Initiation of Marijuana Use: Trends, Patterns, and Implications

7. Discussion

Marijuana has continued to be the most commonly used illicit drug among both youths and adults. The rate of marijuana initiation increased during the late 1960s and early 1970s, with a peak in 1976 and 1977 (21.0 per 1,000 potential new users). After that period, the rate of new marijuana users decreased to 8.5 in 1990, followed by an increase to 16.8 in 1996, then a decrease to 13.6 in 1999. Aggregated 1999 and 2000 data indicate that there were approximately 2.5 million new marijuana users annually in 1997-1998 and 2.0 million new users in 1999. Although the number of new users had decreased in 1999, it was still above the levels for 1989 and 1990 (1.4 million each year).

Incidence rates of marijuana use were generally higher among males than among females. Among males, the highest peak was noted in 1976 and 1977 (close to 24.0). Since the early 1980s, incidence rates declined, and the most recent peak for males was in 1997 (19.3). Similar to the pattern of males, the incidence rate for females was lower during the 1980s. The most recent peak for females was in 1996 (15.5). For both genders, the rate in 1999 was lower than the rates from 1996 to 1998. The estimated mean age at first marijuana use was slightly younger among males than females, but the trend suggested an increasing similarity for both genders in some recent years. The mean age of first marijuana use remained at 17 years during most of the 1990s.

Recent initiates were much more likely to be among youths and young adults than among adults aged 21 or older. The annual age- and gender-specific incidence rate per 1,000 potential users was highest among youths aged 15 to 17. Young adults aged 18 to 20 had slightly higher incidence rates than youths aged 12 to 14, while adults aged 21 or older had a very low rate of new use. Recently, American Indians/Alaska Natives appeared to have particularly high rates of new users and younger age of first use than members of other racial/ethnic groups.

Recent initiation of marijuana use also varied by States and age groups. Colorado, Delaware, Massachusetts, New Hampshire, and Vermont were ranked in the category with the highest incidence rate for the overall, youth, and young adult groups. New Mexico had the highest rate for the overall and youth groups. Minnesota was in the highest category for the overall and young adult groups. By comparison, Louisiana had the lowest rate of recent new users for the overall, youth, and young adult groups. In addition, Texas and Utah had the lowest rate of recent initiation among youths and young adults.

Marijuana initiation varied somewhat by month of year. Overall, the annual average number of marijuana initiates per day was 6,186 (3,197 and 2,989, respectively, for males and females). Among males, the highest number of first marijuana users per day was noted in June and July (around 4,300 initiates per day). Among females, increased rates of initiates were in January (3,519 initiates per day) and July (3,625 initiates per day). Of all recent initiates who started using marijuana before age 18, June (11.6 percent of all recent initiates) and July (12.0 percent) also had the most new users, while adult-onset marijuana users tended to have initiated the use in January (10.9 percent), December (9.2 percent), or July (9.1 percent).

Any prior use of alcohol and/or cigarettes was strongly related to the onset of recent marijuana initiation, regardless of the age when alcohol or cigarettes were first used. Of all recent marijuana initiates, 60 percent used both alcohol and cigarettes before starting their marijuana use. Fewer than 10 percent had never used alcohol and cigarettes at the time of marijuana initiation, and the remaining recent initiates had used either alcohol only (16.6 percent) or cigarettes only (14.8 percent) prior to their first marijuana use.

Multiple logistic regression of persons aged 12 to 25 found that younger persons under age 21, males, blacks, American Indians/Alaska Natives, and persons living in the West or metropolitan areas were likely to become new marijuana users recently. For some young adults aged 18 to 25, the assumption of adult roles (e.g., becoming married) may protect them from marijuana initiation. Multiple logistic regression also confirmed that the risk of recent marijuana initiation increased with increasing age among youths aged 12 to 14, but the risk decreased with increasing age among those aged 15 to 25.

Logistic regression models, holding constant demographic characteristics, revealed that earlier age at first use of marijuana, particularly an onset before age 15, predicted increased odds for the following: (a) heavy use of marijuana and other illicit drugs, (b) dependence on or abuse of alcohol and/or other illicit drugs, and (c) dependence on marijuana or other illicit drugs in the past year.

These national estimates on incidence and trend data provide useful information for planning and delivering substance abuse services, as well as baseline data for further investigations. The strength of the association between alcohol and cigarette use with later marijuana initiation suggests that reducing the onset of alcohol or cigarette use may have some positive influences on the prevention of marijuana initiation. Some new marijuana users, particularly early-onset users, may have a significant impact on the future demand for substance abuse treatment because of their increased odds of progressing into heavier drug involvement and probably other drug-related health risk behaviors. Substance abuse service needs for young marijuana users warrant further study because recent data reveal a younger population of drug users seeking treatment for marijuana abuse (OAS, 2001c). Early identification of subgroups vulnerable to heavier substance use and increasing their access to substance abuse services may prevent them from more serious substance use involvement and decrease subsequent societal costs related to substance use or abuse. These national estimates reveal emerging trends of new marijuana use by some subgroups that could be targeted for prevention (e.g., blacks and American Indians/Alaska Natives).

The pathways to illicit drug use and/or dependence involve multiple factors and processes. To better inform policy for substance use prevention, further research should focus on (a) identifying the processes that account for subgroup variations in illicit drug use and/or dependence and (b) understanding the factors that explain the developmental sequence of substance use involvement, including determinants of early initial use of marijuana or other substances.

In conclusion, the data indicate a decreased rate of new marijuana use in 1999. Youths aged 15 to 17 had the highest rate of becoming new marijuana users in recent years. Whites were less likely than blacks to start to use marijuana recently, and young American Indians/Alaska Natives appeared to have an elevated risk of becoming new marijuana users in recent years. Further analysis of these and future NHSDA data will help to confirm the findings from the 1999 and 2000 surveys and to track changes in the trend of new marijuana use by at-risk subgroups.

This page was last updated on June 16, 2008.

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.