Acculturation and drug addiction stigma among Latinos and African Americans: An examination of a church-based sample

Abstract

Background

Substance use patterns among Latinos likely reflect changes in attitudes resulting from acculturation, but little is known about Latinos’ attitudes regarding drug addiction.

Methods

We surveyed a church-based sample of Latinos and African Americans (N=1,235) about attitudes toward drug addiction and socio-demographics. Linear regression models compared Latino subgroups with African-Americans.

Results

In adjusted models, Latinos had significantly higher drug addiction stigma scores compared to African Americans across all subgroups (U.S.-born Latinos, β = 0.22, p<.05; foreign-born Latinos with high English proficiency, β = 0.30, p<.05; and foreign-born Latinos with low English proficiency, β =0.49, p<.001). Additionally, Latinos with low English proficiency had significantly higher mean levels of drug use stigma compared Latinos with high proficiency (both foreign-born and U.S.-born).

Discussion

In this church-affiliated sample, Latinos’ drug addiction stigma decreases with acculturation, but remains higher among the most acculturated Latinos compared to African-Americans. These attitudes may pose a barrier to treatment for Latino drug users.

Introduction

Migration to the United States appears to have a significant influence on the substance use patterns of Latino immigrants. Longer residence in the United States (U.S.) is consistently associated with higher risk for substance use and substance use disorders across Latino samples encompassing different ages and countries of origin (e.g., Mexicans) (1–5). Indeed, migration appears to be such a powerful predictor of substance use and abuse that it continues to shape migrants’ substance use behavior even after they return to their country of origin, and even among persons who never migrated but have family members that did. For example, several studies using large representative samples of Mexicans living in Mexico demonstrated higher lifetime prevalence of alcohol and other drugs use (e.g., marijuana, cocaine) among return migrants and relatives of migrants compared to those with no migration history, (6–8) even after controlling for sociodemographic differences associated with migration and drug use (e.g., age, gender, marital status, educational attainment). Greater availability of drugs in the U.S. relative to Mexico also seems to account for some, but not all, of the increase in drug use that occurs across generations in the U.S. (6).

Changes in patterns of drug use related to migration may reflect broader changes in attitudes to or moral valuations of drug use, which occur as part of acculturation to American culture and society. Specifically, acculturation involves changes in attitudes, values, and behaviors that result from social and psychological exchanges between groups from different cultures(9). Numerous studies using language-based measures of acculturation demonstrate greater drug use and abuse among the most acculturated Latinos (2, 10–23), suggesting Latino migrants have more opportunities for use and may develop more liberal norms of substance use in the U.S. than were prevalent in their country of origin (8). However, there are few studies on drug use norms or beliefs specifically, including the extent to which Latino immigrants in the U.S. are averse to drug use and to drug users in general.

An important index of cultural change with respect to substance use that has powerful public health implication is the extent of stigma towards substance users, i.e. the extent to which drug users are devalued and viewed within a community as tainted (24). Stigma in the community has public health implications since it marginalizes people with the stigmatizing attribute, creating incentives for them to conceal their stigma-related behavior or condition rather than seeking help for it (25). Stigma might be particularly important for migrant communities because it may exacerbate existing fears of authority and lead to avoiding treatment and more adverse consequences for users. Further, drug abuse stigma has been shown to influence HIV-related attitudes among adults (26, 27), such that stigmatization toward people living with HIV/AIDS (PLWHA) is associated with stigmatization toward vulnerable groups to HIV transmission, including drug users (28). Greater drug abuse stigma has also been shown to be associated with lower levels of drug use among youth (29). Yet we know of no study that has examined drug abuse stigma among Latinos. Part and parcel of the problem is a lack of instruments that assess this construct, with most studies on substance use stigma solely concentrating on alcohol or on drug abuse-related stigma held by health providers working with at-risk populations (30–32). Only two instruments, both of which assessed attitudes toward alcoholics, have been translated into Spanish and validated in Spanish-speaking samples (33, 34).

The aim of this paper is to report on a general drug addiction stigma scale and examine the extent to which proxies for acculturation (e.g., nativity) are associated with drug addiction stigma in a community sample of Latino and African Americans. Specifically, this study uses baseline data from a pilot intervention study addressing HIV-related attitudes and behaviors conducted in collaboration with African American and Latino churches in a high HIV prevalence area of Long Beach in Los Angeles County. Focusing on this church-based sample in Long Beach is important for several reasons. First, the prevalence of illegal drug use during 2010–2012 among persons aged 12 or older is higher in Long Beach (12.05%) compared to LA County (10.17%) and nationwide (8.90%) (35). Second, religion has been identified as having a protective effect against substance use across the lifecourse continuum, given that some religions explicitly prohibit substance use (36) or prescribe behavioral norms that may discourage substance use (37). Third, Latinos and African Americans have high rates of membership in religious congregations (38, 39), therefore churchgoers represent an important segment of these two communities. Lastly, focusing on African Americans and Latinos provides important evidence on whether drug addiction stigma may have a distinct moral valence (i.e., intrinsically aversive) in immigrant vs. U.S.-born groups. Most importantly, the comparison group is people who live in the same community, rather than a group with which Latinos have little direct contact. An issue with research that postulates distinct culturally-based beliefs and norms among different ethnic groups is the comparison group used in the study. Specifically, there is a major assumption in this field—that immigrants are acculturating to white middle class values, norms and behaviors (40, 41) despite the fact that immigrants tend to settle in large, ethnically-diverse, urban cities (42). These features of the settlement context are crucial to our understanding of who migrants see as their relevant peers or models for life in the new culture.

Methods

Survey

Participants completed a survey as part of a study evaluating a new pilot intervention, the Facilitating Awareness to Increase Testing for HIV (FAITH) Project, which is a multi-faceted, congregation-based intervention to reduce HIV-related stigma (29). All of the participating churches were located in and near Long Beach, a city within Los Angeles County. We selected churches purposively to represent diversity in size and denomination: two African American Baptist churches, two Latino Pentecostal churches, and one large Latino Roman Catholic church. Church coordinators and other congregational leaders at 5 churches (3 Latino and 2 African American) helped to promote the survey within each congregation, and English and Spanish language group survey sessions managed by survey administrators were conducted at church sites during regularly scheduled ministry meetings and/or after religious services. Overall participation rate across the 5 churches were 73% at baseline and 79% at follow-up, each calculated as a proportion of regular church attendees at the time of the survey. Participants received a $20 gift card and a meal for completing each survey. The research protocol was approved by the RAND Human Subjects Protection Committee. Further, all study participants were adults (i.e., 18 years and older) and provided informed consent prior to the survey.

Measures

Drug addiction stigma

We modified Ronzani’s alcoholism stigma scale to assess drug addiction stigma (30). Our adapted version is composed of 5 items and explores stereotypes regarding people who use drugs and moral issues popularly associated with addiction (e.g., weakness in character). A 5-point Likert scale was used by participants to indicate level of agreement with each statement. Scoring ranged from 1 to 5, with higher scores signaling greater stigmatizing attitudes towards drug addiction. The adapted scale exhibited good internal consistency among Latinos (Cronbach's α=0.88) and African Americans (Cronbach’s α=0.91).

Race-ethnicity/nativity/language (study groups)

Given that race-ethnicity, nativity, and English language fluency influence acculturation, we used these variables to create 4 study groups: 1) African Americans (reference group); 2) U.S.-born Latinos who reported speaking English “well” or “very well;” 3) foreign-born Latinos who reported speaking English “well or “very well;” and 4) foreign-born Latinos who reported speaking English “not well” or “not at all.”

Other sociodemographic variables were defined as gender (male, female), marital status (single/never married, married/living with partner, divorced/separated/widowed), age (18–20, 31–50, 51+), income ($9,999 or less, $10,000–$29,999, $30,000–$49,999, $50,000–$69,999, $70,000+), education (less than high school, high school, more than high school), and number of household members.

Knowing someone with HIV has been shown to be associated with lower HIV-related stigma (43–45) and may be associated with stigma towards drug addiction. To explore this, we asked respondents whether they know anyone (friends, family, co-workers, others) who “has HIV or AIDS or has died of HIV” (Yes/No).

Data Analysis

Factor analyses

To ensure that the adapted 5-item drug addiction stigma scale measures a single dimension, we conducted a multi-group one-factor confirmatory factor analysis (CFA) using Mplus mean and variance adjusted weighted least squares algorithm (WLSMV) that is appropriate for categorical response items (46). Model fit , that is, the degree to which the model accurately reflects the data, was evaluated with commonly used model fit indices (RMSEA≤.08, TLI≥.95, CFI≥.95) (47). Models are described as closely fitting the data when these criteria are satisfied. Item loadings and thresholds were constrained to equality across the four groups: (1) African Americans (n=419), (2) US born Latinos with high English proficiency (n=205), (3) foreign born Latinos with high English proficiency (n=194), and (4) foreign born Latinos with low English proficiency (n=393). The four groups were modeled as predictors of the drug use stigma latent variable with African American’s coded as the reference group.

Imputation procedures

Rates of missing data for our study variables were relatively low, ranging from 1% to 5% per item. Before conducting multivariate analyses, IVEware 0.2 was utilized to impute missing data by sequential regression multivariate imputation (38). Five imputed datasets were created for the baseline survey and all analyses adjust appropriately using standard rules for aggregating results across multiply imputed data sets (39).

Multivariate analyses

Multivariate linear regression models were fit to examine the independent associations between study group and drug addiction stigma, while controlling for other socio-demographic characteristics and knowing someone with HIV. A p-value of <.05 was considered statistically significant for all analyses.

Results

Table 1 provides an overview of participant characteristics for the entire sample (n=1235) and by study group. There were a number of socio-demographic differences among the groups, most notably: the younger age of U.S.-born Latinos; the higher income and education among African Americans; the much higher proportion of married/living with partner among both foreign-born Latino subgroups; and the larger household size of all Latino subgroups. Further, while almost half of African Americans knew someone who was HIV positive (49.4%), the large majority of Latinos (72.3% to 81.7%) did not know someone who was HIV positive irrespective of place of birth or English proficiency. Finally, the mean drug use stigma score varied across groups, with a range of 2.7–3.7 such that African Americans had the lowest mean levels of drug use stigma and Foreign-born Latinos with low English proficiency had the highest mean.

Table 1.

Participant Characteristics

		All (N=1,235)	African Americans (n=425)	Latino; U.S.-born (n=206)	Latino; Foreign- born and High English proficiency (n=199)	Latino; Foreign- born and Low English proficiency (n=405)
Age (%)	18–30	29.2	26.0	74.9	23.0	11.8
	31–50	45.6	41.7	20.1	60.6	55.2
	51+	25.2	32.3	5.0	15.4	32.9
Gender (%)	Female	63.2	64.9	64.1	52.4	66.1
	Male	36.8	35.1	35.9	47.6	33.9
Race/ethnicity; nativity/English language proficiency (%)	African American	34.4	---	---	---	---
	Latino; U.S.-born	16.8	---	---	---	---
	Latino; Foreign-born and High English proficiency	16.0	---	---	---	---
	Latino; Foreign-born and Low English proficiency	32.9	---	---	---	---
Income (%)	$9,000 or less	22.0	18.1	26.7	16.0	26.8
	$10–$29,000	32.3	20.6	34.9	29.2	44.7
	$30,000–$49,000	22.0	19.3	21.1	30.9	20.9
	$50,000–$69,000	10.6	17.4	6.7	13.3	4.2
	$70,000 or more	13.1	24.6	10.7	10.5	3.4
Education (%)	<High School	32.0	6.8	12.7	33.6	67.5
	High School	28.3	24.6	42.6	31.8	23.0
	>High School	39.7	68.6	44.7	34.6	9.5
Marital Status (%)	Single, never married	31.1	39.9	64.9	19.1	10.3
	Married/living with partner	53.4	37.3	28.3	69.5	73.3
	Divorced/separated/widowed	15.5	22.8	6.8	11.5	14.3
Household size mean (SD)	Number of family members	4.4 (3.2)	3.3 (2.2)	5.0 (2.5)	5.2 (5.4)	4.9 (2.8)
Knows someone who is HIV + (%)	No	68.4	50.6	72.3	75.4	81.7
	Yes	31.6	49.4	27.7	24.6	18.3
Drug abuse stigma score mean (SD)		3.2 (1.3)	2.7 (1.2)	3.3 (1.1)	3.4 (1.2)	3.7 (1.2)

Factor Analysis

All factor loadings for the multi-group 1-factor CFA were greater than 0.80 (see Table 2). Model fit indices suggest that the modified items comprise a unidimensional measure of drug use stigma (χ² = 91, df = 17, CFI=.994, TLI=.992, RMSEA=0.061). Among the subgroups, on average, foreign born Latinos with low English proficiency held more stigmatizing attitudes towards people who use drugs than African Americans with nearly one standard deviation difference. They also held more stigmatizing attitudes towards people who use drugs than their U.S. born counterparts, as well as foreign-born Latinos with high English proficiency.

Table 2.

Drug Abuse Stigma Factor Loadings and Mean Differences Across Groups

	Factor Loadingsa	Group Mean (SE)
Item
Drug addicts are morally weak people	0.93
Drug addicts are people with no will-power	0.90
Drug addiction is a sign of weakness in character	0.83
Drug addicts do not care about their problems	0.83
Drug addicts do not want to quit using drugs	0.82
Group*
US born Latinos with high English proficiency		0.57 (0.10)
Foreign born Latinos with high English proficiency		0.61 (0.10)
Foreign born Latinos with low English proficiency		0.99 (0.08)

^aAfrican Americans were the reference group; group mean differences are in units of standard deviations from the African American group

Multivariate Regression Analysis

After adjusting for age, gender, income, household size, education, marital status, and knowing someone with HIV, Latinos still had significantly higher drug addiction stigma scores compared to African Americans across all subgroups (U.S.-born Latinos, β = 0.22, p<.05; foreign-born Latinos with high English proficiency, β = 0.30, p<.05; and foreign-born Latinos with low English proficiency, β =0.49, p<.001) (see Table 3). Further, we tested for significant differences between the Latino groups in our model and found that foreign-born Latinos with low English proficiency had significantly higher mean levels of drug use stigma compared to their foreign-born and U.S. born high English proficiency counterparts (p <0.05 and 0.01, respectively).

Table 3.

Multivariate associations between participant characteristics and drug abuse stigma scale

		All (N=1,235)
		B (95% CI)‡
Study Group	African American	---
	Latino; U.S.-born	0.22 (0.00, 0.44)*
	Latino; Foreign-born and High English proficiency	0.30 (0.07, 0.53)*
	Latino; Foreign-born and Low English proficiency	0.49 (0.25, 0.72)***
Gender	Female	---
	Male	0.16 (0.05, 0.27)**
Income	$9,000 or less	---
	$10,000–$29,000	−0.10 (−0.25, 0.05)
	$30,000–$49,000	−0.09 (−0.26, 0.07)
	$50,000–$69,000	−0.17 (−0.38, 0.05)
	$70,000 or more	−0.23 (−0.44, −0.03)*
Education	<High School	0.21 (0.04, 0.38)*
	High School	0.15 (0.01, 0.29)*
	>High School	---
Marital Status	Single, never married	0.18 (0.03, 0.33)*
	Married/living with partner	---
	Divorced/separated/widowed	0.16 (0.00, 0.32)*
Household size	Number of family members	−0.02 (−0.08, 0.04)
Knows someone who is HIV +	No	---
	Yes	−0.05 (−0.16, 0.07)

^‡Models control for church-level differences

^*p<0.05;

^**p<0.01;

^***p<0.001

Older adults (i.e., 51+) and men had higher drug use stigma scores, as did those with less education (less than high school, β = 0.21, p<.001 and high school graduates, β = 0.15, p<.001, compared to those with more than high school). Single/never married and divorced, separated, or widowed respondents had higher drug use stigma scores compared to those married or living with a partner (β = 0.18, p<.001 and β = 0.16, p<.001, respectively). Finally, those with annual household incomes of $70,000 or more had significantly lower stigma scores (β = −0.23, p<.001) compared to those with $9,000 or less, and higher numbers of family members in the household was associated with lower stigma scores (β = −0.02, p<.001).

Discussion

Our findings demonstrate that, overall, church-affiliated Latinos have significantly higher rates of stigmatizing attitudes towards drug addiction compared to church-affiliated African Americans. Further, we found that the size of the differences were largest among foreign-born Latinos with low English proficiency, who are less acculturated than their U.S. and foreign born high English proficiency counterparts. Because both African Americans and Latinos were sampled in religious settings, the differences we observed cannot be attributed simply to differences in religious commitment or practice. Religious views about drug use are also likely salient for both groups. Therefore, we interpret the higher stigmatizing attitudes towards drug addiction among Latinos compared to African Americans as evidence that Latinos may hold a distinct set of cultural norms regarding drug addiction. We also interpret the higher stigmatizing attitudes towards drug addiction among immigrant Latinos with low English proficiency compared to their US.-born and foreign-born counterparts to suggest that such cultural norms are likely influenced by one’s social and cultural context and can change when the cultural context changes. For example, drug use in the U.S. has been described as a function of more permissive social norms and attitudes in regard to this behavior (48), and drug use among the general adult population in Latin America is generally lower compared to the U.S.(49) This is also consistent with research documenting higher rates of drug use, abuse and dependence among U.S.-born Latinos compared to their immigrant counterparts (1).

Yet further work is needed to understand the higher stigmatizing attitudes towards drug addiction among foreign-born Latinos with low English proficiency versus those foreign-born with high English proficiency in our sample, and how this difference might be a result of the acculturative process. Scholarship on stigma theory suggests that the element of “moral experience,” or what is most at stake for actors in a local social world, provides a new interpretive lens by which to understand the behaviors of both the stigmatized and stigmatizers (50). If the shift in moral experience occurs quickly, that could explain why we did not find any significant differences between U.S and foreign-born Latinos with high English proficiency. Language acquisition is thought to be one of the first behaviors that change in the process of acculturation (51) and has been shown to be a powerful behavioral marker of acculturation, with many language measures shown to explain most of the variance in acculturation scales (52). We believe this is a useful lens for future research that may help unpack the differences between Latinos and African Americans, as well as U.S.-born and immigrant Latinos.

In general for all of our Latino subgroups, stigmatizing attitudes toward drug addiction might derive not only from greater exposure to distinct drug use patterns and/or different value systems but also from the fact that drug addiction threatens important values across these different groups in qualitatively distinct fashions. Drug addiction may be viewed as individual behavior that requires substantial economic resources and thus may threaten the emotional and economic commitment immigrants have made to their families in their countries of origin – resulting in attitudes that depict drug addiction as particularly odious. Yet another unexplored aspect of the Latino immigrant ethos as it relates to drug addiction stigma is the historical and political particulars that have shaped the drug trade and drug trafficking in Latin America (53, 54). In particular, the escalating violence associated with the rise of the drug trade in many Latin American countries may fundamentally shape more recent immigrant Latinos’ attitudes towards drug addiction. Further, key harm reduction interventions (e.g., needle and syringe exchange and opioid substitution therapy) are not available in most Latin American countries, which would provide people who use drugs with options that help to minimize risks from continuing to use drugs(55). These issues could be explored in future research that measures beliefs about drug addiction and people who use drugs that are held by people in culturally distinct populations, together with additional items that are designed to measure beliefs (e.g., about trafficking and violence) that may be especially salient for specific subpopulations.

Our work also has important public health implications. For example, to the extent that stigmatizing attitudes towards drug addiction influence behavior, higher drug use stigma might be protective by discouraging drug use. However, these same stigmatizing attitudes might be harmful by making addicts less likely to disclose and seek care. For example, several studies have found that even after substance users enter the treatment process, the stigma associated with substance use stays with the user (56, 57). Another public health implication of our finding has to do with the relationship between drug addiction stigma and HIV-related stigma (26). Specifically, those with negative attitudes towards injection drug users held more stigmatizing AIDS attitudes (26, 28). Further, drug injection users affected by HIV cite the stigma of drug injection use as a major reason for delaying medical care (58, 59).

Our study has some limitations that should be noted. Given that all study participants had some church involvement, we cannot draw conclusions about the cultural norms among African Americans and Latinos who have little or no religious involvement. Second, we used a combination of proxy measures (i.e., nativity, language) to approximate the acculturation continuum. This approach has been criticized because it assumes that acculturation is a zero-sum phenomenon (9) (e.g., English language acquisition may not necessarily mean a loss of Spanish language preference). However, as noted above, language acquisition is thought to be one of the first behaviors that change in the process of acculturation (51) and has been shown to be a powerful behavioral marker of acculturation (52). Another strength is that our study also uses African Americans as the referent category instead of Whites, which is important given the scholarship on differences in acculturation typologies and the fact that Latino immigrants are more likely to reside in large-inner cities where Whites may not be the dominant group.

Conclusions

We found drug addiction stigma is higher among immigrant Latinos with low English proficiency compared to their U.S. - and foreign-born counterparts. Moreover, we found that African Americans had significantly lower drug addiction stigma than all three of our Latino subgroups. This suggests that culturally-based attitudes and norms can be protective, but they may also serve as a barrier to treatment. It seems important to continue this line of work among African Americans and Latinos not recruited through churches, especially since the adapted scale had good psychometric properties in this sample. This line of research might also warrant monitoring the consequences of drug-related stigma as well as, potentially, the effectiveness of any efforts to reduce it.