Skip To Content

Return to Table of Contents (TOC) TOC  

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 5 Prevalence of Substance Use by Race/Ethnicity, Age, and Other Sociodemographic Characteristics

5.1 Introduction

This chapter extends the analyses of Chapter 4 by examining the effects of other sociodemographic characteristics on racial/ethnic patterns of substance use, alcohol dependence, and need for illicit drug abuse treatment. While Chapter 4 focused strictly on gender and age patterns of substance use and other drug-related outcomes within racial/ethnic subgroups, this chapter examines how substance use within each racial/ethnic subgroup varies according to region, population density, language of interview, family income, health insurance coverage, and household member’s receipt of welfare. Additional control variables in our analysis are meaningfully defined only for adolescents or only for adults: For adolescent respondents aged 12 to 17, we also examine the effects of school dropout status and family structure (whether or not the adolescent lives with two biological parents). For adult respondents aged 18 and older, we also examine the effects of educational attainment, marital status, employment status, and number of own children. (See Chapter 2 for the operational definition of each of the socioeconomic control variables.)

Two principal questions motivated the analyses of this chapter:

·Are socioeconomic and other sociodemographic differences in the prevalence of substance use and other drug-related outcomes similar across all racial/ethnic subgroups? Alternatively, such differences may be smaller within some subgroups than within others. For example, some subgroups may be especially effective in discouraging substance use by subgroup members whose socioeconomic characteristics would otherwise be associated with a high prevalence of substance use.

·To what extent do the sociodemographic differences among the racial/ethnic subgroups account for the subgroups' different patterns of substance use, alcohol dependence, and need for illicit drug abuse treatment? For example, the higher prevalence of heavy alcohol use among Mexicans compared to non-Hispanic whites (Chapter 4) might result from the large differences between Mexicans and non-Hispanic whites in family income, educational attainment, receipt of welfare, and other socioeconomic variables (see Chapter 3). If so, we would expect Mexicans and non-Hispanic whites with similar socioeconomic characteristics to have roughly the same prevalence of heavy alcohol use.

The more refined analyses of this chapter made it necessary to use a less detailed classification of race/ethnicity, a seven-category classification in place of Chapter 4's eleven-category classification. Two major changes reduced the number of categories from eleven to seven: First, this chapter does not present estimates for Native Americans because of insufficient data. Even in the less detailed analyses of Chapter4, we found that the combined 1991-93 NHSDA sample of Native Americans (n = 416) was often too small to yield reliable prevalence estimates. Second, this chapter does not present separate estimates for three small subgroups of Hispanics—Caribbeans, Central Americans, and South Americans—because separate estimates for these subgroups were generally too imprecise to be reported. Rather than reporting separate estimates, we combine these three small Hispanic subgroups with the "Hispanic-other" subgroup of Chapter 4 to form an expanded "Hispanic-other" subgroup. Thus, the seven racial/ethnic subgroups analyzed in this chapter are Asian/Pacific Islanders, Cubans, Mexicans, Puerto Ricans, Hispanic others (now including Caribbeans, Central Americans, and South Americans, as well as individuals who were classified as other Hispanics in Chapter 4), non-Hispanic blacks, and non-Hispanic whites. Except for the omission of Native Americans and the newly-constituted Hispanic-other subgroup, the subgroups in this chapter are the same ones that were analyzed in Chapter 4.

The main conclusions of this chapter are as follows:

·Regional differences. In the total surveyed population, the West is about 40% higher than the three other regions in measures of illicit drug use and need for illicit drug abuse treatment, while the South is about 15% lower than the other regions in past-year alcohol use (60% using in the past year versus about 70%). Regional differences in cigarette use and heavy alcohol use are small in magnitude. The data provide little evidence that regional differences within racial/ethnic subgroups are markedly different from regional differences in the total surveyed population. For example, among individuals aged 12 and older residing in the West, the percentages using any illicit drug in the past year equal about 7.5% among Asian/Pacific Islanders, 14% among Mexicans, 23% among Puerto Ricans, 9.7% among other Hispanics, 19% among non-Hispanic blacks, and 17% among non-Hispanic whites. Among individuals aged 12 and older residing in the North Central, the corresponding percentages equal 3.9%, 11%, 10%, 6.6%, 16%, and 10% (Table 5.30).

·Differences by population density. Relative to the total surveyed population, individuals residing in metropolitan areas larger than 1 million in population have moderately higher prevalences of past-year alcohol use, illicit drug use, and need for illicit drug abuse treatment. They have a moderately lower prevalence of past-year and heavy cigarette use, and heavy alcohol use; and about the same prevalence of alcohol dependence. However, differences by population density appear fairly distinctive within most Hispanic subgroups. For example, within the Mexican subgroup, the estimated prevalences of past-year illicit drug use, need for illicit drug abuse treatment, and alcohol dependence are higher outside metropolitan areas than within metropolitan areas.

·Differences by language of interview. Within every Hispanic subgroup, the percentages reporting past-year use of licit and illicit substances, need for illicit drug abuse treatment, andheavy cigarette use are higher among individuals who responded in English than among individuals who responded in Spanish. For illicit drug use, need for illicit drug abuse treatment, and heavy cigarette use, this difference is generally twice as high. Differences by language of interview in the percentages who were alcohol-dependent and who used alcohol heavily are not statistically significant.

·Differences by family income. In the total surveyed population, the prevalence of past-year alcohol use is positively (directly) associated with family income, while the prevalences of past-year cigarette use, past-year illicit drug use, need for illicit drug abuse treatment, alcohol dependence, heavy cigarette use, and heavy alcohol use are negatively (inversely) associated with family income. Within Hispanic subgroups, however, these outcomes tend to be less strongly associated with family income than in the total surveyed population. For example, in the total surveyed population, the percentage who were dependent on alcohol declined from about 3.9% among individuals with family incomes of less than $20,000 to about 3.2% among individuals with family incomes of more than $40,000, but, within the Mexican subgroup, about 5.7% and 5.6% respectively of individuals were alcohol-dependent at each income level (Table 5.70).

·Differences by health insurance coverage and receipt of welfare. Both in the total surveyed population and within racial/ethnic subgroups, individuals who are not covered by health insurance or who reside in households where a family member is receiving welfare have relatively high prevalence of most substance abuse and drug-related outcomes. As in the case of family income, these associations appear less strong within Hispanic subgroups than in the total surveyed population.

·Differences between school dropouts and adolescents still in school. Both in the total surveyed population and within racial/ethnic subgroups, individuals aged 12 to 17 who have dropped out of school have a high prevalence of most substance use and drug-related outcomes relative to same-aged individuals still in school.

·Differences by family structure. Both in the total surveyed population and within most racial/ethnic subgroups, individuals aged 12 to 17 who are living with zero or one biological parent have a high prevalence of substance use and drug-related outcomes relative to same-aged individuals who are living with two biological parents. Only within the Cuban subgroup do the data suggest no large differences by family structure.

·Differences by educational attainment. In the total surveyed population aged 18 and older, the prevalences of past-year cigarette use, past-year illicit drug use, need for illicit drug abusetreatment, heavy cigarette use, and heavy alcohol use increase with educational attainment to a peak among individuals with 9 to 11 years of schooling. These prevalences then decline with further increases in educational attainment, while the prevalence of past-year alcohol use increases regularly with educational attainment to a peak among individuals with more than 12 years of schooling (Table 5.22). These patterns appear generally consistent across racial/ethnic subgroups.

·Differences by marital status. In the total surveyed population aged 18 and older, the following behaviors are more prevalent among never-married individuals than among widowed/ divorced/separated individuals: past-year cigarette use, past-year alcohol use, past-year illicit drug use, need for illicit drug abuse treatment, alcohol dependence, and heavy alcohol use. Furthermore, except for past-year alcohol use, the prevalences of these behaviors are higher among widowed/divorced/ separated individuals than among currently married individuals. For example, about 26% of never-married individuals used any illicit drug in the past year, compared with 11% of widowed/divorced/ separated individuals and 7.2% of married individuals. The prevalence of heavy cigarette use is higher among widowed/divorced/separated individuals (20%) than among either never married (14%) or currently married (14%) individuals (Table 5.82). These patterns appear generally consistent across racial/ethnic subgroups.

·Differences by employment status. In the total surveyed population aged 18 and older, the prevalences of past-year cigarette use, past-year illicit drug use, need for illicit drug abuse treatment, alcohol dependence, heavy cigarette use, and heavy alcohol use are highest among unemployed individuals, second-highest among employed individuals, and lowest among individuals who are not in the labor force. The prevalence of past-year alcohol use is roughly equal between employed and unemployed individuals and about 25% lower among individuals who are not in the labor force (Table 5.22). These patterns appear generally consistent across racial/ethnic subgroups.

·Differences by number of own children. In the total surveyed population aged 18 and older, the prevalences of past-year illicit drug use, need for illicit drug abuse treatment, alcohol dependence, and heavy alcohol use among individuals with no children of their own are about 50% higher than among those with one or more children. In contrast, the prevalences of past-year and heavy cigarette use and past-year alcohol use do not differ significantly by number of own children. These patterns appear invariant across racial/ethnic subgroups. For example, among individuals with no children, the percentages of Asian/Pacific Islanders, Cubans, Mexicans, Puerto Ricans, Hispanic others, non-Hispanic blacks, and non-Hispanic whites who used any illicit drug in the past year are about 10%, 10%, 17%, 17%, 11%, 15%, and 13%, respectively, while, among individuals with one or more children, the corresponding percentages are about 1.7%, 5.8%, 8.7%, 9.6%, 5.8%, 12%, and 10%, respectively (Table 5.32).

·The role of socioeconomic differences in accounting for racial/ethnic patterns of substance use. Without exception, none of the socioeconomic variables analyzed in this chapter fully accounts for racial/ethnic differences in past-year substance use, alcohol dependence, need for illicit drug abuse treatment, and heavy cigarette and alcohol use. When racial/ethnic subgroups are compared with respect to the prevalences of each of the nine drug-related outcome variables of this chapter, the patterns of racial/ethnic differences within particular socioeconomic subclasses, such as family income levels and levels of educational attainment, are generally similar to the patterns observed in the total surveyed population for that group. For example, when the comparison of racial/ethnic subgroups is restricted to individuals with family incomes greater than $40,000 or to individuals with more than 12 years of schooling, Mexicans are still the highest of the seven subgroups in past-year illicit drug use and Asian/Pacific Islanders are still the lowest, just as in the total surveyed population (Table 5.32). In future research, multiple regression and related techniques might be applied to quantitatively assess how much of racial/ethnic variations in substance use can be attributed to socioeconomic differences among the subgroups.

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

5.2 Cigarettes

Table 5.10 presents the estimated percentages of individuals aged 12 and older in the total surveyed population (which excludes Native Americans in this chapter [ Because Native Americans made up less than 1% of the total surveyed population (Table 2.2), their exclusion from the analyses of this chapter has little effect on estimates for the total surveyed population. This chapter's conclusions about patterns of substance use and other outcomes in the total surveyed population would be unaltered if Native Americans were included in the analyses.] ) and within each of the seven racial/ethnic subgroups who reported using cigarettes in the past year, by region, density, language of interview, family income, health insurance coverage, and receipt of welfare. (Also shown are percentages by gender and age, which are similar to estimates reported in Table 4.1, except that Table 5.10 uses the less refined seven-category classification of race/ethnicity.) Table 5.11 presents similar estimates for individuals aged 12 to 17, including estimates by school dropout status and family structure. Table 5.12 presents similar estimates for individuals aged 18 and older, including estimates by educational attainment, marital status, employment status, and number of own (i.e., biological) children.

For the total surveyed population, the first column of Table 5.10 shows that past-year cigarette use is:

· slightly higher in the South (32%) than in the Northeast (30%) and West (30%);

·higher among individuals who are not in metropolitan statistical areas (MSAs) than among those

who are in MSAs with populations greater than 1 million (34% versus 29%);

·higher among those who used English in the interview than among those who used Spanish (31% versus 26%);

·higher among individuals with family incomes of $40,000 or less than among individuals with

family incomes of greater than $40,000 (33% to 34% versus 28%);

·much higher among individuals who do not have health insurance coverage than among those who do (44% versus 29%); and

·much higher among individuals who have a family member in the household receiving welfare than among those who do not (46% versus 30%).

Many of these socioeconomic differences for the total surveyed population are discussed in detail in previous NHSDA reports (e.g., SAMHSA, 1997a). Similar patterns are observed in the subpopulations aged 12 to 17 (first column of Table 5.11) and 18 and over (first column of Table 5.12), except that, in the subpopulation aged 12 to 17, the North Central (21%) is significantly higher in past-year cigarette use than the South (18%) and West (19%), and there are no significant differences by family income, health insurance coverage, or welfare receipt. The first column of Table 5.11 also shows that, in the total surveyed population aged 12 to 17, past-year smoking is higher among individuals who are not living with two biological parents than among those who are (24% versus 16%), and higher among school dropouts than among those who are still in school (45% versus 18%). [ In Table 5.11 and subsequent tables of this chapter, differences by school dropout status are subject to varying interpretations because school dropouts aged 12 to 17 are generally older than non- dropouts aged 12 to 17, and because age is positively associated with cigarette and other substance use during adolescence. Higher substance use among school dropouts might be due to the higher average age of school dropouts .] The first column of Table 5.12 also shows that, in the population aged 18 and older, past-year smoking is:

·higher among individuals with 9 to 11 years of schooling than among individuals with more or

fewer years of schooling (45% versus 35% or less);

·higher among widowed/divorced/separated (38%) and among never-married individuals (39%)

than among those currently married (28%);

·higher among unemployed individuals (50%) than among individuals who are employed (34%) or not in the labor force (25%); and

·slightly higher among individuals with one or more of their own children than among individuals with no children of their own (33% versus 31%).

Within racial/ethnic subgroups (second through eighth columns of Tables 5.10, 5.11, and 5.12), most socioeconomic differences in past-year cigarette use are similar to those observed in the total surveyed population. [ As shown by the asterisks in Tables 5.10- 5.12, this broad conclusion is weakened by our inability to analyze some within-subgroup socioeconomic differences in smaller racial/ethnic subgroups, because of low precision estimates for one or more categories of a control variable. Similar qualifications apply to subsequent tables of this chapter.] Regardless of racial/ethnic subgroup, among individuals aged 18 and older, the prevalence of past-year cigarette use was relatively high for those who resided outside of metropolitan areas with populations of more than 1 million, responded to the NHSDA interview in English rather than in Spanish, were not covered by health insurance, had a family member in the household who received welfare, or had 9 to 11 years of formal schooling (Table 5.12).

Yet there are also some important racial/ethnic differences in the effects of socioeconomic variables on past-year cigarette use:

·Racial/ethnic differences by region and population density. Differences by region and population density are relatively small, although statistically significant, in the total surveyed population; however, they are generally not significant within racial/ethnic subgroups. For example, within the Mexican, Hispanic other, and non-Hispanic black subgroups, past-year cigarette use is approximately equal between those residing in metropolitan areas with populations of greater than 1 million and those residing outside metropolitan areas.

·Racial/ethnic differences by family income. Within the Cuban, Puerto Rican, and other Hispanic subgroups, the percentage who used cigarettes in the past year is not significantly different between individuals with family incomes less than $20,000 and individuals with family incomes more than $40,000. Within the Asian-American and non-Hispanic black subgroups, individuals with family incomes lower than $20,000 are more likely to smoke than individuals with family incomes of $20,000 or higher, but there is no significant difference between the two highest income categories, $20,000-$40,000 and greater than $40,000.

·Racial/ethnic differences by receipt of welfare and health insurance. Differences between individuals in households that received and did not receive welfare are generally smaller withinHispanic subgroups (except Cubans) than in the total surveyed population. For example, within the Mexican subgroup, 33% of those in households that received welfare were past-year cigarette users, compared with 29% of those in households that did not receive welfare, while, in the total surveyed population, the corresponding percentages are 46% and 30%. Similarly, differences between individuals covered by health insurance and those not covered are smaller within Hispanic subgroups (except Cubans) than in the total surveyed population.

None of the socioeconomic variables analyzed in this chapter fully accounts for racial/ethnic differences in past-year cigarette use. For example, in comparing adolescents aged 12 to 17 from the seven subgroups analyzed in this chapter, non-Hispanic whites have the highest overall percentage of past-year cigarette users (22% versus 19% in the total surveyed population aged 12 to 17); and non-Hispanic white adolescents are also generally the highest when racial/ethnic subgroups are compared within particular regions, population densities, levels of family income, and other socioeconomic subclasses (see Table 5.11). Similarly, regardless of socioeconomic category, Asian/Pacific Islanders are consistently among the lowest in past-year cigarette use of the seven subgroups, both in the subpopulation aged 12 to 17 (Table 5.11) and in the subpopulation aged 18 and older (Table 5.12). 

5.3 Alcohol

Table 5.20 presents the estimated percentages of individuals aged 12 and older in the total surveyed population (excluding Native Americans) and within each of the seven racial/ethnic subgroups who reported using alcohol in the past year by region, density, language of interview, family income, health insurance coverage, and receipt of welfare. Table 5.21 presents similar estimates for individuals aged 12 to 17, including estimates by school dropout status and family structure. Table 5.22 presents similar estimates for individuals aged 18 and older, including estimates by educational attainment, marital status, employment status, and number of own (i.e., biological) children.

For the total surveyed population, the first column of Table 5.20 shows that past-year alcohol use is higher:

· in the Northeast, North Central, and West (each about 70%) than in the South (59%);

·among individuals who reside in metropolitan statistical areas (MSAs) with populations larger than 1 million compared to those who live outside MSAs (70% versus 60%);

·among those who used English in the interview compared to those who used Spanish (67% versus 57%);

·among individuals with family incomes of $40,000 or more compared to individuals with family

incomes less than $20,000 (73% versus 58%); and

·among individuals in households where no family member received welfare compared to

individuals in households that did have a family member receiving welfare (67% versus 61%).

Except for the positive associations between alcohol use and male gender, alcohol use and young adulthood, and alcohol use and English language use, the associations between past-year alcohol use and the socioeconomic variables presented in Table 5.20 are in the opposite directions from the associations between past-year cigarette use and the same variables (previous section). Moreover, while past-year cigarette use is negatively associated with health insurance coverage, past-year alcohol use has no significant association with health insurance coverage, either among adolescents aged 12 to 17 (Table 5.21) or among adults aged 18 and older (Table 5.22).

The first column of Table 5.21 also shows that, in the total surveyed population aged 12 to 17, past-year alcohol use, like past-year cigarette use, is higher among individuals who are not living with two biological parents than among those who are (42% versus 32%), as well as being higher among school dropouts than those still in school (54% versus 36%). The first column of Table 5.22 also shows that, in the population aged 18 and older, past-year alcohol use increases regularly with educational attainment, from about 41% among individuals with 8 or fewer years of schooling to about 79% among individuals with more than 12 years of schooling. In this respect it is unlike past-year cigarette use, which is highest among individuals with 9 to 11 years of schooling (Table 5.12). Past-year alcohol use is higher among individuals who have never been married (79%) than among widowed, divorced, or separated individuals (60%) or married individuals (70%), and higher among employed individuals (77%) than among individuals who are unemployed (73%) or not in the labor force (55%).

Within racial/ethnic subgroups (second through eighth columns of Tables 5.20, 5.21, and 5.22), most socioeconomic differences in past-year alcohol use are similar to those observed in the total surveyed population. For example, regardless of racial/ethnic subgroup, past-year alcohol use increases regularly with educational attainment (Table 5.22).

Yet there are also some important racial/ethnic differences in the effects of socioeconomic variables on past-year alcohol use:

·Racial/ethnic differences by region and population density. The South's relatively low percentage using alcohol in the past year holds true within the Cuban, Mexican, non-Hispanic black, and non-Hispanic white subgroups but not within other racial/ethnic subgroups. Also, within each of the four Hispanic subgroups, individuals who reside in metropolitan areas with populations greater than 1 million are not significantly higher in past-year alcohol use than individuals who reside outside metropolitan areas.

·Racial/ethnic differences by family income. The overall positive association between past-year alcohol use and family income does not hold for Asian/Pacific Islanders. Within the Asian/PacificIslander subgroup aged 12 and older, about 55% of those with family incomes greater than $40,000 used alcohol in the past year, as compared with 56% of those with family incomes less than $20,000 (Table 5.20).

·Racial/ethnic differences by receipt of welfare. Within the non-Hispanic black and non-Hispanic white subgroups, there are no significant differences in past-year alcohol use by whether or not the individual resided in a household that received welfare.

·Racial/ethnic differences by family structure. For individuals aged 12 to 17 in all Hispanic subgroups except Cubans, those not living with two biological parents are significantly higher in past-year alcohol use than individuals living with two biological parents (Table 5.21). Among Cubans, there is no significant difference between the groups.

None of the socioeconomic variables analyzed in this chapter fully accounts for racial/ethnic differences in past-year alcohol use. The prevalence of past-year alcohol use is generally highest among non-Hispanic whites and lowest among Asian/Pacific Islanders, both in the total surveyed population and within subclasses defined by socioeconomic variables.

5.4 Any Illicit Drug

Table 5.30 presents the estimated percentages of individuals aged 12 and older in the total surveyed population (excluding Native Americans) and within each of the seven racial/ethnic subgroups who reported using any illicit drug in the past year, by region, density, language of interview, family income, health insurance coverage, and household member’s receipt of welfare. Table 5.31 presents similar estimates for individuals aged 12 to 17, including estimates by school dropout status and family structure. Table 5.32 presents similar estimates for individuals aged 18 and older, including estimates by educational attainment, marital status, employment status, and number of own (i.e., biological) children.

For the total surveyed population, the first column of Table 5.30 shows that past-year use of any illicit drug is higher in the West (16%) than in the Northeast (11%), North Central (10%), and South (11%); higher among individuals in metropolitan statistical areas (MSAs) with populations greater than 1 million than among those outside MSAs (13% versus 11%); higher among those who used English in the interview than among those who used Spanish (12% versus 5.1%); and higher among individuals with family incomes less than $20,000 than among individuals with family incomes of more than $40,000 (14% versus 10%). It is also much higher among individuals who do not have health insurance coverage than among those who do (20% versus 11%); and much higher among individuals with a member of the household receiving welfare than without (22% versus 11%). Except for the relatively high percentage using any illicit drug in the West, and the positive association between any illicit drug use and population density, socioeconomic differences in past-year use of any illicit drug are similar to socioeconomic differences in past-year cigarette use.

Similar patterns are observed in the subpopulations aged 12 to 17 (first column of Table 5.31), and 18 and over (first column of Table 5.32), except that in the 12 to 17 group there are no significant differences by population density, family income, or health insurance coverage. The first column of Table 5.31 also shows that, in the total surveyed population aged 12 to 17 , past-year use of any illicit drug is higher among individuals who are not living with two biological parents than among those who are (18% versus 11%), and the difference between school dropouts and those still in school (31% versus 27%) is statistically significant. The first column of Table 5.32 also shows that, in the population aged 18 and older, past-year use of any illicit drug is higher among individuals with 9 to 11 years of schooling than among individuals with fewer or more years of schooling (15% versus 12% or less); higher among never-married individuals (26%) than among widowed/divorced/separated (11%) or married individuals (7.2%); higher among unemployed individuals (24%) than among individuals who are employed (13%) or not in the labor force (6.6%); and higher among individuals with no children of their own than among individuals with one or more (13% versus 9.4%). The differences by education, marital status, and employment status are similar to corresponding differences in past-year cigarette use (Table 5.12).

Within racial/ethnic subgroups (second through eighth columns of Tables 5.30, 5.31, and 5.32), most socioeconomic differences in past-year use of any illicit drug use are similar to those observed in the total surveyed population. For example, in every racial/ethnic subgroup, individuals with more than 8 years of education, never-married individuals, unemployed individuals, individuals with no children of their own, and individuals who spoke English at the interview have relatively high risks of using illicit drugs. Within racial/ethnic subgroups as in the total surveyed population, differences by population density and by family income appear to be modest.

Yet there are some important racial/ethnic differences in the effects of socioeconomic variables on past-year use of any illicit drug:

·Racial/ethnic differences by region. The West is highest within most racial/ethnic subgroups, including Mexicans, Puerto Ricans, Hispanic others, non-Hispanic blacks, and non-Hispanic whites. However, among Asian/Pacific Islanders, the West (7.5%) is significantly higher than the South (3.4%) but not significantly different from the Northeast (8.7%) or North Central (3.9%).

·Racial/ethnic differences by family structure. Differences by family structure in adolescents’ past-year use of any illicit drug are smaller within Hispanic subgroups (except Mexicans) than in the total surveyed population (Table 5.31). For example, within the Cuban subgroup, the percentage of past-year illicit drug users equals 12% among adolescents in mother/father families, compared with 11% among those in other family types.

·Racial/ethnic differences by receipt of welfare and health insurance coverage. Differences between those receiving and not receiving welfare and between those covered and not covered by health insurance appear smaller among Asian/Pacific Islanders and within most Hispanic subgroups (except Cubans) than in the total surveyed population.

None of the socioeconomic variables analyzed in this chapter fully accounts for racial/ethnic differences in past-year use of any illicit drug. For example, Asian/Pacific Islanders, Cubans, and other Hispanics are consistently relatively low in past-year illicit drug use, not only when racial/ethnic subgroups as a whole are compared, but also when those subgroups are compared within particular regions, population densities, income levels and other socioeconomic variables addressed in this chapter. It is important to keep in mind that the relatively low levels of this measure for these subgroups indicate only their comparison to other racial/ethnic subgroups, and do not suggest that the amount of illicit drug use is not worthy of attention. This is particularly true for a category as heterogeneous as that of Asian/Pacific Islander.

5.5 Marijuana

Table 5.40 presents the estimated percentages of individuals aged 12 and older in the total surveyed population (excluding Native Americans), and within each of the seven racial/ethnic subgroups, who reported using marijuana in the past year, by region, density, language of interview, family income, health insurance coverage, and household members’ receipt of welfare. Table 5.41 presents similar estimates for individuals aged 12 to 17, including estimates by school dropout status and family structure. Table 5.42 presents similar estimates for individuals aged 18 and older, including estimates by educational attainment, marital status, employment status, and number of own (i.e., biological) children.

For the total surveyed population, the first column of Table 5.40 shows that socioeconomic differences in past-year marijuana use are similar to corresponding differences in the past-year use of any illicit drug (Table 5.30). Past-year use of marijuana is higher:

·in the West (12%) than in the Northeast (8.6%), North Central (8.0%), and South (8.2%);

·among individuals in metropolitan statistical areas (MSAs) with populations greater than 1 million than among those outside MSAs (10% versus 7.7%);

·among those who used English in the interview than among those who used Spanish (9.1% versus 2.6%); and  among individuals with family incomes less than $20,000 than among individuals with family incomes of more than $40,000 (11% versus 7.6%).

It is also much higher among individuals who do not have health insurance coverage than among those who do (16% versus 7.8%); and much higher among individuals with a member of the household receiving welfare than among those without (17% versus 8.5%).

Similar patterns are observed in the subpopulations aged 12 to 17 (first column of Table 5.41) and 18 and over (first column of Table 5.42), except that in the subpopulation aged 12 to 17, there are no significant differences by population density, family income, or health insurance coverage. The first column of Table 5.41 also shows that, in the total surveyed population aged 12 to 17, past-year use of marijuana is higher among individuals who are not living with two biological parents than among those who are (13% versus 7.0%), and higher among school dropouts than among those still in school (28% versus 8.9%). The first column of Table 5.42 also shows that, in the population aged 18 and older, past-year use of marijuana is:

·higher among individuals with 9 to 11 years of schooling than among individuals with fewer or

more years of schooling (12% versus 9% or less);

·much higher among never-married individuals (22%) than among widowed/divorced/separated

(7.6%) or married individuals (4.9%);

·higher among unemployed individuals (20%) than among individuals who are employed (10%) or not in the labor force (4.5%); and

·higher among individuals living with no children of their own than among individuals living with one or more (11% versus 6.6%).

Understandably, since marijuana is the most commonly used illicit drug, socioeconomic differences in past-year marijuana use are similar to socioeconomic differences in past-year use of any illicit drug (Table 5.30). Within racial/ethnic subgroups (second through eighth columns of Tables 5.40, 5.41, and 5.42), most socioeconomic differences in past-year marijuana use are similar to those observed in the total surveyed population. For example, in every racial/ethnic subgroup, the prevalence of marijuana use is relatively high among individuals with more than 8 years of education, never-married individuals, unemployed individuals, individuals with no children of their own, and individuals who spoke English at the interview. Within racial/ethnic subgroups as in the total surveyed population, differences by population density and by family income appear to be modest.

Yet there are some important racial/ethnic differences in the effects of socioeconomic variables on past-year use of marijuana:

·Racial/ethnic differences by region. The West is highest within most racial/ethnic subgroups, including Asian/Pacific Islanders, Mexicans, Puerto Ricans, non-Hispanic blacks, and non-Hispanic whites. However, among other Hispanics, there are no significant regional differences.

·Racial/ethnic differences by family structure. Differences by family structure in adolescents' past-year marijuana use (Table 5.41) are less pronounced within the Puerto Rican, other Hispanic, and non-Hispanic black subgroups than in the total surveyed population. For example, within the Puerto Rican subgroup, the estimated percentage using marijuana in the past year equals 13% both among adolescents in mother/father families and among adolescents in all other family types.

·Racial/ethnic differences by household member’s receipt of welfare and health insurance coverage. As in the analysis of past-year use of any illicit drug (Tables 5.30- 5.32), differences between those receiving and not receiving welfare and between those covered and not covered by health insurance appear smaller within the Asian/Pacific Islander and within most Hispanic subgroups (except Cubans) than in the total surveyed population.

None of the socioeconomic variables analyzed in this chapter fully accounts for racial/ethnic differences in past-year use of marijuana. For example, the same subgroups that are relatively low in past-year marijuana use when compared as a whole in Chapter 4 (Asian/Pacific Islanders, Cubans, and other Hispanics) are also relatively low when compared within subclasses of the socioeconomic variables examined in this chapter.

5.6 Cocaine

Table 5.50 presents the estimated percentages of individuals aged 12 and older in the total surveyed population (excluding Native Americans) and within each of the seven racial/ethnic subgroups who reported using cocaine in the past year, by region, density, language of interview, family income, health insurance coverage, and receipt of welfare. Table 5.51 presents similar estimates for individuals aged 12 to 17, including estimates by school dropout status and family structure. Table 5.52 presents similar estimates for individuals aged 18 and older, including estimates by educational attainment, marital status, employment status, and number of own (i.e., biological) children.

For the total surveyed population, the first column of Table 5.50 shows that socioeconomic differences in past-year cocaine use are similar to corresponding differences in the past-year use of any illicit drug (Table 5.30), and in the past-year use of marijuana (Table 5.40). Past-year use of cocaine is higher in the West (3.9%) than in the Northeast (2.5%), North Central (2.1%), and South (2.0%); higher among individuals in metropolitan statistical areas (MSAs) with populations greater than 1 million than among those outside MSAs (2.9% versus 2.0%); higher among those who used English in the interview than among those who used Spanish (2.5% versus 1.7%), and higher among individuals with family incomes less than $20,000 than among individuals with family incomes of more than $40,000 (3.2% versus 2.0%).It is also higher among individuals who do not have health insurance coverage than among those who do (5.6% versus 2.0%); and higher among individuals who have a member of the household receiving welfare than among those who do not (5.7% versus 2.3%).

Similar patterns are observed in the subpopulations aged 12 to 17 (first column of Table 5.51) and 18 and over (first column of Table 5.52), except that, in the subpopulation aged 12 to 17, there are no significant differences by population density, family income, or receipt of welfare. The first column of Table 5.51 also shows that, in the total surveyed population aged 12 to 17, past-year use of cocaine is higher among individuals who are not living with two biological parents than among those who are (1.8% versus 0.7%), and higher among school dropouts than among those still in school (8.7% versus 0.9%). The first column of Table 5.52 also shows that, in the population aged 18 and older, past-year use of cocaine is higher among individuals with 9 to 11 years of schooling than among individuals with fewer or more years of schooling (4.1% versus 2.9% or less); much higher among never-married individuals (6.6%) than among widowed/divorced/separated (2.5%) or married individuals (1.4%); higher among unemployed individuals (8.4%) than among individuals who are employed (2.9%) or not in the labor force (1.0%); and higher among individuals with no children of their own than among individuals with one or more (3.2% versus 2.0%).

Within racial/ethnic subgroups (second through eighth columns of Tables 5.50, 5.51, and 5.52), most socioeconomic differences in past-year cocaine use appear similar to those observed in the total surveyed population. For example, in every racial/ethnic subgroup, individuals with more than 8 years of education, never-married individuals, unemployed individuals, individuals with no children of their own, and individuals who spoke English at the interview have relatively high prevalences of cocaine use. Within racial/ethnic subgroups as in the total surveyed population, differences by population density and by family income appear to be modest.

Yet there are some important racial/ethnic differences in the effects of socioeconomic variables on past-year use of cocaine:

·Racial/ethnic differences by region. The West is highest within most racial/ethnic subgroups, including Asian/Pacific Islanders, Mexicans, Puerto Ricans, non-Hispanic blacks, and non-Hispanic whites. However, among Cubans and other Hispanics, there are no significant regional differences. For Cubans in particular, this may be a result of their extreme concentration in the South.

·Racial/ethnic differences by family structure. Differences in adolescents' past-year cocaine use by family structure (Table 5.51) are smaller within the Cuban, Puerto Rican, and other Hispanic subgroups than in the total surveyed population.

·Racial/ethnic differences by receipt of welfare and health insurance coverage. As in the analyses of past-year use of any illicit drug (Tables 5.30- 5.32) and past-year marijuana use(Tables 5.40- 5.42), differences between those with a household member receiving welfare and without, and between those covered and not covered by health insurance, appear smaller within the Asian/Pacific Islander and most Hispanic subgroups than in the total surveyed population.

None of the socioeconomic variables analyzed in this chapter fully accounts for racial/ethnic differences in past-year use of cocaine. For example, as with the measures of past-year use of any illicit drug and marijuana, Asian/Pacific Islanders, Cubans, and other Hispanics are consistently relatively low in past-year cocaine use. This is true not only in comparisons between racial/ethnic groups as a whole, but also in comparisons between subclasses of racial/ethnic groups defined by levels of socioeconomic variables.

5.7 Need for Illicit Drug Abuse Treatment

Table 5.60 presents the estimated percentages of individuals aged 12 and older in the total surveyed population (excluding Native Americans) and within each of the seven racial/ethnic subgroups who needed illicit drug abuse treatment, by region, density, language of interview, family income, health insurance coverage, and receipt of welfare. (See Chapter 2 for a discussion of the criteria used to define illicit drug abuse treatment need.) Table 5.61 presents similar estimates for individuals aged 12 to 17, including estimates by school dropout status and family structure. Table 5.62 presents similar estimates for individuals aged 18 and older, including estimates by educational attainment, marital status, employment status, and number of own (i.e., biological) children.

For the total surveyed population, the first column of Table 5.60 shows that illicit drug abuse treatment need is:

·higher in the West (3.8%) than in the Northeast (2.2%), North Central (2.3%), and South (2.6%);

·higher among individuals in metropolitan statistical areas (MSAs) with populations greater than 1 million than among those outside MSAs (3.0% versus 2.3%);

·higher among those who used English in the interview than among those who used Spanish (2.7% versus 1.3%);

·higher among individuals with family incomes less than $20,000 than among individuals with

family incomes of more than $40,000 (3.6% versus 2.1%);

·higher among individuals who do not have health insurance coverage than among those who do (5.3% versus 2.3%); and

·higher among individuals with a family member in the household who receives welfare than among those without (7.4% versus 2.5%).

Similar patterns are observed in the subpopulations aged 12 to 17 (first column of Table 5.61) and 18 and over (first column of Table 5.62), except that, in the subpopulation aged 12 to 17, there are no significant differences by population density, family income, or health insurance coverage. The first column of Table 5.61 also shows that, in the total surveyed population aged 12 to 17, illicit drug abuse treatment need is higher among individuals who are not living with two biological parents than among those who are (5.3% versus 2.3%) and higher among school dropouts than among those still in school (13% versus 3.2%). The first column of Table 5.62 also shows that, in the population aged 18 and older, treatment need is higher among individuals with 9 to 11 years of schooling than among individuals with fewer or more years of schooling (4.5% versus 2.9% or less); higher among never-married individuals (6.0%) than among widowed/divorced/separated (2.9%) or married individuals (1.4%); higher among unemployed individuals (8.4%) than among individuals who are employed (2.7%) or not in the labor force (1.4%); and higher among individuals with no children of their own than among individuals with one or more (3.0% versus 2.1%). The differences by education, marital status, and employment status are similar to corresponding differences in past-year use of any illicit drug, marijuana, and cocaine (Tables 5.30-5.52).

Within racial/ethnic subgroups (second through eighth columns of Tables 5.60, 5.61, and 5.62), most socioeconomic differences in illicit drug abuse treatment need are similar to those observed in the total surveyed population. For example, in every racial/ethnic subgroup, the following have relatively high prevalences of need for treatment: individuals with 9-11 years of education, never-married individuals, unemployed individuals, individuals with no children of their own, and individuals who spoke English at the interview. Within racial/ethnic subgroups as in the total surveyed population, differences by population density and by family income appear to be modest.

Yet there are some important racial/ethnic differences in the effects of socioeconomic variables on the need for illicit drug abuse treatment:

·Racial/ethnic differences by region. The West is highest within most racial/ethnic subgroups, including Asian/Pacific Islanders, Puerto Ricans, non-Hispanic blacks, and non-Hispanic whites. However, within the Mexican subgroup, the estimated percentage needing treatment in the West is not significantly different from the percentages needing treatment in the North Central and South.

·Racial/ethnic differences by family structure. Differences by family structure in adolescent illicit drug abuse treatment need (Table 5.61) are smaller within Hispanic subgroups (except Mexicans) than in the total surveyed population. For example, within the Cuban subgroup, the percentage needing treatment is approximately equal both among adolescents in mother/father families (2.8%) and among those in other family types (3.0%).

·Racial/ethnic differences by receipt of welfare. Differences between those receiving and not receiving welfare appear smaller within most Hispanic subgroups than in the total surveyed population.

None of the socioeconomic variables analyzed in this chapter fully accounts for racial/ethnic differences in illicit drug abuse treatment need. For example, Asian/Pacific Islanders are consistently relatively low in need for illicit drug use treatment and Mexicans are consistently relatively high, both in comparisons across racial/ethnic subgroups as a whole and across segments of those racial/ethnic subgroups defined by socioeconomic variables. 

5.8 Alcohol Dependence

Table 5.70 presents the estimated percentages of individuals aged 12 and older in the total surveyed population (excluding Native Americans) and within each of the seven racial/ethnic subgroups who reported alcohol dependence, by region, density, language of interview, family income, health insurance coverage, and receipt of welfare. Table 5.71 presents similar estimates for individuals aged 12 to 17, including estimates by school dropout status and family structure. Table 5.72 presents similar estimates for individuals aged 18 and older, including estimates by educational attainment, marital status, employment status, and number of own (i.e., biological) children.

For the total surveyed population, the first column of Table 5.70 shows that socioeconomic differences in alcohol dependence are more similar to socioeconomic differences in the past-year use of any illicit drug (Table 5.30), marijuana (Table 5.40), and cocaine (Table 5.50) than to socioeconomic differences in the past-year use of alcohol (Table 5.20). Dependence on alcohol is higher in the West (4.9%) than in the Northeast (2.9%), North Central (3.2%), and South (3.1%); higher among individuals with family incomes less than $20,000 than among individuals with incomes of more than $40,000 (3.9% versus 3.2%); higher among individuals who do not have health insurance coverage than among those who do (5.5% versus 3.1%); and higher among individuals with a family member in the same household receiving welfare than among those without (4.8% versus 3.4%).

In the subpopulation aged 12 to 17 (first column of Table 5.71), most patterns are different: Individuals outside MSAs are higher than those within MSAs (3.6% versus 2.9% or less), individuals who used English in the interview are higher than those who used Spanish (2.7% versus 0.7%), and there are no significant differences by family income, health insurance coverage, or receipt of welfare. The first column of Table 5.71 also shows that, in the total surveyed population aged 12 to 17, alcohol dependence is not significantly different between individuals who are not living with two biological parents and those who are, or between school dropouts and those still in school. The first column of Table 5.72 also shows that, in the population aged 18 and older, alcohol dependence is higher among individuals with 9 to 11 years of schooling than among individuals with fewer years of schooling (4.0% versus 2.4%); higher among never-married individuals (7.6%) than among widowed/divorced/separated (3.0%) or married individuals (2.4%);higher among unemployed individuals (6.6%) than among individuals who are employed (4.0%) or not in the labor force (2.1%); and higher among individuals with no children of their own than among individuals with one or more children (4.3% versus 2.5%).

Within racial/ethnic subgroups (second through eighth columns of Tables 5.70, 5.71, and 5.72), differences by region, population density, family income, family structure, and language of interview are generally modest, as in the total surveyed population, and, with few exceptions, such differences are not statistically significant within subgroups. Yet the data do suggest that some socioeconomic differences in alcohol dependence depend upon race/ethnicity:

·Racial/ethnic differences by population density. Within the Mexican subgroup, individuals outside MSAs are significantly more likely to be alcohol-dependent than individuals within MSAs (in the Mexican population aged 12 or older, 12% versus 5.9% or less—see Table 5.70). The same pattern of higher Mexican alcohol dependence outside MSAs holds in the subpopulations aged 12 to 17 (Table 5.71) and aged 18 and older (Table 5.72).

None of the socioeconomic variables analyzed in this chapter fully accounts for racial/ethnic differences in alcohol dependence. For example, Asian/Pacific Islanders and Cubans are consistently relatively low in alcohol dependence and Mexicans are consistently high both in the total surveyed population and within subclasses defined by socioeconomic variables.

5.9 Heavy Smoking

Table 5.80 presents the estimated percentages of individuals aged 12 and older in the total surveyed population (excluding Native Americans) and within each of the seven racial/ethnic subgroups who reported smoking heavily (a pack or more per day,) by region, density, language of interview, family income, health insurance coverage, and receipt of welfare. Table 5.81 presents similar estimates for individuals aged 12 to 17, including estimates by school dropout status and family structure. Table 5.82 presents similar estimates for individuals aged 18 and older, including estimates by educational attainment, marital status, employment status, and number of own (i.e., biological) children.

For the total surveyed population, the first column of Table 5.80 shows that the North Central region (15%) and South (15%) are significantly higher in heavy smoking than the West (12%), but not significantly higher than the Northeast (14%). Heavy smoking is also higher among individuals who are not in metropolitan statistical areas (MSAs) than among those who are in MSAs with populations greater than 1 million (17% versus 11%); higher among those who used English in the interview than among those who used Spanish (14% versus 3.3%); higher among individuals with family incomes of $40,000 or less than among individuals with family incomes of greater than $40,000 (15% or higher versus 12%); higher among individuals who do not have health insurance coverage than among those who do (21% versus 13%); and higher among individuals who receive welfare than among those who do not (22% versus 13%). In general, socioeconomic differences in heavy smoking are similar to socioeconomic differences in past-year cigarette use (Table 5.10).

Similar patterns are observed in the subpopulations aged 12 to 17 (first column of Table 5.81) and 18 and over (first column of Table 5.82), except that, in the subpopulation aged 12 to 17, there are no significant differences by family income, health insurance coverage, or region (except that the North Central _2.1%_ is significantly higher than the Northeast _1.0%_). The first column of Table 5.81 also shows that, in the total surveyed population aged 12 to 17, heavy smoking is higher among individuals who are not living with two biological parents than among those who are (2.4% versus 1.1%), and much higher among school dropouts than among those who are still in school (14% versus 1.3%). The first column of Table 5.12 also shows that, in the population aged 18 and older, heavy smoking is higher among individuals with 9 to 11 years of schooling than among individuals with more or fewer years of schooling (25% versus 17% or less); higher among widowed/divorced/separated individuals (20%) than among never-married (14%) and currently married individuals (14%); higher among unemployed individuals (24%) than among individuals who are employed (16%) or not in the labor force (12%); and slightly higher among individuals with one or more of their own children than among individuals with no children of their own (16% versus 14%). These socioeconomic differences are again quite similar to corresponding differences in past-year smoking (Tables 5.11- 5.12).

Within racial/ethnic subgroups (second through eighth columns of Tables 5.80, 5.81, and 5.82), most socioeconomic differences in heavy smoking are similar to those observed in the total surveyed population. Regardless of racial/ethnic subgroup, individuals aged 18 and older had relatively high prevalences of heavy smoking if they resided outside metropolitan areas with populations of more than 1 million, responded to the NHSDA interview in English rather than in Spanish, had between 9 and 12 years of schooling, were unemployed, or resided in households that received welfare (Table 5.82). Regardless of racial/ethnic subgroup, individuals aged 12 to 17 had relatively high prevalences of heavy smoking if they were school dropouts or lived in households where both biological parents were not present (Table 5.81). Within racial/ethnic subgroups as in the total surveyed population, differences by region, family income, marital status, and number of own children are relatively small, and these differences are generally not statistically significant within subgroups. As in the case of past-year cigarette use (Section 5.1), differences in heavy smoking by health insurance coverage and receipt of welfare appear smaller within Hispanic subgroups, except Cubans, than in the total surveyed population.

None of the socioeconomic variables analyzed in this chapter fully accounts for racial/ethnic differences in heavy smoking. For example, in comparing adolescents aged 12 to 17 and adults aged 18 and older across the seven subgroups analyzed in this chapter, non-Hispanic whites have the highest overall percentages of heavy smokers; and non-Hispanic white adolescents are also generally the highest when racial/ethnic subgroups are compared within particular regions, population densities, languages of interview, levels of family income, and other socioeconomic subclasses. 

5.10 Heavy Alcohol Use

Table 5.90 presents the estimated percentages of individuals aged 12 and older in the total surveyed population (excluding Native Americans) and within each of the seven racial/ethnic subgroups who reported heavy alcohol use in the past month, by region, density, language of interview, family income, health insurance coverage, and receipt of welfare. Table 5.91 presents similar estimates for individuals aged 12 to 17, including estimates by school dropout status and family structure. Table 5.92 presents similar estimates for individuals aged 18 and older, including estimates by educational attainment, marital status, employment status, and number of own (i.e., biological) children.

For the total surveyed population, the first column of Table 5.90 shows that, regardless of region, population density, or language of interview, about 5% of individuals aged 12 and older were heavy alcohol users. Heavy drinking is higher among individuals with family incomes of less than $20,000 among individuals with family incomes of greater than $40,000 (6.2% versus 4.5%), and higher among individuals who do not have health insurance coverage than among those who do (8.6% versus 4.6%). The difference between individuals in households receiving and not receiving welfare is not statistically significant (6.0% versus 5.1%).

In the subpopulations aged 12 to 17 (first column of Table 5.91) and 18 and over (first column of Table 5.92), heavy drinking has a modest negative association with family income but most socioeconomic differences are not statistically significant. Among adolescents aged 12 to 17, heavy drinking is higher among those who are not living with two biological parents than among those who are (2.1% versus 1.3%), and also higher among school dropouts than among those still in school (6.2% versus 1.5%). Among adults aged 18 and older, heavy drinking is slightly more common among those with 9 to 11 years of schooling than among those with more or fewer years (6.8% versus 5.7% or less). Individuals who have never married (11%), are unemployed (11%), have no children of their own (6.5%), have no health insurance coverage (9.4%), or reside in households receiving welfare (7.0%) also have high prevalences of heavy drinking relative to the total surveyed population.

Within racial/ethnic subgroups (second through eighth columns of Tables 5.90, 5.91, and 5.92), most socioeconomic differences in heavy drinking are similar to those observed in the total surveyed population. Almost without exception, adults aged 18 and older within each racial/ethnic subgroup appeared to have relatively high prevalences of heavy drinking when their family income was $40,000 or less, if they had never married, if they had no children of their own, or if they lived in households where someone received welfare, but most socioeconomic differences within racial/ethnic subgroups were not statistically significant. As in the total surveyed population, differences in heavy drinking by region and population density appear to be modest within racial/ethnic subgroups.

Tables:  Percentages Using Substance by Race/Ethnicity and Sociodemographic Characters:  1991-1993 

Previous Page Page Top TOC Next Page

This page was last updated on May 19, 2008.