Substance Use Among Foreign-Born Youths in the United States: Does the Length of Residence Matter?

The prevention of substance use is a critical component of health promotion among youths. Of the 72.3 million youths under 18 years of age in the United States in 2000, 2.8 million were foreign-born.1 Research has suggested that foreign-born youths experience increasing risk of substance use as they become assimilated into US society (i.e., become acculturated).2,3 This study provides the first national estimates of the prevalence of substance use among foreign-born youths aged 12 to 17 years and explores the association between acculturation, defined as the length of residence in the United States, and substance use.

METHODS

Data from the 1999 and 2000 National Household Survey on Drug Abuse (NHSDA) were used.4–6 The NHSDA is an ongoing nationally representative survey of the noninstitutional civilian population aged 12 years and older.4 Data were collected by a combination of computer-assisted personal interview and audio computer-assisted self-interview techniques to enhance privacy. Respondents could answer in English or Spanish. Weighted response rates were 91 percent for household screening and 80 percent for youth interviews.5,6 The interview response rate for youths was 86 percent among Hispanics and 82 percent among non-Hispanic blacks. Only 0.2 percent of selected youths did not respond because of language barrier.

Youths born outside the United States were classified as foreign-born and others were classified as US-born, based on the question, “Were you born in the United States?” Foreign-born respondents were also asked, “In what country or US territory were you born?” and “About how long have you lived in the United States?” The sample of 50 947 youths represented 23.2 million youths in the nation, of which 7.1 percent were foreign-born. Among the foreign-born youths, 28.4 percent were born in Mexico, 5.1 percent in Germany, 4.6 percent in the Philippines, and 3.0 percent in India, Vietnam, and Korea (North and South).

Prevalence estimates were computed for past-month use of cigarettes, alcohol (any, binge, and heavy), marijuana, and other illicit drugs (cocaine, heroin, hallucinogens, inhalants, and nonmedical use of prescription-type pain relievers, tranquilizers, stimulants, and sedatives). Past-month use is defined as use at least 1 time during the 30 days before the interview. Cigarette use includes smoking all or part of a cigarette. Binge alcohol use is having 5 or more drinks on the same occasion at least once in the past 30 days. Heavy alcohol use is having 5 or more binge days in the last 30 days. Comparisons of prevalence estimates between foreign-born and US-born youths were made using t-tests. Multiple logistic regressions were run with SUDAAN software (Research Triangle Institute, Research Triangle Park, NC) to assess the effect of length of residence with controls for gender, age, race/ethnicity, family income, region, and population density.

RESULTS

Substance-use prevalence rates were lower (P < .05) among foreign-born youths than among US-born youths, especially for youths aged 16 to 17 years (Table 1 ▶). Consistent with previous studies of acculturation, we found heterogeneity among Hispanic subgroups.7,8

TABLE 1—

Percentage of Youths Ages 12 to 17 Years Reporting Past-month Substance Use, by Country of Birth Status: 1999 and 2000 National Household Surveys on Drug Abuse

	All		US–born		Foreign-born
	%	SE	%	SE	%	SE	Pa
Cigarette use	14.2	0.21	14.6	0.22	8.7	0.65	<.001
Gender
Male	13.8	0.27	14.1	0.29	9.1	0.89	<.001
Female	14.6	0.29	15.1	0.31	8.3	0.98	<.001
Age, y
12–13	3.8	0.19	3.9	0.20	2.1	0.65	.011
14–15	13.2	0.31	13.6	0.33	7.1	1.04	<.001
16–17	25.6	0.43	26.6	0.46	14.9	1.28	<.001
Race/ethnicity
Non-Hispanic White/other	16.5	0.27	16.6	0.28	13.1	1.64	.035
Non-Hispanic Black	7.3	0.40	7.4	0.40	4.8	2.79	.343
Non-Hispanic Asian/Pacific Islander	8.6	0.94	8.8	1.27	8.3	1.50	.837
Hispanic Mexican	11.3	0.57	12.6	0.71	6.8	0.78	<.001
Hispanic other	10.9	0.92	11.7	1.11	8.6	1.53	.101
Alcohol use	16.5	0.21	16.8	0.22	12.4	0.70	<.001
Gender
Male	16.5	0.29	16.7	0.30	14.2	1.02	.018
Female	16.4	0.29	16.9	0.31	10.6	0.95	<.001
Age, y
12–13	4.5	0.20	4.6	0.21	3.2	0.65	.036
14–15	15.6	0.35	16.0	0.36	9.5	0.96	<.001
16–17	29.3	0.42	30.0	0.44	21.9	1.53	<.001
Race/ethnicity
Non-Hispanic White/other	18.2	0.26	18.3	0.26	14.2	1.45	.005
Non-Hispanic Black	9.3	0.45	9.1	0.46	13.0	2.88	.191
Non-Hispanic Asian/Pacific Islander	9.9	0.96	10.2	1.24	9.4	1.42	.664
Hispanic Mexican	17.3	0.69	18.6	0.80	12.8	1.24	<.001
Hispanic other	16.0	1.06	16.9	1.18	13.6	2.18	.179
Binge alcohol use	10.3	0.17	10.5	0.18	6.8	0.56	<.001
Gender
Male	11.3	0.24	11.5	0.25	8.6	0.87	.002
Female	9.3	0.22	9.6	0.23	4.9	0.64	<.001
Age, y
12–13	1.9	0.13	2.0	0.14	1.0	0.43	.033
14–15	9.2	0.27	9.5	0.28	5.0	0.76	<.001
16–17	19.8	0.37	20.4	0.39	12.8	1.21	<.001
Race/ethnicity
Non-Hispanic White/other	11.7	0.21	11.8	0.21	7.9	1.17	.001
Non-Hispanic Black	4.6	0.30	4.6	0.31	5.1	2.06	.805
Non-Hispanic Asian/Pacific Islander	4.6	0.61	4.5	0.77	4.9	1.06	.750
Hispanic Mexican	11.2	0.58	12.4	0.67	7.0	0.93	<.001
Hispanic other	9.6	0.95	9.9	1.07	9.0	1.73	.652
Heavy alcohol use	2.5	0.09	2.6	0.10	1.2	0.23	<.001
Gender
Male	3.1	0.14	3.2	0.15	1.7	0.37	<.001
Female	1.9	0.10	2.0	0.11	0.7	0.28	<.001
Age, y
12–13	0.2	0.04	0.2	0.04	0.1	0.05	.006
14–15	1.7	0.12	1.8	0.13	0.4	0.15	<.001
16–17	5.7	0.23	6.0	0.24	2.8	0.60	<.001
Race/ethnicity
Non-Hispanic White/other	3.0	0.12	3.0	0.12	1.8	0.52	.018
Non-Hispanic Black	0.8	0.11	0.8	0.11	0.5	0.53	.658
Non-Hispanic Asian/Pacific Islander	0.7	0.25	0.8	0.35	0.5	0.33	.519
Hispanic Mexican	2.7	0.33	3.1	0.40	1.5	0.46	.009
Hispanic other	1.8	0.31	1.9	0.39	1.4	0.74	.580
Marijuana use	7.2	0.15	7.5	0.16	3.5	0.44	<.001
Gender
Male	7.8	0.21	8.1	0.22	3.9	0.57	<.001
Female	6.6	0.20	6.9	0.21	3.1	0.68	<.001
Age, y
12–13	1.3	0.12	1.3	0.12	0.5	0.23	.001
14–15	6.9	0.24	7.2	0.25	3.4	0.67	<.001
16–17	13.4	0.33	14.1	0.34	5.9	0.85	<.001
Race/ethnicity
Non-Hispanic White/other	7.7	0.18	7.8	0.19	5.2	0.91	.006
Non-Hispanic Black	6.0	0.38	6.0	0.37	6.6	3.06	.850
Non-Hispanic Asian/Pacific Islander	4.4	0.73	5.2	1.11	3.3	0.91	.207
Hispanic Mexican	6.9	0.45	8.3	0.56	1.9	0.42	<.001
Hispanic other	6.7	0.73	8.0	0.92	2.8	0.91	<.001
Other illicit drug use	4.6	0.11	4.6	0.12	3.6	0.40	.013
Gender
Male	4.2	0.15	4.2	0.16	3.5	0.54	.189
Female	4.9	0.17	5.0	0.18	3.7	0.61	.037
Age, y
12–13	2.6	0.15	2.7	0.15	2.2	0.55	.444
14–15	4.8	0.20	4.8	0.21	4.0	0.77	.291
16–17	6.3	0.24	6.4	0.25	4.2	0.68	.002
Race/ethnicity
Non-Hispanic White/other	4.8	0.14	4.8	0.14	3.7	0.81	.178
Non-Hispanic Black	3.4	0.25	3.4	0.25	4.2	1.63	.607
Non-Hispanic Asian/Pacific Islander	3.0	0.51	2.9	0.61	3.2	0.81	.807
Hispanic Mexican	5.6	0.44	6.3	0.52	3.0	0.63	<.001
Hispanic other	4.2	0.59	3.9	0.58	4.9	1.35	.478

Notes. “Non-Hispanic White/other” includes Non-Hispanic American Indians, Alaska Natives, and persons reporting more than 1 race. “Hispanic other” includes Puerto Rican, Cuban, Central or South American, Hispanic Caribbean, or other Hispanic.

^aTest of difference between US–born prevalence and foreign-born prevalence.

Logistic regression showed that differences in prevalence between foreign- and US-born youths diminished with increasing length of residence in the United States among foreign-born youths (Table 2 ▶). Foreign-born youths who had resided in the United States for less than 5 years had lower (P < .05) prevalence than did US-born youths for all the substances. Prevalence estimates for foreign-born youths in the United States for 10 or more years were not significantly different from estimates for US-born youths, except for heavy alcohol use. Separate regressions for ages 12 to 14 years and 15 to 17 years yielded similar results.

TABLE 2—

Adjusted Odds Ratios for Past-Month Substance Use by Foreign-Born Youths, by Length of Residence in the United States: 1999 and 2000 National Household Surveys on Drug Abuse

	Less Than 5 Years in the United States		5 to Less Than 10 Years in the United States		10 or More Years in the United States
Dependent Variable (Past-Month Use)	Odds Ratio	(95% CI)	Odds Ratio	(95% CI)	Odds Ratio	(95% CI)
Cigarette use	.41	(0.29, 0.56)	.52	(0.37, 0.74)	.82	(0.63, 1.06)
Alcohol use	.47	(0.35, 0.62)	.68	(0.52, 0.90)	.88	(0.72, 1.06)
Binge alcohol use	.39	(0.28, 0.56)	.56	(0.39, 0.81)	.82	(0.62, 1.07)
Heavy alcohol use	.19	(0.08, 0.46)	.60	(0.26, 1.39)	.57	(0.34, 0.96)
Marijuana use	.11	(0.05, 0.22)	.31	(0.18, 0.56)	.78	(0.57, 1.08)
Other illicit drug use	.58	(0.35, 0.95)	.68	(0.44, 1.03)	.88	(0.62, 1.26)

Notes. CI = confidence interval. Model covariates are gender, age, race/ethnicity, family income, region, and population density. Reference group is US–born youths.

An additional analysis used the language selected by the respondent for the survey interview as an alternative measure of acculturation among Hispanics. For each of the substances, we found that the prevalence estimate among Hispanics was higher (P < .05) for those who responded in English than for those who responded in Spanish.

DISCUSSION

This study reinforces the findings of previous studies reporting lower rates of substance use among foreign-born youths compared with US-born youths, but increased risk of use as they become acculturated.2,3,7,9,10 A better understanding of these results could be gained by studying how acculturation interacts with known risk and protective factors for substance use.11 Acculturation occurs through contacts with parents and peers, formal education, and exposure to media such as television, movies, and magazines. Each of these could influence a youth’s propensity to use substances. In addition, access to substances of abuse may be greater for youths who are more fluent in English or more in touch with local customs for obtaining substances. Research in this area should help prevention planners design programs that appropriately consider acculturation to reduce substance use among immigrant youth.

A limitation of this study is that it is based on self-reports; therefore, prevalence estimates may be subject to underreporting. This could vary by country of birth as well as by acculturation. Views of substance use may differ across cultures.