Skip To Content
Click for DHHS Home Page
Click for the SAMHSA Home Page
Click for the OAS Drug Abuse Statistics Home Page
Click for What's New
Click for Recent Reports and Highlights Click for Information by Topic Click for OAS Data Systems and more Pubs Click for Data on Specific Drugs of Use Click for Short Reports and Facts Click for Frequently Asked Questions Click for Publications Click to send OAS Comments, Questions and Requests Click for OAS Home Page Click for Substance Abuse and Mental Health Services Administration Home Page Click to Search Our Site

Prevalence of Substance Use Among Racial & Ethnic Subgroups in the U.S.

Chapter 4 

4.1 Introduction

This chapter compares individuals in eleven racial/ethnic subgroups (see Chapter 2) with respect to the prevalence of cigarette, alcohol, marijuana, cocaine, and any illicit drug use during the past year, need for illicit drug abuse treatment, alcohol dependence, heavy smoking (a pack of cigarettes or more a day), and heavy drinking (five or more drinks per occasion on 5 or more days in the past 30 days). We first describe racial/ethnic differences in the total combined 1991-93 NHSDA sample and then analyze the racial/ethnic differences within subclasses defined by gender and age (age groups 12 to 17, 18 to 24, 25 to 34, and 35 and older).

The main conclusions of this chapter are as follows:

·Cigarette use. Compared with other racial/ethnic subgroups of the U.S. household population aged 12 and over, Native Americans (53%), Puerto Ricans (33%), and non-Hispanic whites (32%) had a relatively high prevalence of past-year cigarette smoking. Asian/Pacific Islanders (22%) had a relatively low prevalence, as did Caribbeans (21%) and Central Americans (18%). Cubans, Mexicans, South Americans, other Hispanics, and non-Hispanic blacks are intermediate. The ranking of subgroups with respect to past-month heavy cigarette smoking is similar.

·Illicit drug use and need for illicit drug abuse treatment. The rankings of racial/ethnic subgroups with respect to past-year illicit drug use (including use of marijuana and cocaine) and need for illicit drug abuse treatment are similar, but not identical, to their rankings with respect to past-year cigarette use and past-month heavy cigarette use. A relatively high prevalence was found among Native Americans (20% used an illicit drug in the past year), Mexicans (13%), Puerto Ricans (13%), and non-Hispanic blacks (13%). A relatively low prevalence was found among Asian/Pacific Islanders (6.5%), Caribbeans (7.6%), Central Americans (5.7%), and Cubans (8.2%). The prevalence was intermediate among South Americans, other Hispanics, and non-Hispanic whites.

·Alcohol use and dependence on alcohol. South Americans (74%) and non-Hispanic whites (69%) are relatively likely to have used alcohol in the past year. In contrast, Asian/Pacific Islanders are relatively unlikely to have done so (53%), as are Central Americans (51%), Puerto Ricans (60%), and non-Hispanic blacks (55%). The rankings of the subgroups with respect to past-month heavy alcohol use and alcohol dependence are similar to each other, but somewhat different from their rankings with respect to past-year alcohol use. Specifically, the prevalence of dependence on alcohol was relatively high among Native Americans (5.6%) and Mexicans (5.6%), and relatively low among Asian/Pacific Islanders (1.8%), Caribbeans (1.9%), and Cubans (0.9%). The prevalence of heavy alcohol use was relatively high among Mexicans (6.9%) and non-Hispanic whites (5.3%), while Asian/Pacific Islanders (0.9%), Caribbeans (2.5%), Central Americans (2.2%), Cubans (2.8%), and South Americans (3.0%) were relatively low.

·Gender differences. Differences between males and females are generally similar across racial/ethnic subgroups for substance use, alcohol dependence, and need for illicit drug abuse treatment. With the possible exception of Native Americans (no significant differences between males and females), males in each of the eleven subgroups are more likely than females in the same subgroup to use substances, to be dependent on alcohol, and to need illicit drug abuse treatment. Gender differences in heavy alcohol use and in alcohol dependence are especially large, with males in the total surveyed population more than twice as likely as females to be dependent on alcohol in the total surveyed population (4.9% versus 2.1%). Gender differences in both heavy alcohol use and alcohol dependence are larger among Mexicans than among other racial/ethnic subgroups. Also, for Cubans and Asian/Pacific Islanders, males were three times as likely as females to report smoking a pack or more of cigarettes per day.

·Age patterns. Age patterns of substance use, dependence on alcohol, and need for illicit drug abuse treatment are generally similar across the eleven racial/ethnic subgroups, with the possible exception of Native Americans (insufficient data). Each of the measures analyzed in this chapter rises to a peak prevalence among 18- to 25-year-olds and then declines, with the exception of heavy cigarette use, which peaks in the 26 to 34 age group. Unlike other measures, cigarette and alcohol use, especially heavy smoking (15% at ages 35 and older), persist at relatively high levels among individuals aged 35 and older. Relative to the total surveyed population, non-Hispanic blacks have low prevalence of use of cigarettes, marijuana, and cocaine at ages 12 to 17, and high prevalence of use of these substances at ages 35 and older. The age patterns reported in this chapter may reflect differences among individuals born during different historical periods (i.e., birth cohort effects) as well as the effects of aging.

The remaining sections of this chapter discuss these findings in detail.

4.2 Patterns of Specific Substance Use

Table 4.1 presents the estimated percentages of individuals in the eleven racial/ethnic subgroups who responded affirmatively to each of nine measures of substance use, dependence, and need for treatment (see section 2.1). In the total surveyed population aged 12 and older in 1991 through 1993, about 31% used cigarettes in the past year, 66% used alcohol, 12% used any illicit drug, 9.0% used marijuana, and 2.5% used cocaine. About 2.7% of the total surveyed population were in need of illicit drug abuse treatment,3.5% were dependent on alcohol, 14% smoked heavily in the past month, and 5.1% drank alcohol heavily in the past month.

Some racial/ethnic subgroups deviate markedly from the substance use patterns of the total surveyed population. Table 4.1 suggests that Native Americans are the highest of the eleven subgroups on six of the nine measures—past-year cigarette use (53%), past-year use of any illicit drug use (20%), past-year marijuana use (15%), past-year cocaine use (5.2%), need for illicit drug abuse treatment (7.8%), and heavy past-month cigarette use (24%). However, given the small sample size of Native Americans (n = 411), the difference between Native Americans and the next highest subgroup is statistically significant for only one of these six measures: past-year cigarette use. (The statistical test results reported in Chapters 4 and 5 are based on two-tail differences-of-proportions tests at the standard .05 significance level; see Appendix A for details.) In general, the small Native American sample size often implies that differences between Native Americans and other subgroups are not statistically significant in this report, even when the differences are substantial in magnitude. Thus, we urge caution in interpreting differences between Native Americans and other subgroups presented in Table 4.1 and in subsequent tables of this chapter.

Nevertheless, the differences between Native Americans and other subgroups are sufficiently large to establish Native Americans as one of the highest subgroups with respect to each of the nine measures, except past-year alcohol use (64% of Native Americans versus 66% of the total population), and past-month heavy alcohol use (4.6% versus 5.1%). For example, with respect to past-year marijuana use, Native Americans (15%) are significantly higher than Asian/Pacific Islanders (4.7%), Caribbeans (5.6%), Central Americans (2.7%), Cubans (5.9%), South Americans (8.4%), and non-Hispanic whites (8.9%).

Mexicans are significantly higher in alcohol dependence (5.6%) and past-month heavy drinking (6.9%) than any other subgroup, except Native Americans (5.6% and 4.6%, respectively). Mexicans are also relatively high in past-year cocaine use (3.9% versus 2.5% in the total target population), and need for illicit drug abuse treatment (3.6% versus 2.5% in the total target population). However, Mexicans are not appreciably higher than other Americans on other measures. For example, Mexicans are about as likely as the total surveyed population to have used any alcohol in the past year (64% versus 66% in the total target population).

Other subgroups are relatively high on particular measures. Non-Hispanic whites and South Americans are relatively high on measures of licit substance use, (both past-year and heavy measures of cigarette and alcohol use for non-Hispanic whites, and past-year alcohol use for South Americans). Puerto Ricans and non-Hispanic blacks are relatively high on several measures of illicit substance use (for Puerto Ricans they are any illicit drug use, marijuana and cocaine use in the past year, and need for illicit drug abuse treatment), (for non-Hispanic blacks they are any illicit drug use, marijuana use, and need for illicit drug abuse treatment).

Four racial/ethnic subgroups are relatively low on all nine measures of substance use, dependence, and need for treatment: Asian/Pacific Islanders, Caribbeans, Central Americans, and Cubans. Central Americans are the lowest of all eleven subgroups on seven of the nine measures, all except alcohol dependence and heavy alcohol use. Cubans are the lowest in alcohol dependence (0.9% versus 3.5% in the total surveyed population). Asian/Pacific Islanders are lowest in heavy alcohol use (0.9% versus 5.1% in the total surveyed population).

4.3 Cigarettes

Table 4.2 presents the estimated percentages of individuals in the eleven racial/ethnic subgroups who reported using cigarettes in the past year by gender and age. In the total surveyed population (first row of Table 4.2), males were nearly 20% more likely than females to have used cigarettes in the past year (34% versus 28%). Overall, past-year cigarette use increased with age to a peak prevalence of about 40% among individuals aged 18 to 25 before declining to about 29% among individuals aged 35 and older.

In Table 4.2 and in subsequent tables of this chapter, there are a number of possible explanations of the age pattern of prevalence estimates, which include differences among individuals born during different periods of history (i.e., birth cohort effects) as well as effects of aging. Strong assumptions are often required to analytically separate the effects of cohort, age, and historical period in analyses of cross-sectional data (Johnson, 1985). This report does not attempt to analytically separate these different effects.

Gender and age patterns of cigarette use within racial/ethnic subgroups are generally similar to the corresponding patterns in the total surveyed population, but there are exceptions. The gender difference in past-year cigarette use is relatively large among Asian/Pacific Islanders (14% of females versus 30% of males), and among most Hispanic groups (e.g., 22% of Mexican females versus 36% of Mexican males). On the other hand, males and females are not significantly different among Caribbeans (21% versus 22%) and other Hispanics (25% versus 27%). Among Caribbeans, Central Americans, Cubans, and non-Hispanic blacks, past-year cigarette smoking attains its peak prevalence relatively late in the life cycle, at ages 26 to 34 rather than at ages 18 to 25.

Despite some distinctive age patterns of substance use within racial/ethnic subgroups, the ranking of subgroups by prevalence of past-year smoking is largely the same within each of the age groups 12 to 17, 18 to 25, 26 to 34, and 35 and older. Native Americans, Puerto Ricans, and non-Hispanic whites have relatively high prevalence at every age, while Asian/Pacific Islanders, Caribbeans, and Central Americans have relatively low prevalence at every age. The main exception is non-Hispanic blacks, who are among the lowest subgroups in smoking at ages 12 to 17 (9.3%) and among the highest at ages 35 and older (34%). Based on the University of Michigan's Monitoring the Future survey, the age pattern of cigarette use among non-Hispanic blacks is better interpreted as a birth cohort effect than as an effect of aging. This is because the prevalence of cigarette smoking among black high school students consistently declined between about 1981 and 1992 (Johnston et al., 1995). 

4.4 Alcohol

Table 4.3 presents the estimated percentages of individuals in the eleven racial/ethnic subgroups who reported using alcohol in the past year, by gender and age. In the total surveyed population (first row of Table 4.3), males were about 15% more likely than females to have used alcohol in the past year (71% versus 62%). Past-year alcohol use increased with age to a peak prevalence of about 80% among individuals aged 18 to 34, before declining to about 64% among individuals aged 35 and older.

In every racial/ethnic subgroup except Native Americans, males were significantly more likely than females to use alcohol. The significant gender differences range from a low of about 8 percentage points among non-Hispanic whites (73% of males versus 65% of females) to a high of about 24 percentage points among Caribbeans, Central Americans, and Mexicans. Age patterns of alcohol use were also similar across subgroups: without exception, prevalence is substantially higher at ages 18 to 25 and 26 to 34 than at ages 12 to 17 and 35 and older. Some subgroups attain their peak prevalence at ages 18 to 25 (Asian/Pacific Islanders, other Hispanics), while others attain their peak prevalence at ages 26 to 34 (Caribbeans, Central Americans, Cubans, Puerto Ricans). However, in every subgroup, prevalence at ages 26 to 34 is no more than 20% higher or lower than prevalence at ages 18 to 25. Thus, the percentage using alcohol in the past year is roughly constant between the end of adolescence and the beginning of middle age. Regardless of age group, South Americans and non-Hispanic whites are relatively high in past-year alcohol use, while Asian/Pacific Islanders, Central Americans, Puerto Ricans, and non-Hispanic blacks are relatively low.

4.5 Any Illicit Drug

Table 4.4 presents the estimated percentages of individuals in the eleven racial/ethnic subgroups who reported using any illicit drug in the past year, by gender and age. In the total surveyed population (first row of Table 4.4), males were about 40% more likely than females to have used any illicit drug in the past year (14% versus 10%). Past-year use of any illicit drug increased with age to a peak prevalence of about 27% among 18- to 25-year-olds before declining to about 6% among individuals aged 35 and older.

In every racial/ethnic subgroup, males were more likely than females to have used illicit drugs in the past year, except among Native Americans, Caribbeans, and other Hispanics, three subgroups with no significant difference between males and females. The significant gender differences range from a low of about 4 percentage points among non-Hispanic whites (14% of males versus 10% of females), to a high of about 9 percentage points among Puerto Ricans and South Americans.

In the total surveyed population, Mexicans, Puerto Ricans, and non-Hispanic blacks were relatively high in past-year illicit drug use (around 13%), while Asian/Pacific Islanders, Caribbeans, Central Americans, and Cubans were relatively low (5.7% to 8.2). Native Americans were relatively high at ages 12 to 17 (23%) and 26 to 34 (34%), but data for Native Americans are too sparse to provide reliable estimates at ages 18 to 25 and 35 and older. Other Hispanics have a fairly distinctive age pattern of past-year illicit drug prevalence: relatively high at ages 25 and younger (17% at ages 12 to 17, versus 13% in the total population), and relatively low at ages 35 and older (3.0% at ages 35 and older, versus 5.9% in the total population).

4.6 Marijuana

Table 4.5 presents the estimated percentages of individuals in the eleven racial/ethnic subgroups who reported using marijuana in the past year, by gender and age. In the total surveyed population (first row of Table 4.5), males were about 70% more likely than females to have used marijuana in the past year (11% versus 6.7%). Past-year use of marijuana increased with age, to a peak prevalence of about 23% among 18- to 25-year-olds, before declining to about 4% among individuals aged 35 and older. Given that marijuana is by far the most commonly used illicit drug (SAMHSA, 1997a), it is not surprising that gender and age patterns for marijuana are similar to the corresponding patterns for any illicit drug (see section 4.5).

In every racial/ethnic subgroup males are more likely than females to have used marijuana in the past year, except among Native Americans and other Hispanics, two subgroups with no significant difference between males and females. In every subgroup males are more than twice as likely as females to be past-year marijuana users, except among Native Americans (16% of females versus 14% of males), other Hispanics (9.2% and 8.9%), non-Hispanic blacks (7.7% of females versus 14% of males), and non-Hispanic whites (6.7% versus 11%). Among Asian/Pacific Islanders (2.0% versus 7.7%) and South Americans (4.2% versus 13%), males are at least three times as likely as females to be past-year marijuana users.

In every age group, Puerto Ricans and non-Hispanic whites are relatively high in past-year marijuana use, while Asian/Pacific Islanders, Caribbeans, Central Americans, and Cubans are relatively low. Native Americans are relatively high at ages 12 through 34, but data for Native Americans are too sparse to yield a reliable estimate at ages 35 and older. Non-Hispanic blacks are relatively high at ages 26 and older (5.8% among individuals aged 35 and older) but about average at younger ages. Mexicans, South Americans, and other Hispanics, are about average in past-year marijuana use.

4.7 Cocaine

Table 4.6 presents the estimated percentages of individuals in the eleven racial/ethnic subgroups who reported using cocaine in the past year, by gender and age. In the total surveyed population (first row of Table 4.6), males were about twice as likely as females to have used cocaine in the past year (3.5% versus 1.7%). Past-year use of cocaine increased with age to a peak prevalence of about 6.3% among 18- to 25-year-olds before declining to about 1.1% among individuals aged 35 and older.

Racial/ethnic differences in past-year cocaine use by gender and age appear similar to the corresponding differences for any illicit drug (section 4.5) and for marijuana (section 4.6). Yet the low prevalence of cocaine use relative to marijuana and any illicit drug use makes it more difficult to detect statistically significant differences (see section 2.4). Males are significantly more likely than females to use cocaine in every subgroup, with the possible exceptions of Native Americans (insufficient data to estimate prevalence among females), South Americans (no significant male-female difference), and other Hispanics (no significant male-female difference).

Overall past-year cocaine use is relatively high among Native Americans (5.2%), Mexicans (3.9%), Puerto Ricans (3.7%), and non-Hispanic blacks (3.1%), but the age pattern of cocaine use appears different in these four subgroups: Among Native Americans and Mexicans, the peak prevalence occurs at ages 18 to 25 (11% and 7.1%, respectively), while among Puerto Ricans and non-Hispanic blacks, the peak prevalence occurs at ages 26 to 34 (9.3% and 6.1%, respectively). At ages 18 to 25, past-year cocaine use is also high among other Hispanics (12%) and among non-Hispanic whites (7%). In every age group, past-year cocaine use is relatively low among Asian/Pacific Islanders, Caribbeans, and Central Americans.

4.8 Need for Illicit Drug Abuse Treatment

Table 4.7 presents the estimated percentages of individuals in the eleven racial/ethnic subgroups who were in need of illicit drug abuse treatment, by gender and age. In the total surveyed population (first row of Table 4.7), males were about 50% more likely than females to need illicit drug abuse treatment (3.5% versus 2.1%). Treatment need increased with age to a peak prevalence of about 6.0% among 18- to 25-year-olds before declining to about 1.4% among individuals aged 35 and older.

The racial/ethnic subgroups that are relatively high in need for illicit drug abuse treatment are the same subgroups that are relatively high in past-year use of any illicit drug (see section 4.5): Native Americans (7.8%), Mexicans (3.6%), Puerto Ricans (3.7%), and non-Hispanic blacks (3.9%). Gender and age patterns of treatment need also appear very similar to the patterns observed for any illicit drug use, but it is more difficult to find statistically significant racial/ethnic differences in Table 4.7, because of the relatively low prevalence of illicit drug abuse treatment need (see section 2.4). Regardless of gender and age, Asian/Pacific Islanders (1.7%), Caribbeans (1.6%), Central Americans (1.5%), and South Americans (1.7%) appear relatively low in need for illicit drug abuse treatment.

4.9 Alcohol Dependence

Table 4.8 presents the estimated percentages of individuals in the eleven racial/ethnic subgroups who reported dependence on alcohol, by gender and age. In the total surveyed population (first row of Table 4.8), males were about twice as likely as females to be dependent on alcohol (4.9% versus 2.1%). Alcohol dependence increased with age to a peak prevalence of about 8.0% among 18- to 25-year-olds before declining to 2.2% among individuals aged 35 and older.

Gender differences and age patterns of alcohol dependence within racial/ethnic subgroups appear generally similar to the corresponding patterns in the total population, though the low prevalence of alcohol dependence implies that there is relatively little statistical power to detect racial/ethnic differences (see section 2.4). The main exception to the overall pattern is that the gender difference among Mexicans (8.4% - 2.6% = 5.8 percentage points) is significantly larger than the gender difference in any other racial/ethnic subgroup, except Central Americans (5.4% - 0.8% = 4.6 percentage points).

Racial/ethnic differences in alcohol dependence are much more similar to racial/ethnic differences in any illicit drug use (Table 4.4) than to racial/ethnic differences in past-year alcohol use (Table 4.3). The relatively high subgroups are Native Americans (5.6%) and Mexicans (5.6%), while the relatively low subgroups are Asian/Pacific Islanders (1.8%), Caribbeans (1.9%), and Cubans (0.9%).

4.10 Heavy Smoking

Table 4.9 presents the estimated percentages of individuals in the eleven racial/ethnic subgroups who reported smoking a pack or more of cigarettes per day during the past month, by gender and age. In the total surveyed population (first row of Table 4.9), males were about one-third more likely than females to smoke heavily (16% versus 12%). Heavy smoking increased with age to a peak prevalence of about 17% among 26- to 34-year-olds and remained high among individuals aged 35 and older (15%).

In some racial/ethnic subgroups, the percentage smoking heavily appears higher among females than among males, or higher at ages 35 and older than at ages 26 to 34, but these anomalies are generally not statistically significant. Overall, gender differences and age patterns of heavy smoking within racial/ethnic subgroups appear similar to the corresponding patterns in the total population. An exception occurs among South Americans, where the percentage smoking heavily is significantly higher at ages 35 and older (12%) than at ages 26 to 34 (3.3%), which might reflect either a cohort effect or an aging effect among South Americans.

Native Americans (24%), Puerto Ricans (12%), and non-Hispanic whites (16%) are relatively high in heavy smoking, while Asian/Pacific Islanders (4.8%), Caribbeans (3.6%), Central Americans (2.3%), and Mexicans (4.7%) are relatively low. The ranking of subgroups with respect to heavy smoking prevalence is similar to the ranking with respect to past-year cigarette use (Table 4.1).

4.11 Heavy Alcohol Use

Table 4.10 presents the estimated percentages of individuals in the eleven racial/ethnic subgroups who reported using alcohol heavily during the past month (five or more drinks per occasion on five or more occasions during the past 30 days), by gender and age. In the total surveyed population (first row of Table 4.9), males were 4.3 times more likely than females to use alcohol heavily (8.6% versus 2.0%). Heavy alcohol use increased with age to a peak prevalence of about 11% among 18- to 25-year-olds before declining to 7.1% among 26- to 34-year-olds and 3.8% among individuals aged 35 and older.

Gender differences and age patterns of heavy alcohol use are generally similar across racial/ethnic subgroups: In each subgroup, males are much more likely than females to drink heavily, while young adults aged 18 to 34 are more likely to drink heavily than either adolescents aged 12 to 17 or older adults aged 35 and older. As in the results for alcohol dependence (Table 4.8), the gender difference among Mexicans is significantly larger than the gender difference in any other racial subgroup. Mexican males are about seven times more likely than Mexican females to use alcohol heavily (12% versus 1.7%), and about one-third more likely than the males of any other racial/ethnic subgroup to use alcohol heavily (12% versus 8.8% among non-Hispanic white males, the second highest). Yet Mexican females are no more likely to use alcohol heavily than females in the total surveyed population (1.7% versus 2.0%). Among the Hispanic subgroups, Central Americans, Cubans, and South Americans also have estimated ratios of male-to-female prevalence that are higher than the estimate of 4.3 for the total surveyed population. Even so, both the gender difference, measured in percentage points, and the overall prevalence of heavy alcohol use are significantly higher among Mexicans than in other racial/ethnic subgroups.

Mexicans (6.9%) and non-Hispanic whites (5.3%) are relatively high in heavy alcohol use, while Asian/Pacific Islanders (0.9%), Caribbeans (2.5%), Central Americans (2.2%), Cubans (2.8%), and South Americans (3.0%) are relatively low. Non-Hispanic whites are not significantly higher than Native Americans (4.6%), other Hispanics (4.9%), and non-Hispanic blacks (4.7%). The ranking of subgroups with respect to heavy alcohol use is similar to the ranking with respect to alcohol dependence (Table 4.8).

     Return to Table of Contents (TOC)

This page was last updated on May 19, 2008.