Alcohol use trajectories of adult Latinx immigrants during their first decade in the United States

Abstract

Background:

Despite important research advances on the alcohol use behaviors of Latinx immigrants, the bulk of existing evidence has focused on alcohol use patterns after immigrant’s arrival to the US. The present study examines pre- to post immigration alcohol use trajectories among Latinx immigrants of diverse national origins throughout their first decade in the United States.

Methods:

Data for this longitudinal study was collected between 2007 and 2017. At baseline, retrospective pre-immigration alcohol use data was collected on 527 Cuban, South American and Central American participants ages 18–34 who immigrated to the US less than one year prior. Three subsequent follow-up assessments over the past 10 years have collected data on post-immigration alcohol use.

Results:

Overall decreases in pre- to post-immigration alcohol use were found. Gender, income, and national origin were associated with distinct drinking patterns after immigration.

Conclusion:

Findings underscore the importance of examining pre-immigration experiences among Latinx immigrants, in addition to longitudinal alcohol use patterns after immigration. This knowledge is important not only from an epidemiological perspective, but for the development of effective culturally relevant alcohol interventions targeting Latinx immigrants early in the immigration process.

1. Introduction

Prior research indicates that Latinx immigrants are less likely to engage in alcohol misuse compared to their US-born counterparts. Substantial evidence has also revealed an escalation of alcohol use and misuse among Latinx immigrants as their time in the United States (US) increases (Castañeda et al., 2019; Lui & Zamboanga, 2018). It has been widely posited that the lower rates of alcohol misuse among foreignborn Latinx relates in part to the healthy immigrant effect which states that those who migrate (both in their homes and receiving country) are more likely to be psychologically and physically healthy and more motivated to improve their quality of life (Flores & Brotanek, 2005; Rubalcava, Teruel, Thomas, & Goldman, 2008). The escalation of alcohol use among Latinx immigrants as their time in the US increases has been linked to the acculturation process. Broadly, acculturation is the process of adjusting to a cultural setting distinct from one’s place or culture of origin, including changes in cultural practices, values, and identifications (Schwartz, Unger, Zamboanga, & Szapocznik, 2010). Namely, greater adherence to US practices and beliefs that tend to have more permissive of alcohol use norms may lead to the adoption alcohol use behaviors and exposure to social groups where alcohol misuse is accepted (Salas-Wright, Clark, Vaughn, & Córdova, 2015; Salas-Wright et al., 2018) (Table 1).

Table 1.

Results of GEE models and mixed effects zero-inflated negative binomial (ZINB) model for alcohol use from baseline (T1) to third follow-up (T4).

Logistic (Zero-Inflation) portion of ZINB model
Independent Variables		Alcohol Use Frequency aIRRa [95% CI]	Alcohol Use Quantity Estimateb [95% CI]	Binge Drinking Days Estimatec [95% CI]
Time:
T2 vs. T1				1.41 [0.96, 2.07]
T3 vs. T1				1.66 [1.16, 2.38]**
T4 vs. T1				3.33 [2.04, 5.45]***
Gender:       Male vs. Female				0.37 [0.27, 0.50]***
Age at baseline				1.02 [0.99, 1.05]
Household income (in thousands of US$)				1.00 [0.99, 1.01]
Country of origin:
Central America vs. Cuba				0.80 [0.56, 1.14]
South America vs. Cuba				0.42 [0.29, 0.61]***
GEE models, Negative-Binomial portion of ZINB modeld
Independent Variables		Alcohol Use Frequency aIRRa [95% CI]	Alcohol Use Quantity Estimateb [95% CI]	Binge Drinking Days Estimatec [95% CI]
Time:
T2 vs. T1		0.70 [0.53, 0.91]**	0.77 [0.62, 0.95]*	0.66 [0.54, 0.82]***
T3 vs. T1		0.94 [0.71, 1.24]	0.75 [0.62, 0.92]**	0.81 [0.67, 0.99]*
T4 vs. T1		0.66 [0.49, 0.89]**	0.77 [0.57, 1.05]	0.76 [0.52, 1.12]
Gender:      Male vs. Female		1.86 [1.40, 2.47]***	2.79 [2.20, 3.55]***	2.10 [1.57, 2.81]***
Time*Gender:
T2*Male		1.27 [0.92, 1.75]	0.86 [0.63, 1.17]
T3*Male		0.82 [0.59, 1.15]	0.76 [0.59, 0.99]*
T4*Male		1.51 [1.08, 2.11]*	0.59 [0.43, 0.82]**
Age at baseline		0.99 [0.97, 1.01]	1.00 [0.98, 1.02]	0.99 [0.96, 1.01]
Household income (in thousands of US$)		1.01 [1.00, 1.01]***	1.00 [1.00, 1.00]	1.00 [0.99, 1.00]
Country of origin:
Central America vs. Cuba		0.86 [0.70, 1.06]	1.41 [1.08, 1.86]*	0.91 [0.67, 1.23]
South America vs. Cuba		1.70 [1.36, 2.11]***	1.52 [1.22, 1.90]***	1.59 [1.16, 2.18] **

Notes:

^a.Estimates are adjusted incident risk ratios (aIRRs) from GEE model with negative binomial distribution.

^b.Estimates are exponential correlation coefficients from GEE model with log link.

^c.Estimates for the Zero-Inflated portion of ZINB are adjusted odds ratios (aORs); while estimates for the Negative-Binomial portion of ZINB are aIRRs.

^d.Estimates for predicting alcohol use frequency and quantity were from GEE models. Estimates for predicting binge drinking days were from ZINB model with two components: Logistic (zero-inflated) and Negative-Binomial portions. Because GEE models do not include a zero-inflated portion, results for the zero-inflated portion were reported for binge drinking days only.

^*p < 0.05;

^**p < 0.01;

^***p < 0.001.

Yet, little is known about how alcohol use behaviors change from pre- to post-immigration among Latinx immigrants. The bulk of existing evidence has focused on alcohol use patterns of Latinx immigrants after arrival to the US (Caetano, Mills, Vaeth, & Reingle, 2014; Salas-Wright et al., 2018). These studies have also concentrated on drinking patterns of Latinx immigrants (predominantly of Mexican origin) living in the US for extended periods of time (Reingle, Caetano, Mills, & Vaeth, 2014). The pre- to post-immigration alcohol use trajectories among Latinx immigrants over a substantial time period (e.g., 10 years) remains relatively unknown.

Previous studies conducted by this investigative team suggest decreases in post-immigration alcohol use quantity and frequency among Latinx immigrants during their initial years in the US (Sanchez et al., 2015). Gender specific-effects also revealed that compared to women, men report greater alcohol use quantity and frequency prior to immigration, along with steeper declines in drinking behaviors after immigration. Researchers have posited that decreases in alcohol use during the initial years in the US could be explained by deterrence theory; specifically, immigrants willing to go through the stressors of immigration and start a new life in the US may be deterred from behaviors such as alcohol misuse that could place them at greater risk for legal consequences, and in some cases potential deportation (Sanchez et al., 2015; Salas-Wright et al., 2018). The present study expands the current literature to examine alcohol use trajectories in a cohort of Latinx immigrants from pre- to post-immigration during their first decade in the US. Based on acculturation theory, we anticipate that increased time in the US will be positively associated with greater alcohol use overtime, and these trajectories will vary across demographic factors (i.e, gender, national origin).

From a public health perspective, identifying and understanding the link between alcohol use and length of time living in US is significant given that excessive alcohol use/abuse among US-born and adult Latinx immigrants is a pressing public health concern (NIAAA, 2019). Compared to non-Latinx Whites, the US Latinx population experiences disproportionate rates of at-risk drinking. For instance, Latino men also have the highest rates of age-adjusted alcohol-related mortality (41.6 per 100,000) compared to non-Latino white men (34.8), non-Latino white women (10.8) and Latina women (6.7) (Mejia de Grubb, Salemi, Gonzalez, Zoorob, & Levine, 2016). White Latino men also have the highest rates of alcohol-related cirrhosis of all ethnic/racial groups with age adjusted death rate from cirrhosis in this group being almost twice that of White and Black men (NIAAA, 2019; Stinson, Grant, & Dufour, 2001; Yoon & Chen, 2016). As such, examining how patterns in alcohol use quantity, frequency, and binge drinking change from the pre- to post-immigration context is critical for informing prevention efforts early in the immigration process.

While cultural differences between distinct Latinx national groups are well-known, previous research on alcohol use among Latinx immigrants has primarily focused on Mexican immigrant populations (Borges et al., 2011; Worby and Organista, 2013). Far less is known about the alcohol use trajectories of Cuban and South and Central American immigrants (Sanchez et al., 2015). Existing studies suggest higher rates of alcohol use among South and Central American Latina women compared to their Mexican counterparts. Conversely, Mexican men report higher levels of alcohol use and dependence than their South American, Central American and Caribbean counterparts (NIAAA, 2019; Vaeth, Caetano, & Rodriguez, 2012). While Mexicans remain the largest US Latinx immigrant group, shifts in immigration patterns within the past decade have indicated steep increases in immigrants from Central and South America arriving in the US (Noe-Bustamante, 2019). For instance, according to the 2017 American Community Survey, immigrants from some South American countries such as Venezuela has seen an increase of 76% to 421,000, while immigrant from Central American countries such as Guatemala has increased by 37% to 1.4 million (Noe-Bustamante, 2019).

To this end, the present study aims to examine the pre- to post-immigration alcohol use trajectories of Latinx immigrants from Cuba and South and Central America during the first decade in the US. Our overarching hypothesis is that Latinx immigrants will experience decreases in alcohol use during their initial years in the US (informed by deterrence theory) but will show increased alcohol use as their time in the US increases (informed by acculturation theory). Based on previous findings we anticipate gender-specific effects with increased rates of alcohol use over time being stronger among men compared to women.

2. Methods

Data for the present study were obtained from the first prospective community-based cohort study to examine the pre- to post-immigration alcohol use trajectories among young adult recent Latinx immigrants during their first decade in the US. The study was conducted in a large metropolitan city in the South Eastern United States. The sample consisted of 527 recent Latinx immigrants. Inclusion criteria were: (a) being between the ages of 18 and 34 years old; (b) self-identifying as Latinx; (c) having immigrated to the U.S. from a Latin American country within one year prior to baseline assessment; (d) residing within Miami-Dade County; and (e) intending to stay in the U.S. for at least 3 years.

Respondent-driven sampling was the primary recruitment strategy. This technique is an effective strategy in recruiting participants from difficult-to-reach populations (Salganik & Heckathorn, 2004). Given that the baseline sample consisted of recently arrived immigrants, many of whom had either temporary or undocumented immigration status, respondent driven sampling was deemed as the most appropriate sampling approach. We asked each participant (the seed) to refer three individuals in his/her social network who met eligibility criteria. Seeds were recruited via fliers and in-person throughout Miami-Dade County neighborhoods and businesses with substantial Latinx recent immigrant populations, community-based agencies serving Latinx recent immigrants, and during Latinx health fairs in Miami-Dade County. This referral procedure was followed for three legs for each initial participant (seed), at which point a new seed would begin, thus limiting the number of participants that were socially interconnected. This process was undertaken to avoid skewing the respondent sample (Salganik & Heckathorn, 2004).

Baseline (T1) data were collected commencing in 2007 and consisted of retrospective pre-immigration information (i.e., participants reported alcohol use in the 90 days before coming to the U.S.). Three follow-ups, T2, T3, T4, across the past 10 years, have collected post-immigration data. Specifically, T2 and T3 data were collected in two subsequent follow-up assessments that were 12 months apart between 2008 and 2010. Average time of residence in the US was 2 years at T2 and 3 years at T3. T4 data began in 2017 due to a gap in funding, at which point, participants had been residing in the US for approximately 10 years. Specifically, two study protocols were approved by the institutional review board of a large public university in Southeastern US; the first protocol was for the initial study (T1-T3) and a subsequent protocol for the follow up study (T4). Participants were consented during the original and follow-up study. Due to consistent and on-going retention efforts the RLIS currently yields an exceptional retention rate of over 95% across a 10-year span. Retention efforts included a) dedicated retention coordinator that employed a computerized participant information and assessment schedule database to maintain routine contact with participants every 3 months (e.g., project newsletters, birthday cards, and U.S. and Latin American holidays); b) a combination of mail, email, text reminders, and confirmation calls to help minimize missed assessment; c) the use of corroborative local and international contacts (e.g., sibling, friend, cousin) provided by participants with informed consent. For their participation, study participants were compensated as follows: T1 $50; T2 $30; T3 $60; T4 $50. Further methodological details of this investigation are detailed elsewhere (De La Rosa, Babino, Rosario, Martinez, & Aijaz, 2012, De La Rosa, Dillon, Sastre, & Babino, 2013).

2.1. Measures

2.1.1. Sociodemographic

A demographics form assessed, in part, participants’ age and annual income in the past 12 months as continuous variables. Annual household income in participants’ country of origin prior to immigration was converted to US dollar value. Gender was assessed dichotomously (0 = female, 1 = male), and country of origin was measured categorically as (1 = Cuban, 2 = South American, 3 = Central American).

2.1.2. Alcohol use

The Timeline Follow-Back interview (Sobell & Sobell, 1995) was used to assess alcohol use in the past 90 days. Daily alcohol use was measured in number of standard drinks per day. Consistent with prior research, alcohol use frequency was calculated by summing the total number of days alcohol was consumed, while alcohol use quantity was calculated by the average number of standard drinks on days alcohol was consumed (Sanchez et al., 2015). Binge drinking was assessed by number of binge drinking days in the past 90 days. Based on National Institute of Alcohol Abuse and Alcoholism binge drinking days were: four or more drinks (for females) and five or more drinks per occasion (for males).

2.1.3. Time

Time was a categorical variable with four categories that T1, T2, T3, and T4.

2.2. Statistical analysis

Separate models examined change over time for our three primary outcomes: drinking frequency, drinking quantity, and binge drinking. Drinking frequency and binge drinking were both count measures that were positively skewed with overdispersion and considerable zero inflation. Therefore, Poisson, Negative Binomial, Zero-Inflated Poisson, and Zero-Inflated Negative Binomial (ZINB) models were used to examine optimal goodness of fit for change in the two outcomes overtime. We compared Akaike information criterion (AIC) for generalized linear mixed effects modeling (GLMM), quasi-likelihood under the independence model criterion (QIC) for generalized estimating equation (GEE) modeling, model convergence status, and the estimated overdispersion parameters. A negative binomial model with GEE for drinking frequency, and ZINB model for binge drinking with GLMM, were selected as the final models, respectively. A random effect was included to the GLMM to adjust for the individual variation. Drinking quantity was analyzed using GEE model with gaussian distribution and log link due to the continuous and skewed distribution of the variable (p of Kolmogorov-Smirnov test < 0.01).

All models were adjusted for gender, age, household income, and country of origin. These variables were selected based on epidemiological research with Latinx populations that suggests alcohol use may differ across these demographic factors (Castañeda et al., 2019). National study findings indicate Cubans have the lowest alcohol use rates compared to other Latinx subgroup. As such, Cubans were used as the reference group in the present analyses (Castañeda et al., 2019; NIAAA, 2019). To examine potential interaction effects between time lived in the U.S. and gender, this interaction effect was reported in addition to the main effects. Interpretation of results were based on the values and corresponding 95% confidence intervals (CIs) of the adjusted odds ratios (aORs), adjusted incidence rate ratios (aIRRs), or adjusted estimated correlation coefficients from the GEE model with log link. All analyses were performed using SAS 9.4 software (SAS Institute Inc., 2016), with 0.05 level of significance.

3. Results

3.1. Participant characteristics

A total of 288 males (54.6%) and 239 (45.4%) females participated in the study. At baseline (T1) average participant age was 27 years (SD = 4.98), with a mean time in the US of 6.75 months (SD = 3.12). Approximately 22% were unauthorized immigrants, while 47% had an education level of a high school diploma or less. The study sample represented recent immigrants from seventeen Spanish-speaking countries in Latin America and the Caribbean (Cuba, Dominican Republic, Mexico, Guatemala, Honduras, El Salvador, Nicaragua, Panama, Colombia, Venezuela, Argentina, Ecuador, Peru, Bolivia, Paraguay, Uruguay, Chile). For data analytic purposes country/region of origin were aggregated as follows: 42.1% Cuban, 29.4% Central American, and 27.1% South American. The median household income increased from 4 thousand US dollars at T1 (pre-immigration) to 59 thousand US dollars at T4 (10 years since immigration).

3.2. Alcohol use frequency

Results revealed a significant reduction in number of drinking days from pre-immigration to post-immigration at (T2) (~2 years since immigration) (aIRR = 0.70, p = 0.007) and (T4) (~10 years since immigration) (aIRR = 0.66, p = 0.006). Alcohol use frequency was 1.8 times greater in males compared to females (aIRR = 1.86, p < 0.001). Higher household income was associated with greater alcohol use frequency, with drinking days increasing by 1% for every $1,000 increase in household income (aIRR = 1.01, p < 0.001). Compared to Cubans, immigrants from South America reported greater alcohol use frequency (aIRR = 1.70, p < 0.001). The interaction between gender and time was significant (type III p = 0.003, Fig. 1), indicating different trends in drinking frequency by gender over time. Specifically, alcohol frequency decreased among men from pre-immigration through the first three years in the US. No significant change over time was found among females.

Fig. 1.

Alcohol use frequency and quantity by gender over time.

3.3. Alcohol use quantity

Analogous to our findings for alcohol use frequency, time, gender, and country of origin were significantly associated with alcohol use quantity. Statistically significant declines in drinking quantity were found from pre-immigration (T1) to post-immigration (T2, T3). Specifically, alcohol use quantity significantly decreased by 23% at Time 2 (~2 years since immigration) (β = 0.77, p = 0.013) and 25% at Time 3 (~3 years since immigration) (β = 0.75, p = 0.005). Alcohol use quantity was 2.8 times greater for males compared to females (β = 2.79, p < 0.001). Results revealed a significant interaction between gender and time (type III p = 0.013). Particularly, two of the three interaction effects between gender and time were significant, including gender at T3 and T4, indicating declines in pre- to post-immigration alcohol use quantity among males, with no significant changes among females (Fig. 1). Central American and South American immigrants reported greater alcohol use quantity compared to Cubans (βs = 1.41 and 1.52, p < 0.05).

3.4. Binge drinking

The ZINB model for predicting number of binge drinking days included a Zero-Inflated component to assess abstinence from binge drinking and a Negative Binomial component to assess the number of binge drinking days. Results from the Zero-Inflated portion of the analysis indicated greater rates of abstinence from binge drinking at post-immigration (T2, T3) compared to pre-immigration (aORs = 1.66 and 3.33, p < 0.01). As anticipated, males were less likely to abstain from binge drinking compared to females (aOR = 0.37, p < 0.001). South American immigrants were also less likely to abstain from binge drinking compared to Cuban immigrants (aOR = 0.42, p < 0.001).

The Negative Binomial portion of the model revealed number of binge drinking days were significantly reduced after immigration (T2 & T3) (aIRRs = 0.66, and 0.81, p < 0.05). Rates of binge drinking among males were as high as 2.1 times greater than females (aIRR = 2.10, p < 0.001). South American immigrants reported higher binge drinking rates compared to Cuban immigrants (aIRR = 1.59, p = 0.004). No significant interactions between gender and time were found. Significant random effect (1.46, p < 0.001) suggests considerable heterogeneity across individuals with respect to the number of binge drinking days.

4. Discussion

Findings from the present study challenge the current understanding of alcohol use trajectories among Latinx immigrants. Previous studies have suggested time of residency in the US is associated with increased alcohol use among Latinx immigrants (Lui & Zamboanga, 2018; Salas-Wright et al., 2018). Earlier research in this area guided the development of widely used theories such as the healthy immigrant effect (Flores & Brotanek, 2005). Conversely, our study reveals overall decreases in pre- to post-immigration alcohol use quantity and frequency, suggesting an emerging theoretical paradigm that requires further exploration. We are encouraged that the present study findings provide a potential platform for the development of sophisticated conceptual frameworks with which to build future research investigating pre- to post-immigration alcohol use trajectories among Latinx immigrants.

Our findings suggest that certain demographic factors may be linked with changes in pre- to post-immigration alcohol use in this population. Rather than indicating typical patterns of increased alcohol use among females (and little change in males) over time, the evidence revealed decreases in pre- to post-immigration alcohol use frequency for men with no significant change among women. We posit that these decreases could potentially be related to deterrence theory, namely immigrant’s motivations to abide by alcohol use regulations and its consequences in the US. Stringent societal laws in the US (e.g., prohibition of driving under the influence) may not be enforced in the country/region of origin and could serve as a way of monitoring and directing behaviors by recent Latinx immigrants that prevents unwanted run ins with the criminal justice system (De La Rosa et al., 2012).

With regard to gender, and consistent with previous literature, males reported greater alcohol use frequency, quantity, and binge drinking compared to females. These findings confirm the need for alcohol use intervention strategies that target specific Latinx immigrant subgroups, accounting for differences such as gender and national origin.

Overall decreases in binge drinking were found from pre-immigration through first three years in the US. However, no significant differences between pre- and post-immigration binge drinking were evident 10 years after immigration, suggesting increases in problematic alcohol use to pre-immigration levels as Latinx immigrants’ time in the US increases. Notably, an increasing trend towards pre-immigration rates of alcohol use frequency was also observed among males. These findings suggest that the initial years after immigration may be a key time for prevention interventions among recent Latinx immigrants with a history of alcohol misuse.

In relation to income, participants with lower income reported lower alcohol use frequency. While these results seem paradoxical, it may be that limited financial ability to purchase alcohol and a desire to “stay under the radar” may have health-promoting effects for Latino immigrants despite the stressors associated with these challenging circumstances.

Regarding national origin, Cuban immigrants reported the lowest rates of alcohol use compared to Latinx immigrants of Central and South American origins. These findings are in line with previous evidence indicating lower rates of alcohol use among Cubans, when compared to other Latinx subgroups (Castañeda et al., 2019). While previous national studies have combined South and Central American immigrants into one group, we disaggregated them into two distinct subgroups in our analyses. This allowed for a more fine-grained examination of alcohol use patterns by national origin. Results showed South American immigrants had significantly higher alcohol use quantity, frequency, and binge drinking rates compared to Cuban immigrants, whereas Central American immigrants reported greater alcohol use quantity than Cuban immigrants. These findings address a fundamental gap in the literature regarding the drinking patterns of Cuban and South and Central American immigrants—Latinx subgroups that have been understudied but are quickly becoming representative of the US Latino population.

Another factor that should considered when interpreting results from the present study is the context of the receiving community. This study was conducted in a metropolitan city in the South Eastern United States with a large and diverse Latinx immigrant population that tends to enjoy more political and economic advantages compared to other cities (Schwartz et al., 2014; Stepick & Stepick, 2002). Well-established Latinx immigrant receiving communities with dense ethnic enclaves can provide increased availability to culturally and linguistically congruent services, tangible and intangible support systems, and lower levels of discrimination that could ease the process of acculturative stress and its associated health compromising behaviors such as alcohol use. As such, the concentration (or lack thereof) of Latinx immigrants in receiving communities is important to consider—suggesting that immigrant communities that are not well-established may provide very different results compared to those found in the present study (Schwartz et al., 2014).

Additionally, examining the moderating effects of demographic factors such as country of origin or income were outside the scope of this investigation. Future directions should include examining the interaction effects of these and other demographic factors as they relate to pre- to post-immigration alcohol use trajectories in this population.

4.1. Study limitations

Results from this study should be interpreted in light of several limitations. Although respondent-driven sampling is a preferred method to recruit “hard to reach” populations such as recent immigrants it does not ensure a representative sample. Additionally, pre-immigration alcohol use data were collected retrospectively, making this information susceptible to recall errors and under-reporting. However, because well-established timeline follow back procedures were employed (Sobell & Sobell, 1995) this limitation may have been minimized. In addition, documentation status was not included as a covariate in the present study as it was a confounder with national origin. At the time of data collection, the Cuban Adjustment Act of 1966 allowed for any individual who fled Cuba and entered the US to pursue residency one year later. Subsequently, the “Wet foot, Dry foot” law that passed during the Clinton administration in 1995 offered Cubans that reached US soil a clear path to citizenship upon entering the US. As such, all Cuban immigrants in the sample were documented immigrants. At baseline 30% of the South and Central American participants were unauthorized immigrants, by T4 approximately 95 percent of participants held authorized immigrant status. Future studies would benefit from examining how changes in immigration status impact pre- to post-immigration alcohol use among Latinx immigrants. It also remains to be seen how recent normalizations of diplomatic relations between US and Cuba and the repeal of the “Wet Foot, Dry Foot” Act in 2017, impacts the Cuban immigrant experience and subsequent alcohol use trajectories. Lastly, due to gaps in funding, there was a 7-year assessment gap between T3 and T4. Nonetheless, a 95% retention rates was achieved across the study’s 10-year duration. This high retention was primarily due to efforts by the investigative team to retain study participants by continuing to maintain annual contacts with them and updating their contact information until funding was once again attained.

5. Future directions

Despite its limitations, the present study is among the first to report on pre- to post-immigration alcohol use trajectories of Latinx immigrants during their first decade in the US. Future studies identifying how various risk and protective sociocultural factors are associated with these distinct alcohol trajectories is warranted. This knowledge is important not only from an epidemiological perspective, but for the development of effective culturally relevant alcohol interventions targeting Latinx immigrants early in the immigration process.

HIGHLIGHTS.

• Findings revealed overall declines in pre- to post-immigration alcohol use.

• Gender, income, and national origin were associated with distinct drinking patterns.

• Results inform future alcohol intervention research on Latinx immigrants early in the immigration process.