“Heterogeneity in Tobacco-Use Behaviors among U.S. Blacks per Global Region of Origin”

Abstract

Background:

Heterogeneity in tobacco-use behaviors among U.S. blacks by global region of origin and age at immigration was examined.

Methods:

Self-identified black participants from the 2006–2015 Current Population Survey-Tobacco Use Supplement were included (n=47,857). Countries of origin were classified by global regions (U.S., Africa, Europe, West Indies). Multivariable logistic regression models, adjusted for sociodemographic characteristics, were used to examine the association of global region of origin and age at immigration with tobacco-use behaviors.

Results:

Prevalence of current cigarette smoking among U.S. blacks varied significantly by global region of origin (U.S.-born=17.4%, Europe-born=17.7%, Africa-born=4.7%, West-Indies-born=4.9%; p<0.01). Furthermore, foreign-born blacks were less likely than U.S.-born blacks to smoke menthol flavored cigarettes (p<0.05). Participants who immigrated to the U.S. at ≥13 years old were less likely than U.S.-born blacks to start regular smoking as minor (p<0.05).

Conclusion:

Tobacco-use behaviors differ by global region of origin and age at immigration in the U.S. black population.

1. Introduction

With an estimated 45,000 deaths annually attributable to tobacco-related diseases, smoking remains the primary cause of death of African-Americans in the U.S. [1]. In 2015, 16.7% of non-Latino black adults (≥18 years old) in the U.S. reported current cigarette smoking [2]. If existing trends in tobacco use persist, tobacco-related mortality will continue to rise [3]. Therefore, understanding heterogeneity in tobacco-use behaviors among U.S. blacks is essential to develop tailored strategies to reduce tobacco use within this population.

1.1. Theoretical/Conceptual Framework

Although most U.S. blacks are U.S.-born, 3.8 million foreign-born blacks are currently living in the U.S., representing 8.7% of the U.S. black population in 2013, and 16.5% by 2060 [4]. Berry’s acculturation framework posits that individual-level (e.g., age) and group-level (e.g., social norms of origin country) characteristics influence the acculturation process [5]; thus, age of immigration and prevalence of smoking in the origin country may be associated with the likelihood of foreign-born blacks smoking in the host country. Furthermore, experiences of racial/ethnic discrimination in the host country may lead to substance-use behaviors, as shown in previous research among immigrants [6]. Previous studies found that foreign-born blacks had a lower prevalence of current smoking and daily smoking than U.S.-born blacks [7, 8], suggesting that uptake of smoking may be part of the acculturation process in the U.S. Nevertheless, these studies did not examine variations in smoking prevalence among foreign-born blacks. A recent study of U.S. blacks of West Indian, Haitian, Latin American, and African descent found a lower prevalence of smoking among foreign-born blacks, compared with U.S.-born blacks, with variation by age at immigration [9], possibly attributable to differences in norms and attitudes towards smoking between country of origin and the U.S. [10]. It is unclear, however, if this observation holds true for Europe-born blacks, since they were not included in this study. Furthermore, tobacco-use behaviors other than current/daily smoking were not examined in previous studies. For example, it is important to understand variations in age of initiation and menthol cigarette smoking, given that both are negatively associated with smoking cessation [11, 12], especially among racial/ethnic minorities [12]. The aim of this study was to explore whether there is heterogeneity in tobacco-use behaviors (including current cigarette and cigar smoking, time to first cigarette, and menthol cigarette use) among U.S. blacks by global region of origin and age at immigration.

2. Methods

2.1. Participants

Subjects included self-identified black adult participants (≥18 years old) who reported countries of origin in the following global regions: U.S. (including Guam, Puerto Rico, the U.S. Virgin Islands, and other U.S. island areas), Africa (including both North and Sub-Saharan Africa), West Indies, and Europe (n=47,857). Participants born in other global regions were excluded, due to small sample sizes (n=1,033).

2.2. Data Source

Data were from the 2006–07, 2010–11, and 2014–15 Current Populations Survey -Tobacco Use Supplement (CPS-TUS). Waves prior to 2006 were not included, due to inconsistencies in measures (including race and menthol cigarette use). The CPS-TUS is administered as part of the U.S. Census Bureau’s CPS and comprised of a nationally representative sample of approximately 240,000 individuals per wave. The survey is a key source of national, state, and sub-state level data from U.S. households regarding smoking, use of tobacco products, and tobacco-related norms, attitudes, and policies [13].

2.3. Measures

Self-reported country of origin was classified as one of four global regions listed above. Age at immigration into the U.S. was estimated based on age, survey year, and self-reported year of entry into the U.S., and dichotomized as entry at <13 years old vs. at ≥13 years old [9]. Participants who reported currently smoking cigarettes or cigars at least some days were classified as current users of these products. Among established smokers (i.e., >100 cigarettes in a lifetime), participants were classified as former smokers, or not, and whether they began smoking regularly as minor. Among current cigarette smokers, participants reported if they smoked their first cigarette within 30 minutes after waking up, regularly use menthol cigarettes, and tried to quit smoking in the past 12 months. Covariates included survey year, age, educational attainment (no high-school diploma, high-school diploma or GED, some college or associate’s degree, and bachelor’s degree or greater), income (<$35,000, $35,000-$59,999, ≥$60,000, not reported), and U.S. census region.

2.4. Statistical Analysis

CPS-TUS survey weights were used to allow for estimates representative of the non-institutionalized U.S. black population. Multivariable logistic regression models were used to examine the differences in tobacco-use behaviors by global region of origin and by age at immigration separately. All models adjusted for the sociodemographic characteristics listed above, which were forced and retained, regardless of statistical significance. For each analysis, participants with missing outcomes data were excluded from the analysis (see Table 2 for sample sizes). In analyses including age at immigration, 176 participants were excluded because of inability to determine age at immigration based on responses. All analyses were conducted using SAS® version 9.4 (SAS Institute: Cary, North Carolina). Because this was a secondary data analysis using de-identified data, it was deemed exempted from institutional-review-board review, as determined by the National Institutes of Health Office of Human Subjects Research Protections.

3. Results

U.S.-born blacks were older and least likely to have a bachelor’s degree or higher educational attainment, an income ≥$60,000, and to reside in the Northeast (Table 1). Europe-born blacks were most likely to have immigrated to the U.S. at <13 years old. Current cigarette smoking was significantly less prevalent among Africa-born blacks (4.7%) and West-Indies-born blacks (4.9%), compared with U.S.-born blacks (17.4%; p<0.05), whereas Europe-born blacks had a similar prevalence (17.7%) to U.S.-born blacks (Table 2). Compared with U.S.-born black established smokers, Africa-born and West-Indies-born black established smokers were more likely to be former smokers, and Africa-born black established smokers were less likely to start smoking regularly as minor. Compared with U.S.-born current cigarette smokers, Africa-born black current cigarette smokers were less likely to have their first cigarette within 30 minutes after waking up, and Africa-born and West-Indies-born black current cigarette smokers were less likely to regularly smoke menthol cigarettes. Compared with their U.S.-born counterparts, foreign-born blacks, regardless of age at immigration, were less likely to be current cigarette or cigar smokers, and foreign-born blacks were less likely to smoke menthol-flavored cigarettes. In contrast, only those who immigrated at ≥13 years old were more likely, than their U.S.-born counterparts, to become former smokers, less likely to start smoking regularly as minor, and to have their first cigarette within 30 minutes after waking up.

Table 1.

Sociodemographic characteristics of the analytic sample of US blacks from the 2006–2015 Current Populations Survey-Tobacco Use Supplement (CPS-TUS).

	Global Region of Origin
Variable	U.S. (n= 43,560)	Europe (n=192)	Africa (n=1,911)	West Indies (n=2,194)
Survey wave
2006–07	31.4%	33.1%	21.1%	28.4%
2010–11	33.0%	20.8%	35.2%	34.4%
2014–15	35.6%	46.0%	43.8%	37.2%
Mean age, years (SE)a	43.5 (0.1)	36.8 (1.1)	38.4 (0.4)	45.8 (0.4)
Sex
Male	44.1%	40.4%	57.8%	45.2%
Female	55.9%	59.6%	42.2%	54.8%
Educational attainmentb
No high-school diploma	16.4%	9.5%	11.8%	19.9%
High-school diploma or GED	33.8%	25.5%	23.8%	28.9%
Some college or associate’s degree	31.9%	35.1%	27.9%	29.4%
Bachelor’s degree or greater	17.8%	30.0%	36.4%	21.9%
Annual incomeb
Not reported	15.5%	12.9%	15.9%	18.5%
<$35,000	41.5%	27.5%	35.2%	31.1%
$35,000 to $59,999	17.8%	18.4%	20.0%	20.6%
≥$60,000	25.1%	41.2%	28.9%	29.8%
U.S. census regionb
Northeast	14.4%	17.7%	27.3%	54.3%
Midwest	18.3%	9.1%	17.9%	3.0%
South	58.1%	63.2%	39.3%	40.4%
West	9.2%	10.1%	15.5%	2.3%
Age at immigrationb
<13 years old	−	69.2%	9.8%	18.4%
≥13 years old	−	30.8%	90.2%	81.6%

Due to rounding, percentages may not sum to 100%.

^aWeighted Analysis of Variance (ANOVA) test p<0.01.

^bWeighted Chi-square test p<0.05.

Table 2.

Bivariable and multivariable analyses of associations of global region of origin and age of immigration with smoking behaviors of US blacks from the 2006–2015 CPS-TUS.

	Full Sample				Established Smokers				Current Smokers
	Current Cigarette Smoker (n=47,556)		Current Cigar Smoker (n=47,126)		Former Smoker (n=14,011)		Started Regular Smoking as Minor (n=13,175)		First Cigarette within 30 Minutes after Waking Up (n=7,405)		Regularly Menthol Cigarette Smoker (n=7,747)		Tried Quitting in the Past 12 Months (n=7,585)
Variable	%	AOR (95% CI)	%	AOR (95% CI)	%	AOR (95% CI)	%	AOR (95% CI)	%	AOR (95% CI)	%	AOR (95% CI)	%	AOR (95% CI)
Global region of origin
U.S.	17.4%	The referent	3.2%	The referent	40.6%	The referent	45.5%	The referent	47.4%	The referent	74.4%	The referent	44.0%	The referent
Europe	17.7%	1.42 (0.91, 2.23)	2.6%	0.83 (0.19, 3.57)	33.1%	0.89 (0.44, 1.83)	42.1%	1.02 (0.49, 2.12)	42.1%	0.99 (0.44, 2.25)	64.3%	0.54 (0.23, 1.27)	35.0%	0.60 (0.26, 1.39)
Africa	4.7%	0.25 (0.19, 0.33)	0.1%	0.03 (0.01, 0.09)	51.0%	1.63 (1.10, 2.41)	32.7%	0.57 (0.38, 0.86)	25.5%	0.47 (0.25, 0.87)	35.1%	0.19 (0.11, 0.34)	54.2%	1.41 (0.83, 2.39)
West Indies	4.9%	0.24 (0.19, 0.30)	1.1%	0.41 (0.24, 0.70)	55.3%	1.63 (1.19, 2.23)	45.4%	0.78 (0.55, 1.09)	34.2%	0.59 (0.35, 1.01)	54.0%	0.47 (0.29, 0.76)	46.4%	1.13 (0.70, 1.82)
Age at immigrationa
<13 years old	9.1%	0.55 (0.39, 0.78)	0.8%	0.21 (0.08, 0.58)	34.9%	1.23 (0.73, 2.08)	50.4%	1.23 (0.71, 2.11)	34.7%	0.68 (0.32, 1.40)	60.6%	0.46 (0.24, 0.89)	42.7%	0.86 (0.45, 1.65)
≥13 years old	4.7%	0.24 (0.20, 0.29)	0.8%	0.27 (0.16, 0.45)	55.6%	1.62 (1.24, 2.11)	38.0%	0.62 (0.47, 0.82)	30.9%	0.55 (0.36, 0.83)	43.6%	0.31 (0.21, 0.46)	49.8%	1.25 (0.85, 1.84)

Each behavior was modeled separately, using U.S.-born blacks as referent, adjusting for sociodemographic factors (age, sex, educational attainment, annual income, and U.S. census region) and survey waves. Boldface indicates statistical significance (p<0.05).

^aFor age at immigration, the sample size for current cigarette smokers was 47,444, for current cigar smokers was 47,104, for former smokers was 14,001, and for started regular smoking as minor was 13,165.

4. Discussion

Previous studies have found that foreign-born blacks were less likely than U.S.-born blacks to be current and daily smokers [7, 8]. Within the foreign-born black populations, West-Indies-and Africa-born blacks were less likely than U.S.-born blacks to be current smokers [9]. Our findings among West-Indies and Africa-born blacks were consistent with these previous studies. In contrast, our finding that Europe-born blacks do not differ from U.S.-born blacks in prevalence of current smoking was novel. This finding appears to contradict the immigrant health paradox (health-protective effect of foreign-born individuals despite lower SES, compared with their U.S.-born counterparts) [14]. Further analysis revealed that 67.4% of Europe-born blacks in our sample migrated to the U.S. at <13 years old, compared with 7.3% of Africa-born and 15.5% of West-Indies-born blacks. Given that age of immigration can be associated with degree of acculturation [15], a high proportion of Europe-born blacks immigrating at <13 years old may be linked to a higher degree of acculturation to U.S. practices, including cigarette smoking. Additionally, the overall prevalence of smoking in Europe is 28%, compared with 17% in the U.S., 13% in Africa and 5–15% [16] in the West Indies. This would suggest that social norms in the country of origin may influence tobacco use among foreign-born blacks [17].

The study findings also suggest that African-born blacks were less likely to smoke their first cigarettes within 30 minutes after waking, and Africa-born and West-Indies-born blacks were less likely to use menthol cigarettes than their U.S.-born counterparts, results that have not be previously reported. Onset of regular smoking during adolescence is associated with higher levels of subsequent nicotine dependence during adulthood [18]. Therefore, the negative association between an African country of origin and time to first cigarette could be explained by the lower proportion of African-born blacks starting regular smoking as minor, compared with U.S.-born blacks. The marketing and promotion of menthol cigarettes have historically targeted African-American communities [19]. Consequently, Africa-born and West-Indies-born blacks may not be exposed to menthol cigarette marketing during childhood, as most immigrated to the U.S. at ≥13 years old. Thus, our findings may imply that the high prevalence of menthol cigarette use among U.S.-born blacks is likely an outcome of marketing, rather than preference.

One study limitation is that, despite aggregating data across multiple years, sample sizes for certain models were small and potentially underpowered. Therefore, our findings should be considered as exploratory, and no conclusion should be drawn from underpowered models. Another limitation is that household income data did not allow for assessment of poverty levels. Other limitations include variations in tobacco-use behaviors within global region of origin, the incapacity to test the interaction between global region of origin and age at immigration due to low statistical power, and limited generalizability to U.S. blacks born outside of the global regions of study.

5. Conclusion

Our findings indicate that tobacco-use behaviors vary within the U.S. black population by global region of origin. Future studies with larger sample sizes are needed to confirm our findings and to examine the potential reasons for such differences, including the role of culture, and tobacco-use perceptions and beliefs, to guide development of tobacco-control interventions tailored to the diversity of the U.S. black population.