1. Introduction - Initiation of Marijuana Use: Trends, Patterns, and Implications Report

Estimates of first-time drug use, referred to as incidence or initiation, provide an important measure of the Nation's drug use problem. They suggest emerging patterns of use and identify periods of heightened risk for an immediate focus on the prevention of substance use, particularly among children and youths. Incidence data also suggest the future burden on substance abuse treatment systems.

This report contains an analysis of the initiation of marijuana use. Marijuana is the most widely used illicit drug in the United States and is in most cases the first illicit drug used by persons who have used an illicit drug. The analysis is based on data from the 1999 and 2000 National Household Surveys on Drug Abuse (NHSDAs). Overall estimates of the prevalence and rate of marijuana initiation based on combined 1999 and 2000 data were released in September 2001 (Office of Applied Studies [OAS], 2001b). Those results showed that, although there was a decrease in incidence from 2.6 million new users in 1996 to 2.0 million in 1999, these levels were still significantly higher than the levels in 1990 (1.4 million). The purpose of this report is to present more in-depth analyses of incidence rates among population subgroups, demographic characteristics and predictors of recent initiates, and consequences of early marijuana initiation. Specifically, this report has four objectives:

The report is organized into seven chapters and two appendices. Chapter 2 describes the data source, measures of key study variables, and statistical methods. Chapter 3 presents incidence rates and trends of marijuana use based on the combined 1999 and 2000 computer-assisted interviewing (CAI) data. Chapter 4 summarizes incidence rates by State using small area estimation (SAE) methods. Chapter 5 examines social and demographic characteristics of recent marijuana initiates and predictors of initiation. This chapter also examines the relationship of marijuana initiation with school status, employment, and marital status among those aged 18 to 25 years. Chapter 6 addresses the relationship between early marijuana use and later drug use patterns, including lifetime and past year use of heroin, cocaine, and psychotherapeutics nonmedically; heavy marijuana use; heavy illicit drug use other than marijuana; abuse of and/or dependence on alcohol or other drugs; marijuana dependence; illicit drug dependence other than marijuana; illicit drug dependence; and alcohol dependence. Chapter 6 also reports findings on the relationship between the age at onset of marijuana use and past year drug dependence among lifetime marijuana users aged 26 or older who also used marijuana in the past year. Chapter 7 provides a summary of overall findings and conclusions and discusses some implications. Appendix A discusses the statistical methods used and the limitations of the NHSDA data, describes the statistical methods for calculating incidence rates and potential biases associated with incidence estimates, and discusses the change in NHSDA measures of substance use initiation and its impact on incidence rate calculation. Appendix B presents selected standard error tables for population estimates in the report. Appendix C provides selected questionnaire pages from the 1999 and 2000 NHSDAs on the demographic and marijuana questions.

Marijuana is the most widely used illicit drug in the United States (OAS, 2001b). According to the 2000 NHSDA, an estimated 14.0 million Americans were current (past month) marijuana users (OAS, 2001b). This represents 6.3 percent of people aged 12 or older and 76 percent of current illicit drug users. Of all current illicit drug users, approximately 59 percent used only marijuana, 17 percent used marijuana and another illicit drug, and the remaining 24 percent used only an illicit drug other than marijuana in the past month (OAS, 2001b).

The NHSDA and the Monitoring the Future (MTF) have shown generally similar long-term trends in the prevalence of substance use among youths, regardless of substantial differences in methodology between the two primary surveys of youth substance use. Between 1999 and 2000, both the NHSDA and MTF found no significant changes in lifetime, past year, and current use of marijuana (Johnston, O'Malley, & Bachman, 2001; OAS, 2001b).

The MTF found that marijuana use rose particularly sharply among 8^th graders in the 1990s, with annual prevalence tripling between 1991 and 1996 (i.e., from 6 to 18 percent) (Johnston et al., 2001). Starting a year later, marijuana use also rose significantly among 10^th and 12^th graders. Following the recent peak in 1996-1997, annual marijuana use declined somewhat in recent years (Johnston et al., 2001).

Although the prevalence of marijuana use has been studied widely, relatively few incidence (first use) data are available. In the first published analysis of national incidence trends, Gfroerer and Brodsky (1992) estimated the number of new users of marijuana and other drugs based on combined data of 1985 to 1991 NHSDAs. They found that fewer than half a million people per year began using marijuana before 1966 and that new use of marijuana began increasing after 1966, reaching a peak in 1973 and declining thereafter. Johnson, Gerstein, Ghadialy, Choi, and Gfroerer (1996) studied the incidence of alcohol, cigarettes, and illicit drugs using data from the 1991 to 1993 NHSDAs. Their investigation found declining trends of marijuana initiation at all ages since at least the late 1970s. However, the mean age of marijuana initiates declined throughout most of the measurement period, from older than 19 years in the mid-1960s to younger than 18 years in the late 1980s and early 1990s. In addition, the rates of marijuana initiation at ages 12 to 17 (youths) and 18 to 25 (young adults) in the early 1990s were still much higher than corresponding rates in the early 1960s.

In recent years, youths aged 12 to 17 have constituted about two thirds of the new marijuana users, with young adults aged 18 to 25 constituting most of the remaining third (OAS, 2001b). Additionally, recent rates of new use among youths in 1996-1998 (averaging 86.4 initiates per 1,000 potential new users) were higher than they had ever been. Nonetheless, rates of new use for both youths and young adults decreased between 1998 and 1999. The average age of marijuana initiation has generally declined since 1965 and remained around 17 years after 1992 (OAS, 2001b).

There are no known prior studies of seasonal patterns in the initiation of substance use. A few studies, however, have looked at seasonal patterns in use. Zingraff and Belyea (1983) suggested a possibility of increased rates of marijuana use during the summer months; other researchers have suggested that variations in activities during the different seasons may influence substance use (McKee, Sanderson, Chenet, Vassin, & Skolnikov, 1998). Kovalenko et al. (2000) studied the seasonality in symptoms of mental and substance use disorders among youths aged 9 to 17 and found a weak seasonality in the counts of symptoms of marijuana use, with estimated zeniths in August and September. The investigators suggested that the possible seasonality in marijuana use may be related to the cycles in school attendance.

On the other hand, one analysis found a lower prevalence of current marijuana during July to September. Using data from 1992-1996 NHSDAs, Huang, Schildhaus, and Wright (1999) examined the seasonality of past month substance use on a quarterly basis. In their logistic regression model controlling for survey year, age, gender, race/ethnicity, and region, current use of the following substances among youths aged 12 to 17 was observed to show seasonal differences: alcohol, an illicit drug except marijuana, marijuana only, an illicit drug, and heavy drinking. Youths were 1.3 times more likely to engage in current marijuana use only in Quarter 4 (October-December) than in Quarter 3 (July-September). Relative to Quarter 3, youths also were 1.2 times more likely to use an illicit drug in Quarter 1 (January-March). Further analyses found that, during Quarter 3, youths were significantly less likely to report being approached by drug dealers in the past month than in the other quarters.

Little research exists on the predictors of marijuana initiation. Van Etten and Anthony (1999) examined the initial opportunity to try marijuana and the transition from first opportunity to first marijuana use using data from the 1979 to 1994 NHSDAs. They found that an estimated 51 percent of U.S. residents had an opportunity to try marijuana. One striking finding is that 43 percent of those with an opportunity went on to first use marijuana within 1 year of the first opportunity (i.e., making a rapid transition). The study also found that males were more likely than females to have an opportunity to use marijuana, but were not more likely to eventually use marijuana once an opportunity was presented. Research has also shown that the risk of initiating marijuana use is associated with age and birth cohort. Chen and Kandel (1995) found that the major risk period for initiation into marijuana was mostly over by age 20. Gfroerer and Epstein (1999) also found that marijuana initiation was unlikely to occur after age 21. Rates of first marijuana use were higher among younger people and cohorts born after World War II than older people and cohorts born before World War II (Johnson et al., 1996; Johnson & Gerstein, 1998).

The onset of marijuana use also is influenced by a variety of personal, family, and community risk and protective factors, such as affiliation with drug-using peers, personality dimensions (e.g., unconventionality), and the parent-child bond (Brook et al., 1999a; Clayton, 1992).

Marijuana has been hypothesized to be a gateway drug for other illicit drug use. Studies by Kandel and other investigators have identified a developmental sequence of drug involvement among youths (Ellickson, Hays, & Bell, 1992; Kandel, Yamaguchi, & Chen, 1992; Yamaguchi & Kandel, 1984). Specifically, the initial use of alcohol and/or cigarettes typically precedes the use of marijuana, which then is followed by the involvement of other illicit drugs. By studying a sample of rural youths, Donnermeyer (1993) also found that early use of alcohol predicted early use of marijuana, which in turn was predictive of early use of other illicit drugs. Studies of age at initiation of drug use confirmed that initiation of alcohol or tobacco typically occurred before marijuana initiation (Costello, Erkanli, Federman, & Angold, 1999; Kosterman, Hawkins, Guo, Catalano, & Abbott, 2000).

Not only does early marijuana use signal an increased risk for hard drug use by grade 10 (Ellickson & Morton, 1999), but it also is associated with drug use problems, dependency, and treatment need (Brook, Richter, Whiteman, & Cohen, 1999b; Clark, Kirisci, & Tarter, 1998; Gfroerer & Epstein, 1999). Among individuals with a history of marijuana dependence, the age at onset of marijuana dependence was younger in the adolescent-onset individuals compared with the adult-onset individuals, and the time from the first use to the onset of dependence also was shorter in the adolescent-onset individuals (Clark et al., 1998). Among middle school students, use of marijuana and other drugs before the age of 12 was found to be associated with engaging in greater numbers of health risk behaviors than among students whose age at onset was 12 years or older or the never users (DuRant, Smith, Kreiter, & Krowchuk, 1999). Early marijuana use is associated with later adolescent problems that limit the acquisition of skills necessary for employment and increased risk of contracting the human immunodeficiency virus (HIV) and using illicit drugs (Brook et al., 1999b). Gfroerer and Epstein (1999) used NHSDA data to examine the impact of marijuana initiation on future drug abuse treatment need and found age at first use of marijuana as the most significant predictor of treatment need in all four age groups (i.e., 12 to 17, 18 to 25, 26 to 34, and 35 or older).

The number of new marijuana users may have a significant impact on the future demand for substance abuse treatment as some new users continue into heavier marijuana use or other illicit drug taking. Consequently, delaying the onset of marijuana initiation could be important in preventing the progression into heavy drug involvement and other drug-related health risk behaviors, as well as in decreasing the social burdens of illicit drug use.

Taken together, studies of marijuana initiation provide vital information for focused prevention programs about the periods of heightened initiation risk, specify subgroups vulnerable to initial use, and generate estimates on treatment needs and future demand for substance abuse treatment.

SAMHSA, an agency in the Department of Health and Human Services, is the Federal Government's lead agency for improving the quality and availability of substance abuse prevention, addiction treatment, and mental health services in the United States.

* Adobe™ PDF and MS Office™ formatted files require software viewer programs to properly read them. Click here to download these FREE programs now