can influence drug use, and many of which do not change substantially on departure from high school.

However, many things do change after high school. New freedoms from the supervision of parents and teachers, as well as new “adult” legal status, provide greater opportunities for drug use than existed during high school. Conversely, new role responsibilities, such as becoming engaged, marrying, becoming pregnant, and having children, all have the potential to inhibit drug use.

In this book we focus less on stability and more on the changes in drug use that occur during young adulthood. Specifically, we examine drug use during the late teens, the twenties, and the early thirties, contrasting levels of use at those ages with the levels of use observed at the end of high school. Most importantly, we focus on just how those changes in drug use seem to be linked to changes in role experiences and responsibilities.

Here we concentrate on four widely used substances—two legally available (to adults), and two illegal. Cigarettes are used regularly by fewer than one-third of young adults, but the large majority of those are daily users who have formed a habit that is very difficult to break, a habit that is widely recognized to carry deadly risks. Alcohol is used by nearly all young adults;although some use it only in moderation, substantial proportions at least occasionally use it to excess, which can involve serious risks to self and others. Marijuana is the most widely used of the illicit drugs; it can have serious consequences itself, and can also provide a “gateway” to the use of other illicit drugs. Cocaine is used by fewer young adults than is marijuana, but can be much more dangerous in its potential for generating acute adverse health effects, and for developing long-term dependency. We see here that there are some important parallels across several of these drugs in terms of their links with post-high-school roles and responsibilities, but we also see some interesting distinctions among the drugs.

DATA QUANTITY, QUALITY, AND BREADTH OF GENERALIZATION

Our findings are drawn from the Monitoring the Future project. For two decades we have been studying the drug use habits of young adults, surveying them first as high-school seniors and then resurveying them throughout their twenties and into their thirties. We have followed them every 2 years as they left their parents’ homes and entered new living arrangements and new roles as college students, employees, spouses, and parents. More than 33,000 young adults surveyed in the high-school classes of 1976–1994, who also completed follow-up questionnaires from 1977 to 1995, contributed to the data presented in this book. Those from the earliest classes participated in seven follow-ups extending to 14 years beyond high school; those from later classes provided proportionately fewer follow-ups. The total number of questionnaires included in these analyses exceeds 139,000, representing more than 100,000 respondent hours.

Lest these samples seem extravagantly large, we should note that some of the drug-using behaviors routinely reported by the Monitoring the Future project are relatively rare and thus require large samples for reliable detection and meaningful examination (Johnston, O’Malley, & Bachman, 1997). Moreover, some of the post-high-school experiences that are the focus of the present analysis also involve fairly small proportions of the total population at any single point in time (e.g., military service, pregnancy, living in a dormitory). Thus a great advantage of the present sample, or compilation of samples, is our ability to focus not just on the most prevalent post-high-school experiences, but also on other experiences that may have substantial impacts on drug use.

In addition to being quite large and extending across two decades, our samples are nationally representative. Because of this broad coverage in terms of both time and geography, the findings are more useful descriptively and more broadly generalizable than is typically the case for panel studies of drug use. Having said that, however, we must also caution that our findings do not represent the entire age cohort. Our initial sampling did not include the approximately 15% of young people who do not graduate from high school, and it also underrepresents frequent absentees (Johnston et al., 1997). Further, although the large majority of those asked to participate in follow-ups did so, the loss of some respondents further restricts our ability to generalize. Finally, for reasons noted later, our present analyses omit the small proportion of respondents who, as high-school seniors, were already married, living away from parents or guardians, had children, or any combination thereof. (The samples and survey methods are described in greater detail in the Appendix, and still more information on sample characteristics is provided in a supplementary publication available from the authors; Bachman, O’Malley, et al., 1996.)

The quality of survey data depends, of course, on the ability and willingness of those surveyed to provide accurate responses. Accuracy is easy when respondents are asked about their marital or parental status, but it can be more difficult when they are asked about their use of various drugs. In the case of cigarettes, young adults tend to use them on a regular daily basis, if they use them at all, and thus can report such use with a good deal of accuracy. However, alcohol use for many is less routine, and therefore more subject to errors in recall. Furthermore, most who use marijuana or cocaine do so infrequently, and the fact that such use is illegal might inhibit reporting. Such problems in self-reports of drug use have been widely discussed and studied, using many different data sets including our own. We say more about these matters in the Appendix, and cite relevant studies. For present purposes, it is important to stress our confidence that reporting problems have not invalidated our findings. That said, we should add that, to the extent that biases remain, we suspect they lie largely in the direction of underreporting.

In summary, these survey findings, like virtually all others, have inevitable limitations in quality and generalizability. Nevertheless, we think the strengths of this study design are substantial, and we believe that the findings reported here provide important information about the early adult experiences of a large majority of high-school graduates during the past two decades. Moreover, the consistency of our key findings across a number of different analyses at different stages in time leaves us confident that many of the basic relationships found for these samples are broadly generalizable to the population of young adults in America.

STUDYING CAUSAL RELATIONSHIPS IN NATURAL SETTINGS

The Monitoring the Future project, like many other social psychological studies, examines individuals (survey respondents) in “natural settings.” This stands in sharp contrast to the experimental method, in which individuals (subjects) are placed in treatment groups by a process of random assignment intended to cancel out “extraneous” individual differences. A fundamental problem with survey studies of individuals in natural settings is that assignment to various social contexts is anything but random. The freedoms and responsibilities experienced in early adult roles (e.g., college, military service, civilian employment; living with parents, a spouse, others) differ from one individual to another largely as a result of the individuals’ own backgrounds, prior behaviors, and preferences, as well as the (nonrandom) choices of many other people. Thus, even when the present analyses show that important differences in drug use are correlated with certain post-high-school experiences, we still have to make informed judgments about whether the changes in drug use were caused by those particular post-high-school experiences.

It is axiomatic that studies in natural settings can provide no certain demonstration of causation—the possibility always remains that other (perhaps unexamined) factors are more fundamental causes, and that our conclusions about causation are therefore spurious. Panel data, obtained by following the same survey respondents across a number of years, can be very helpful in ruling out some such alternative explanations and in demonstrating temporal sequences of events consistent with certain interpretations; however, panel data alone are not sufficient to establish causation with certainty.

It is also axiomatic that studies in natural settings can lead to erroneous conclusions if the analyses fail to include all important explanatory variables (i.e., if the implicit or explicit causal model is misspecified). One cannot argue with that assertion, at least in principle. In practice, however, the problem is how to figure out everything that might be relevant, measure it accurately, and then fit it all into an analysis model without overtaxing the capabilities of either the computer programs or the human interpreters. In the present monograph, our analysis strategy for dealing with this practical problem has three important features. First, we have focused on one broad area—post-high-school roles and experiences—in considerable detail, while attempting to provide adequate statistical controls for other relevant factors. Second, as outlined in chapter 3, we have streamlined our analyses in a number of important respects so as to make them more manageable and approachable. Third, we have chosen analysis methods that are in some respects intentionally neutral or agnostic about causal order, even though we ourselves are not so neutral and argue that some causal interpretations are more plausible than others.

CHAPTER OUTLINE AND PREVIEW OF KEY FINDINGS

Chapter 2 provides a review of relevant literature, summarizing previous findings and relevant theory that have inspired and informed the present work. We then turn to analysis of the Monitoring the Future panel data.

Chapter 3 describes our samples in terms of roles and experiences at each of seven follow-up points, with modal ages ranging from 19 to 32. We document also how some experiences in young adulthood seem to go hand in hand with others. In the first years after high school, for example, those who enter college are more likely than average to leave their parents’ homes, but less likely than average to marry or have children.

Chapters 4 through 7 examine how usage patterns for four substances—cigarettes, alcohol, marijuana, and cocaine—change during young adulthood. We also examine how individuals’ post-high-school drug use differs from their use during senior year, and we link those changes to their roles and experiences during young adulthood.

The final chapter summarizes and integrates our findings, and shows how they support our conclusions. Before leaving this introductory chapter, however, it may be helpful to preview several of the general conclusions that are illustrated again and again in the following chapters.

First, it bears repeating that use of any drug during the post-high-school years is predicted to a very considerable extent by the use of that drug during high school. This stability in drug using behaviors during late adolescence and early adulthood very likely reflects (a) relatively stable individual differences in personality (e.g., sensation seeking, willingness to take risks, nonconformity), as well as (b) relatively stable differences in social environments (e.g., friendships, behaviors and attitudes of friends, values and preferences of parents).

Second, we find that changes in drug use do occur between high school and young adulthood, and some of these changes are clearly linked to the roles and environments of young adulthood. Moreover, a central finding is that the current experiences are what seem to make the difference. Thus, for example, being married is linked with lower drug use, and it seems to make relatively little difference whether one has been married for nearly a decade or for only a year or two.

Third, the different post-high-school roles and experiences are complexly interrelated. Although our multivariate analyses provide some help in disentangling patterns of relationships, problems of overlapping effects remain and require some causal interpretations. Thus, for example, we observe greater than average increases in alcohol use among college students, and we attribute that to the marital status and living arrangements usually associated with college attendance.

Fourth, the impacts of most roles and environments in young adulthood seem widely generalizable. That is, the impacts are much the same whether measured in the first follow-up or two (i.e., late teens or early twenties), or in the sixth or seventh (late twenties or early thirties). Moreover, most of the impacts are very similar for men and women. One notable exception is that pregnancy has substantial impacts on the drug use of women, whereas among men the impacts of having a pregnant spouse are more limited.

Fifth, on average, the usage rates for most substances decline during the mid-twenties, and that seems largely attributable to the fact that increasing proportions of individuals during that period become married, and many take on the additional responsibilities of pregnancy and parenthood. In other words, these age-related changes are not simply “maturation” in general; rather, they reflect the impacts of specific new roles and living arrangements undertaken by increasing proportions of young adults as they move through their twenties.

Finally, we note that although there are many similarities across the four drugs in their relationships with post-high-school experiences, there also are clear differences. For example, patterns of cigarette use are more stable than those involving the other drugs, largely due to the high proportion of cigarette users who are nicotine dependent; another result of this high level of dependency is that cigarette consumption is less likely to be influenced by post-high-school experiences, compared with use of other drugs.

2

Reviewing the Influence of Social Roles on Drug Use During the Transition to Young Adulthood

This investigation is focused on how the new freedoms and new responsibilities in young adulthood affect drug use. In this chapter we review the empirical and theoretical literature concerning change during the transition to young adulthood. This chapter focuses on the influences of adult social roles on alcohol use, drug use, and other problem behaviors. We have attempted to highlight the central studies that inform this field of inquiry, and to incorporate disparate theoretical perspectives on these issues.

THE TRANSITION TO YOUNG ADULTHOOD

The critical developmental transition from adolescence to young adulthood is characterized by an increase in new and demanding social opportunities and expectations. Plans regarding education and occupation, financial independence, living arrangements, or intimate relationships may be made prior to leaving high school, but the actual experience of the transition may be very different from initial aspirations. Success or failure in each of these areas has important consequences throughout one’s life co...