The impact of deportation policies on the substance using behaviors of US-citizen Latinos

Abstract

Background:

Scant data has found that the deportation of a migrant family member has adverse health consequences for their US-citizen family members. However, no study has investigated how the detention or deportations of others (e.g., family, friends) relate to the alcohol and drug-using behaviors of US-citizen Latinos.

Methods:

From April to May 2019, 3,446 adult participants of White, Black, and Latino racial/ethnic descent were recruited to complete a self-administered, structured, online questionnaire. Analyses were restricted to US-citizens only (n=3,282). Using the Alcohol Use Disorders Identification Test-Concise and the Substance Use Brief Screen scale, the two dependent outcome variables were hazardous drinking and symptoms of a drug use disorder (DUD). Three separate multivariable logistic regression models were conducted to explore how Latinos who personally knew a detained or deported migrant, and their relationship to the detainee/deportee (e.g., family, friend), related to each outcome, relative to other racial/ethnic groups.

Results:

Overall 39% of Latinos reported personally knowing a detained or deported migrant. In multivariable analyses, Latinos who knew a detained or deported migrant had significantly higher odds of reporting hazardous drinking and DUD symptoms relative to their White counterparts. Latinos who had a family member detained or deported had 1.23 and 3.91 times the odds of reporting hazardous drinking and symptoms of DUD, respectively, than their White counterparts. Latinos who had a friend detained or deported had also greater odds of reporting DUD symptoms than Whites. Importantly, Latinos who did not personally know a detainee or deportee were not any more likely than Whites to report hazardous drinking and DUD symptoms.

Conclusion:

Findings underscore that detentions and deportations may be influencing the substance using behaviors of US-citizen Latinos and may potentially exacerbate or perpetuate substance abuse-related disparities. Public health and policy implications are discussed.

Introduction

Given the current political climate, the deportation of documented and undocumented migrants from the United States (US) government has received wide media and public attention. However, deportations in the US have long been prevalent. Over a decade of local-, state-, and national-level immigration control and punitive policies have facilitated the mass removal of migrants (Hatzenbuehler et al., 2017; Pinedo, Burgos, & Ojeda, 2014). In 2010, approximately 188,000 migrants were deported from the US (US Department of Homeland Security, 2011). By 2012, deportations reached record numbers, increasing four-fold to approximately 409,000 (Pinedo, Burgos, & Ojeda, 2014). The most at-risk population for being deported has historically been Latino¹ migrants: 90% of all deported migrants are of Latino-origin, with Mexican migrants making up the highest proportion (56%–75%) (US Department of Homeland Security, 2017; Pinedo et al., 2014a). Currently, US immigration and Customs and Enforcement (ICE), which focuses on arresting and apprehending undocumented migrants, has aggressively targeted Latino and migrant communities (Camayd-Freixas, 2009; Lopez et al., 2017). ICE immigration raids have doubled from 2017 alone (Kopan, 2017). In 2019, the largest immigration raid in the history of the US took place in Mississippi when over 680 migrants were detained (Solis & Amy, 2019). Not surprisingly, deportations continue at high rates. Most recently available data indicate that in 2017 over 256,000 migrants were deported, with expectations that deportations will increase in subsequent years (US Department of Homeland Security, 2018).

Increased deportations, coupled with immigration raids and anti-immigration sentiments and policies, have translated into a permanent state of emotional distress for some Latinos (Becerra, Androff, Cimino, Wagaman, & Blanchard, 2013; Hatzenbuehler et al., 2017; Lopez et al., 2017). Such circumstances carry critical health implications for this population. Importantly, US-citizen Latinos are also directly and indirectly affected by deportations. US-citizen Latinos are also vulnerable to being arrested or detained if they are suspected of being undocumented—a process that is rooted in racial profiling (Lopez et al., 2017; Vargas, Sanchez, & Juarez, 2017). US-citizen Latinos are indirectly impacted when their family members are deported, often times being present during their arrests, which can be a traumatic experience (Brabeck, Lykes, & Hunter, 2014; Randolph Capps, 2007; Gulbas et al., 2016; Koball et al., 2015). Scant data with US-citizen Latino children have documented associations between the deportation of a family member and poor mental health outcomes (Allen, Cisneros, & Tellez, 2015; Brabeck et al., 2014; Gulbas et al., 2016; Vargas et al., 2017; Zayas, Aguilar-Gaxiola, Yoon, & Rey, 2015). Other studies have also found negative mental health implications related to immigration raids and the detention and deportation of family members among adult Latinos, regardless of their citizenship status (Lopez et al., 2017; Vargas, Juárez, Sanchez, & Livaudais, 2018). However, how detentions and deportations relate to substance use among US-citizen Latinos has received virtually no attention.

To date, no study has examined if detentions and deportations are influencing the alcohol and drug using behaviors of US-citizen Latinos. Studies that have investigated the relationship between substance use and deportation have exclusively focused on deported migrants in post-deportation settings (e.g., in Mexico and Latin America) (Horyniak, Pinedo, Burgos, & Ojeda, 2017; Ojeda et al., 2011; Pinedo, Burgos, & Ojeda, 2014; Pinedo, Burgos, Robertson, et al., 2014; Pinedo et al., 2018). Due to the emotional and psychological consequences of detentions and deportations, we can broadly speculate that US-citizen Latinos who have been impacted by the detention or deportation of others in their social networks (e.g., family, friends) may be more susceptible to use alcohol and drugs as a coping mechanism. If this is true, US-citizen Latinos who know a detainee or deportee may be at increased risk to developing a substance use disorder. However, this claim remains to be empirically assessed, representing a critical gap in the alcohol and drug use literature. Further, Latinos are disproportionally impacted by alcohol and drug-related problems (Chartier & Caetano, 2010; Guerrero, Marsh, Khachikian, Amaro, & Vega, 2013; Mulia, Ye, Greenfield, & Zemore, 2009; Pinedo, Zemore, Cherpitel, & Caetano, 2017). Thus, the emotional and psychological consequences of detentions and deportations have the potential to exacerbate or perpetuate existing substance use-related disparities. Due to the continued high rates of detentions and deportations and their potential impact on the substance-using behaviors of Latinos, makes this line of research is timely and significant.

We undertook the present study to explore the relationship between hazardous drinking, symptoms of a drug use disorder (DUD), and personally knowing a detained or deported migrant using a national dataset of US-citizens. We hypothesize that Latinos who personally know a detainee or deportee will report greater odds of engaging in hazardous drinking and DUD symptoms than Whites, Blacks, and Latinos who do not personally know a detained or deported migrant. We further hypothesize that Latinos who have had a family member detained or deported will be more likely to report hazardous drinking and DUD symptoms relative to Whites, Blacks, and Latinos who do not personally know a detained or deported migrant. To guide our study, we applied migration as a social determinant of health as our analytic lens (Castañeda et al., 2015). This perspective posits that health outcomes and behaviors are products of larger social and structural forces that are directly associated with the migrant experience (e.g., anti-immigration policies, anti-immigration sentiments, deportations). We expand on this focus by underscoring that US-citizen Latinos, who were born and raised in the US, and therefore not migrants by definition, are also impacted by migration-related factors. Under this framework, deportations can be characterized as a form of forced migration and distinct from voluntary migration. The deportation of others (e.g., family, friends) is characterized as a structural-level determinant. Stressors induced by the fear of a potential or actual deportation is a critical component of the migrant experience, which also extend to their families. Thus, this framework can be used to better understand the substance-using behaviors of US-citizen Latinos within the current social and political climate.

Methods

Study design and population

The present study uses data from the National Social Policy Survey conducted by the University of New Mexico (UNM) Center for Social Policy. The overall objective of this study was to better understand how social, political (including immigration policies), and economic contexts impact the physical and mental health of Americans. Data collection efforts were overseen by Pacific Market Research, a leading survey research firm located in Renton, Washington. Briefly, from April to May 2019, a total of 3,446 adult participants were recruited to complete a self-administered, structured, anonymous online questionnaire. Eligible participants were adults of White, Black, and Latino racial/ethnic descent—the three largest racial/ethnic groups in the US. Potential participants were recruited via two sampling methods. First, emails were obtained from the national voter registration database and potential participants were randomly selected and invited to participate. To reduce bias, a sample of non-registered voters (including non-US-citizens) was also recruited. Participants were randomly sampled using online web-panels from Pure Spectrum and Cint. Web-panels are an emerging and innovative survey method that is cost-sensitive and ideal for reaching a larger number of people, relative to telephone surveys. Web-panels are made up of persons that have already agreed to take part in online surveys, which ensures a high-response rate. Panels are constructed through a combination of targeted invitations to consumers and open recruitment techniques, using a national frame, that have resulted in panels that are representative of the general population and including hard-to-reach segments of the population. This sample of potential participants were asked to confirm that they were not registered to vote before starting the survey. A total of 6,355 potential participants were invited to complete the online questionnaire, 5,734 were eligible, and 3,446 completed the survey. The response rate for the voter registration database sample, the Pure Spectrum web-panel, and the Cint web-panel, was 65%, 82%, and 87%, respectively.

Before starting the survey, participants provided informed consent online. The survey collected data on a diversity of topics including socio-demographics, health related topics (e.g., alcohol use, drug use, mental health), and migration and deportation-related questions, among other topics. The survey was self-administered online, anonymous, contained 146 questions, took on average 24 minutes to complete, and was available in English and Spanish (199 participants took the survey in Spanish). The final dataset was weighted within each racial/ethnic group to match the adult population in the 2017 Census ACS 1-year data file for age, gender, education, nativity, ancestry, and voter registration status. A post-stratification raking algorithm was used to balance each category within +/− 1 percent of the ACS estimates. The study protocol was approved by the Institutional Review Board of the University of New Mexico (UNM). The present analysis is based on de-identified data and therefore the Institutional Review Board of UT Austin required no review or oversight.

Measures

Hazardous drinking and symptoms of DUD.

Our two primary outcome variables were hazardous drinking and symptoms of DUD. To assess hazardous drinking we used the Alcohol Use Disorders Identification Test-Concise (AUDIT-C). The AUDIT-C is a 3-item measure that has high sensitivity for identifying hazardous drinking that is indicative of an alcohol use disorder (AUD) (Frank et al., 2008). Participants were asked how often they drank alcohol: never (0 points), monthly or less, 2–4 times a week (1 point), 2–3 times a week (2 points), or 4 or more times a week (4 points). Next, participants were asked how many alcoholic drinks they consumed on a typical day. Response options included: 1 or 2 drinks (0 points), 3 to 4 drinks (1 point), 5 or 6 drinks (2 points), 7 to 9 drinks (3 points), or 10 or more drinks (4 points). Lastly, participants were asked how often they consume 6 or more drinks on one occasion. Response options included: never (0 points), monthly or less, 2–4 times a week (1 point), 2–3 times a week (2 points), or 4 or more times a week (4 points). Scores were summed with a total possible range of 0 to 12 points. Using the recommended screening thresholds (Frank et al., 2008), male participants who scored 4 or more points and female participants who scored 3 or more points were characterized as engaging in hazardous drinking. This level of drinking poses significantly greater health and social risks. Participants who answered ‘never’ to the first question were not asked the subsequent two questions and were categorized as not being hazardous drinkers.

To screen for symptoms of DUD, we used the illicit and prescription drug use section (two items) of the Substance Use Brief Screen (SUBS) scale. These two items have shown high sensitivity (86%) and specificity (82%) for identifying DUD (McNeely et al., 2015). Participants were asked on how many days in the past 12 months they used: (1) any illegal drugs, including marijuana; and (2) any prescription medications ‘recreationally’ (just for the feeling, or using more than prescribed). Response options included: 3 or more days, 1–2 days, or Never. Per the recommended guidelines (McNeely et al., 2015), participants who answered never to both were characterized as not having symptoms of DUD while those who responded positively to either question were coded as reporting DUD symptoms.

Personally knowing someone who has been detained or deported.

Our primary variables of interest centered on knowing (vs. not knowing) a detained or deported migrant and their relationship (e.g., family, friend) to the participant. Latinos were asked, “Do you personally know someone that has been detained or deported for immigration reasons?” Those who answered affirmatively were characterized as knowing a detainee or deportee. We recognize that being detained and being deported are two different experiences. However, one study tested how knowing a detained migrant versus knowing a deported migrant influenced the mental health of Latinos and found no differences (Vargas et al., 2018). This finding suggests that all phases of the deportation process may have the same emotional consequences. Further, Latinos who reported personally knowing a detainee or deportee, were asked if the detainee or deportee was a: (1) family member, (2) a friend, or (3) someone from work or the community. Using this variable, we characterized Latinos into four mutually exclusive categories, those who (1) do not personally know a detainee or deportee; (2) know a family member that was detained or deported; (3) know a friend that was detained or deported; and (4) know a co-worker or community member that was detained or deported.

Covariates.

Covariates included socio-demographic characteristics. To assess race/ethnicity, participants were first asked to self-report if they were non-Hispanic White, Black or African American, or Hispanic or Latino. Participants were also asked if they considered any part of their race or ethnicity to be African American or Hispanic/Latino (two separate questions). Using these data, participants were categorized as White (i.e., non-Hispanic), Black, or Latino. Following, participants were asked to self-report their biological sex (male vs. female), age, if they were born in the US (yes vs. no), highest educational attainment (less than high school, graduated high school, some college, or graduated college), employment status (employed full/part-time vs. unemployed), and total annual household income (less than $20,000, $20,000 to $39,000, $40,000 to $59,000, $60,000 to $79,000, $80,000 to $99,000, $100,000 to $140,000, or $150,00 or more). We also controlled for state of residence, as research has linked states with more anti-immigration policies with poorer health outcomes among Latinos (Hatzenbuehler et al., 2017; Vargas et al., 2017). Participants self-reported what state they were currently living in.

Psychological distress was also included in the analysis because it has been strongly associated with substance use (Löwe et al., 2010). Psychological distress was measured via the Patient Health Questionnaire (PHQ-4). This is a validated 4-item measure of anxiety and depression. Participants were asked how often they had experienced the following problems over the last two weeks: (1) feeling nervous, anxious, or on edge; (2) not being able to stop or control worrying; (3) little interest or pleasure in doing things; and (4) feeling down, depressed, or hopeless. Response options include: Not at all (0 points), several days (1 point), more than half the days (2 points), or nearly every day (3). Scores were summed, ranging from 0 to 12 points. Participants who scored 3 or more points were coded as having psychological distress.

Analyses

All analyses were weighted using STATA v.15 software. Given the objective of our study, we restricted analyses to US-citizens (n=3,282; 708 Whites, 706 Blacks, 1,868 Latinos). We first compared participants by socio-demographic characteristics and psychological distress stratifying by race/ethnicity and then among Latinos only stratifying by personally knowing (vs. not knowing) a detainee or deportee. Following, we compared frequencies of hazardous drinking and symptoms of DUD by race/ethnicity, among Latinos who personally know (vs. do not know) a detainee or deportee, and by Latinos’ relationship to the detainee or deportee. We tested associations among variables using chi-square tests. Lastly, we conducted three separate multivariable logistic regression models for each outcome (i.e., hazardous drinking and symptoms of DUD) controlling for socio-demographic characteristics and psychological distress. The first model examined the relationship between race/ethnicity on hazardous drinking and symptoms of DUD. Model 2 expanded on Model 1 by separating Latinos into two groups, those who: (1) personally know a detainee or deportee and (2) do not personally know a detainee or deportee. This separation allowed us to evaluate the relationship between knowing and not knowing a detainee or deportee on hazardous drinking and DUD symptoms, relative to other racial/ethnic groups. Model 3 further expanded on Model 2 by separating Latinos according to their relationship to the detainee or deportee: (1) a family member, (2) a friend, or a (3) co-worker or community member. This model allowed us to examine how Latinos’ relationships to detainee or deportee relates to hazardous drinking and symptoms of DUD and in relation to other racial/ethnic groups.

Results

Sample characteristics of participants by race/ethnicity

Table 1 displays weighted sample characteristics. About half the sample was male across all racial/ethnic groups. The majority of participants in our sample were US-born. Latinos were more likely to report having less than a high school education and not having graduated from college than their Black and White counterparts. On average, Latino and Black participants were younger, less likely to be currently employed, and reported less household income than White participants. A higher proportion of Latinos reported psychological distress relative to White and Black participants.

Table 1.

Selected characteristics of US-citizen participants by race/ethnicity, (weighted %, unweighted n), unweighted n=3,282, 2019.

Variable	% Whites (n=708)	% Blacks (n=706)	% Latinos (n=1,868)	P-value
Socio-demographics
Male	49% (405)	47% (268)	50% (695)	0.452
Mean age (SD)	48 (16.74)	41 (16.32)	39 (15.84)	≤0.001
US-born	98% (700)	95% (691)	80% (1,716)	≤0.001
Married	47% (333)	26% (186)	40% (700)	≤0.001
Employed	73% (520)	54% (378)	58% (1,058)	≤0.001
Educational attainment				≤0.001
Less than high school	3% (17)	8% (39)	12% (144)
Graduated high school	27% (193)	25% (179)	29% (572)
Some college	27% (195)	36% (256)	31% (616)
Graduated college	42% (303)	32% (232)	29% (536)
Total Family Income				≤0.001
Less than $20,000	13% (86)	17% (116)	13% (240)
$20,000 - $39,000	23% (155)	28% (196)	23% (462)
$40,000 - $59,000	19% (140)	22% (154)	23% (420)
$60,000 - $79,000	20% (144)	14% (103)	15% (273)
$80,000 - $99,000	9% (62)	5% (38)	9% (166)
$100,000 - $140,000	10% (77)	7% (50)	8% (137)
$150,00 or more	5% (35)	5% (32)	4% (74)
Mental health status
Psychological distress	46% (321)	45% (320)	51% (1,045)	≤0.001

Sample characteristics of Latinos participants

Overall, 39% (n=741) of US-citizen Latinos in our sample reported personally knowing someone who has been detained or deported (Table 2). The socio-demographic characteristics of Latinos in our sample were similar regardless of whether they reported knowing or not knowing a detainee or deportee. However, Latinos in our sample who knew someone who was detained or deported were more likely to be younger (36 years vs. 41 years; p≤0.0001), US-born (84% vs. 78%; p=0.04) and currently employed (62% vs. 56%; p=0.02). Latinos who know a detained or deported migrant (vs. those who do not) were also more likely to report psychological distress (65% vs. 43%; p≤0.001). Among Latinos who personally know a detainee or deportee, 26% had a family member, 43% had a friend, and 31% had a co-worker or community member detained or deported.

Table 2.

Selected characteristics of US-citizen Latino participants by those who personally know a detainee or deportee (weighted %, unweighted n), unweighted n=3,282, 2019.

	Latinos who personally know a detainee or deportee
	No 61% (n=1,127)	Yes 39% (n=741)	P-value
Socio-demographics
Male	51% (420)	48% (275)	0.30
Mean age (SD)	41(16.37)	36(15.12)	≤0.001
US-born	78% (1,029)	84% (687)	0.03
Married	42% (438)	38% (262)	0.13
Employed	58% (607)	62% (451)	0.01
Educational attainment			0.54
Less than high school	12% (88)	11% (56)
Graduated high school	30% (359)	28% (213)
Some college	30% (370)	31% (256)
Graduate college	28% (310)	31% (226)
Total Family Income			0.07
Less than $20,000	15% (165)	9% (75)
$20,000 - $39,000	23% (266)	25% (196)
$40,000 - $59,000	22% (245)	24% (175)
$60,000 - $79,000	14% (165)	15% (108)
$80,000 - $99,000	10% (102)	8% (64)
$100,000 - $140,000	8% (78)	9% (59)
$150,00 or more	3% (43)	4% (31)
Mental health status
Psychological distress	43% (540)	65% (505)	≤0.001
Deportation variablesa
Relationship to deportee			≤0.001
Family member	0% (n=0)	27% (n=197)
Friend	0% (n=0)	43% (n=318)
Co-Worker or Community member	0% (n=0)	31% (n=226)

^aAmong Latinos who personally know a deportee n=741

Prevalence of hazardous drinking and symptoms of a drug use disorder by race/ethnicity and knowing a detainee or deportee

Table 3 reports the prevalence of hazardous drinking and symptoms of DUD by race/ethnicity and among Latinos who personally know a detainee or deportee and their relationship to the detainee or deportee. We found that the prevalence of hazardous drinking was the same for all racial/ethnic groups: 48% of Whites, Blacks, and Latinos in our sample engaged in hazardous drinking. However, when we characterized Latinos into two groups, those who know a detainee or deportee and those who do not, differences emerged. Among Latinos, those who personally know (vs. those who do not know) a detained or deported migrant had a significantly higher prevalence of hazardous drinking (59% vs. 42%; p≤0.0001). This prevalence was also higher than White and Black participants. Latinos who had a friend detained or deported had the highest prevalence of hazardous drinking (66%), followed by those who had a family member detained or deported (59%) and those who had a co-worker or community member (49%) detained or deported (p=0.002).

Table 3.

Prevalence of hazardous drinking and symptoms of a drug use disorder among US-citizen participants by race/ethnicity & status of knowing a detainee or deportee, (weighted %, unweighted n), unweighted n=3,282, 2019.

Variable	Hazardous drinking % (n=1,618)	P-value	Drug use disorder symptoms %(n=997)	P-value
Race/ethnicity		0.929		≤ 0.001
White	48% (345)		23% (166)
Black	48% (339)		32% (222)
Latino	48% (934)		31% (609)
Latinos by knowing a detainee or deporteea		≤ 0.001		≤ 0.001
Does not know a detainee or deportee	42% (486)		21% (237)
Know a detainee or deportee	59% (448)		48% (372)
Latinos’ relationship to the detainee or deporteeb		≤ 0.002		≤ 0.001
Family member was detained or deported	59% (117)		61% (117)
Friend was detained or deported	66% (211)		53% (175)
Co-worker of community member was detained or deported	49% (120)		32% (80)

^aAmong US-citizen Latinos, n=1,868;

^bAmong US-citizen Latinos who reported personally knowing a detainee or deportee, n=741

In terms of drug use (Table 3), a higher proportion of Latinos and Blacks reported symptoms of DUD relative to Whites (31% and 32% vs. 23%, respectively, p≤0.001). Among Latinos, those who personally know (vs. those who do not know) a detainee or deportee were more likely report symptoms of DUD (48% vs. 21%; p≤0.001). Latinos who know a detainee or deportee had a significantly higher prevalence of DUD symptoms than all racial/ethnic groups, with those who have had a family member detained or deported reporting the highest prevalence (61%).

Multivariable logistics regression models examining hazardous drinking and symptoms of a drug use disorder by race/ethnicity and knowing a detainee or deportee

Results from the multivariable logistics regression models showed that when controlling for socio-demographic characteristics and psychological distress, race/ethnicity was unrelated to hazardous drinking (Table 4; Model 1). However, Model 2 revealed that Latinos who personally know a detainee or deportee had higher odds of reporting hazardous drinking (Adjusted Odds Ratio (AOR): 1.37; 95% CI: 1.06–1.77) than their White counterparts. Associations between hazardous drinking and being Black and not knowing a detainee or deportee were non-significant. Further, Model 3 explored the association between Latinos’ relationships with the detainee or deportee and hazardous drinking. Results suggest that Latinos who had friend detained or deported had double the odds (AOR: 1.95; 95% CI: 1.41–2.69) of reporting hazardous drinking than Whites.

Table 4.

Multivariable logistics regression models examining hazardous drinking and symptoms of a drug use disorder by race/ethnicity and knowing a detainee or deportee among US-citizen participants, unweighted n=3,276, 2019.

	Hazardous drinking		Drug use disorder symptoms
Variable	AORa	95% CIb	AORa	95% CIb
Model 1c
Whites (Ref)	-	-	-	-
Blacks	1.07	0.84–1.37	1.56	1.18–2.05
Latinos	0.99	0.80–1.23	1.42	1.10–1.83
Model 2c
Whites (Ref)	-	-	-	-
Blacks	1.07	0.84–1.36	1.56	1.18–2.05
Latinos who do not personally know a deportee	0.81	0.64–1.02	0.85	0.64–1.14
Latinos who personally know a deportee	1.37	1.06–1.77	2.57	1.93–3.41
Model 3c
Whites (Ref)	-	-	-	-
Blacks	1.07	0.84–1.37	1.56	1.18–2.05
Latinos who do not personally know a deported migrant	0.81	0.64–1.02	0.85	0.64–1.14
Latinos, family member was deported	1.23	0.84–1.79	3.91	2.59–5.90
Latinos, friend was deported	1.95	1.41–2.69	3.10	2.20–4.38
Latinos, co-worker or community member was deported	0.96	0.66–1.37	1.38	0.93–2.05

^aAdjusted Odds Ratio;

^bConfidence Interval;

^cControlling for gender, age, marital status, US-born, education, employment status, annual household income, and psychological distress, state of residence

Within the context of illicit drug use, Blacks and Latinos had higher odds of reporting DUD symptoms than Whites (Table 4; Model 1). However, Model 2 suggested that the relationship between DUD symptoms and being Latino was only significant for Latinos who personally know someone who was detained or deported. Latinos who personally know a detainee or deportee had 2.57 times the odds of reporting DUD symptoms than their White counterparts. Finally, Model 3 results suggest that Latinos who had a family member detained or deported have almost four times the independent odds (AOR: 3.91; 95% CI: 2.59–5.90) of reporting DUD symptoms, relative to their White counterparts. Latinos who had a friend (AOR: 3.10; 95% CI: 2.20–4.38) also had significantly higher odds of reporting DUD symptoms than Whites.

Discussion

This is the first study to examine the relationship between knowing a detainee or deportee and alcohol and drug using behaviors among a national sample of US-citizen Latino participants. Overall, 48% of participants engaged in hazardous drinking and 23–31% reported DUD symptoms. These statistics are aligned with national trends. For example, data from the National Survey on Drug Use and Health (NSDUH) suggest that 47% of alcohol users and 32% of adults engaged in hazardous drinking and any illicit drug use, respectively, in the past month (NSDUH, 2017). However, we found that Latinos who have close connections with detained or deported migrants may be potentially high-risk group for substance use disorders. These are novel findings that have not been previously documented in the alcohol and drug use literature. Results add to the limited evidence base suggesting that the health of US-citizen Latinos is also being impacted by deportations and the need to consider migration as a social determinant of health among US-born and US-citizen Latinos (Gulbas et al., 2016; Rojas-Flores, Clements, Hwang Koo, & London, 2017; Vargas et al., 2018; Zayas et al., 2015). Migration-related factors (e.g., consequences of detentions and deportations, immigration raids, anti-immigration sentiments) have largely been ignored by national datasets examining the health of Latinos and studies have rarely considered migration-related factors on the health of US-citizen Latinos—a blind spot in the health disparities and substance use field. Our findings underscore that detentions and deportations may be shaping the substance-using behaviors of US-citizen Latinos, which can potentially perpetuate or exacerbate existing health disparities related to alcohol and drug use among Latinos.

Aligned with our hypothesis, we found that US-citizen Latinos who personally know a detainee or deportee were more likely to report hazardous drinking and DUD symptoms than Whites. This relationship varied depending on Latinos’ relationship with detainees or deportees. Within the context of alcohol, those who have a friend detained or deported had a higher likelihood of reporting hazardous drinking than Whites. Surprisingly, associations between having a family member detained or deported and AUD were null—contrary to our second hypothesis. This finding may be explained by the fact that a higher proportion of Latinos in our sample were more likely to report knowing a friend that was detained or deported than a family member. In terms of drug use, all Latinos who know a detainee or deportee were more likely to report DUD symptoms, regardless of the relationship to the detainee or deportee. Interestingly, a dose-response-like relationship emerged in terms of Latinos’ connectedness to detainees/deportees and DUD symptoms. Latinos who had a family member detained or deported had the highest odds of reporting DUD symptoms, followed by those who had a friend detained or deported, and then those who know a co-worker or community member who was detained or deported. Alarmingly, Latinos with detained or deported family members had four times the independent odds of reporting DUD symptoms relative to Whites. Lastly, it is noteworthy to highlight that Latinos who do not know a detainee or deportee were not any more likely than Whites to report hazardous drinking and DUD symptoms. This finding suggests that US-citizen Latinos who have been impacted by the detention or deportations of others in their social networks may be at increased risk for substance misuse.

To contextualize our findings, detentions and deportations are traumatizing experiences that cause extreme emotional and psychological distress at the individual and community-level. At the individual level, deportations contribute to family separation, social disruption, post-traumatic stress, financial distress (namely when the deported migrant is the breadwinner), severed social support ties, and other emotional stressors (Dreby, 2012; Gulbas et al., 2016; Lopez et al., 2017; Rojas-Flores et al., 2017; Zayas et al., 2015). At the community-level, immigration raids primarily target Latino communities and are conducted by heavily armed officials who forcefully enter peoples’ homes or work places and detain everyone present, regardless of whether the person is the target (Golash-Boza, 2015; Hernández, Lowery, & Hauslohner, 2017; Lopez et al., 2017). These raids are violent, chaotic, and have long lasting emotional effects well after the raid has been conducted. Raids elicit significant fear, intense feelings of being persecuted, and migration-related stress (e.g., fears of the consequences of deportation) among communities—which also extend to US-citizen Latinos (Randy Capps et al., 2015; Golash-Boza, 2015). These individual- and community-level stressors may particularly be intensified among Latinos who personally know someone who has been detained or deported, as they have been directly impacted. The detention and deportation of others may foster circumstances that ultimately increase susceptibility to substance use, particularly as a coping mechanism. More research on how detentions and deportations may be influencing the substance using behaviors of Latinos is warranted.

Some limitations should be considered when interpreting our findings. Participants were recruited online and may be biased towards individuals with more resources (e.g., have a computer) and computer or technology literate. However, web and mobile devices (e.g., smartphones) have become increasingly common among the general population. Due to the sensitive nature surrounding detentions, deportations, and substance use, participants may have been likely to underreport their experiences. However, surveys were completely anonymous and self-administered, thereby reducing fear and potential social desirability bias. Response bias may also be present as participants who were willing to participate may differ from those who did not; no incentive was provided. The cross-sectional study design hinders our ability to make causal inferences between knowing a detainee/deportee and hazardous drinking and DUD symptoms. Future studies should consider a longitudinal design to determine if Latinos are more likely to use, or increase, their substance use after someone they know has been detained or deported. Lastly, the survey did not collect contextual data regarding the detention or deportation of others (e.g., why was the migrant initially detained, when did the detention/deportation occur, which family member was detained). Such data would have assisted in our interpretation of the findings. Despite these limitations, this well-powered dataset allowed us to conduct an in-depth and comparative examination of substance use across race/ethnicity within the context of the current political climate surrounding detentions and deportations, a topic that is significantly understudied.

Conclusion

Overall, our findings raise concerns regarding the future health of Latinos in the US. US-citizen Latinos with migrant family members, friends, and community members represent a significant proportion of the Latino population. More than 5.3 million US-citizen Latino children have at least one family member that is an undocumented migrant (Randy Capps et al., 2015). This figure is likely drastically higher given that this estimate does not account for adult US-citizens, documented migrants (who are also vulnerable to being deported), and other relationships (e.g., non-parental family members, friends, co-workers). US-citizen Latinos are closely intertwined with the migrant population. As policy debates surrounding deportations and immigration in the US, more research on the health effects of detentions and deportations on US-citizen Latinos is critical. Policies should consider the emotional and health impacts of detentions and deportations on the health of US-citizen Latinos—a significant and growing segment of the US population. Prevention strategies should also consider tailored interventions aimed as addressing specific stressor stemming from the detention and deportation of others that may increase vulnerability to substance use.