Alcohol use among two generations of Southeast Asians in the U.S

Abstract

The paper assesses the drinking norms and practices of two generations of Southeast Asians in the East San Francisco Bay Area. Researchers included quantity and frequency measures of current alcohol use and binge drinking and open-ended questions on drinking norms and behaviors in a mixed-method study of tobacco use. The study generated data through in-person interviews with 164 respondents from two urban East Bay neighborhoods. The findings include normative drinking among women and underage youth, with heavy drinking among underage youths particularly troubling. Preventive interventions for this population should consider community-level alcohol problems as well as individual alcohol use disorders.

INTRODUCTION

Among U.S. racial/ethnic groups, Asians are considered to have low prevalence of alcohol use disorders. Drinking rates among U.S. Asians are generally shown to be much lower than the U.S. general population and than other racial/ethnic groups (Sue, 1987; Sue et al., 1979), which has been used as another indicator of Asian Americans’ “model minority” status in the U.S. Consequently, reduction of alcohol use among Asian Americans has not been a high priority in substance abuse prevention programming. Researchers investigating substance use and misuse among Asian Pacific Islanders have taken issue with this generalization on several counts. Firstly, language and cultural barriers to their participation in surveys and clinical studies may result in drinking among Asians being underreported (Kuramoto, 1994; Price et al., 2002; Sakai et al., 2005; Zane and Sasao, 1992; Zane and Huh-Kim, 1998). Secondly, while Asians may drink at lower rates, studies have shown patterns of heavy drinking when they do drink (Amodeo et al., 1997; Chi et al., 1989; Makimoto, 1998; Sue et al., 1979). Finally, the term “Asian” includes multiple ethnic groups whose use of alcohol and other substances may vary (Sue, 1987; Zane and Sasao, 1992). Low alcohol use prevalence rates among the larger East Asian groups (particularly Chinese and Japanese) may mask higher rates among smaller groups (Zane and Huh-Kim, 1998), such as Southeast Asians.

Southeast Asia spans a large geographic area, from mountainous Burma to the thousands of islands that make up the nation of Indonesia and including the Philippines, Malaysia and Brunei as well as the Indochinese region of Thailand, Laos, Cambodia and Vietnam. Hence the term “Southeast Asian” includes a large number of peoples with widely disparate cultures, socioeconomic lifestyles and socio-political histories. Because they are represented in the U.S. in smaller numbers relative to other racial/ethnic groups, Southeast Asians are not typically included in large-scale studies of alcohol and other drug use, and when they are included, small numbers make disaggregated analyses impossible. Some researchers have indicated troubling patterns of substance use and misuse among Southeast Asians, including use of illicit drugs (Nemoto et al., 1999; Westermeyer et al., 1989), misuse of over-the-counter medications (D’Avanzo et al., 1994), and high rates of cigarette smoking (Jenkins et al., 1995; Levin et al., 1988) as well as alcohol misuse (D’Avanzo and Barab, 2000; Wong et al., 2007). Most such reports are based on community samples, and as such the data are difficult to interpolate.

Random sample surveys of Southeast Asians in the U.S. have been expensive to implement due to sampling and recruiting issues. One such study of the mental health and well-being of recent Southeast Asian immigrants, conducted in the 1980s in San Diego, CA, showed high rates of cigarette smoking and mental health disorders, but did not report on alcohol use (Rumbaut, 1989). A more recent random sample survey was conducted in Long Beach, CA, which is home to the largest Cambodian community in the U.S., and did include alcohol use items. The authors of this study found relatively low alcohol use and related disorders (D’Amico et al., 2007). These studies as well as all but a very few of the previous studies of Southeast Asian substance use were based solely on quantitative data and thus lack the explanatory information available through qualitative data, such as contexts of use, culturally-specific practices, and rationales for alcohol use, which may aid in better defining patterns of alcohol use and related problems among Southeast Asians.

In this paper we present quantitative and qualitative findings on alcohol use from a sample of Laotians and Cambodians in California’s San Francisco Bay Area. Laotian and Cambodian refugee immigration to the U.S. began during the Vietnam War in the 1970s. Many Southeast Asians settled in depressed urban areas throughout the United States. Of U.S. states, California has the largest numbers of Laotians and Cambodians; of U.S. metropolitan areas, the San Francisco Bay Area has the largest numbers of Laotians (11,545) and the fourth-largest population of Cambodians (10,552) (www.hmongstudies.org). Participants for this study were drawn from Oakland and Richmond/San Pablo, two urban Bay Area communities with large numbers of Southeast Asian residents.

METHODS

Data on alcohol use was collected in a mixed-method study focusing on norms and practices related to use of tobacco and other substances among two generations of Bay Area Southeast Asians. The sample included 164 respondents and was stratified by gender, smoking status (smoker or non-smoker), generation in the U.S. and by ethnic group (Laotian and Cambodian). Due to difficulties identifying and recruiting males who had never smoked, we included former smokers (no smoking in the past year) in the category of non-smoker. We defined the first generation as persons who were born in Southeast Asia and lived more than 1/4 of their lives there, while the second generation included persons who were born and raised in the U.S. as well the “one and a half” generation, i.e. those who were born abroad but lived more than one-fourth of their lives in the U.S. (Min, 2002). The second generation respondents ranged in age from 15 to 28 with a mean age of 19.3. Although Southeast Asian adolescents may be engaging in substance use at younger ages, youths below age 15 were judged to be too immature for the in-depth semi-structured interview component of the study. Because older Southeast Asians have been underrepresented in the literature due to language and accessibility barriers (Wong et al., 2007), the research team specifically sought to include older respondents from across the span of age groups. As a result, the first generation respondents ranged in age from 30 to 87, with a mean age of 49.2. The mean age of the total sample was 35.4 (Table 1).

Table 1.

Characteristics of respondents, N=164

Variable	N (%)
Cambodian	81 (49)
Laotian	83 (51)
Male	78 (48)
Smoker	82 (50)
First generation	88 (53)
Second generation	76 (46)

Since the legal minimum age to consume alcohol in the U.S. is 21 years, for this paper we also aggregated respondents as either adult (21 years and over) or underage (under 21 years). According to this division, the sample included 113 adults ranging in age from 21 to 87 with a mean age of 43.5, and 51 underage youths ranging in age from 15 to 20 with a mean age of 17.27.

Respondents were recruited by a combination of agency and personal contacts together with snowball referrals. Data was collected using English-language instruments as well as Lao and Khmer versions which were translated and back-translated for accuracy. Interviewers obtained informed consent for all interviews, with additional parent consent obtained for minor respondents. Each respondent received $40 for participating. All research data were identity-coded and securely stored. The Institutional Review Board of the Pacific Institute for Research and Evaluation approved all data collection protocols for the protection of human subjects.

Quantitative data was collected through a brief interviewer-administered survey which included background demographics and standardized measures of tobacco, alcohol and other drug use. The alcohol use items included frequency and number of drinks consumed. Because some non-native-born populations may not consider items such as beer or homebrewed liquors as “alcohol” (Strunin, 1999; Wong et al., 2007) respondents were prompted to consider all alcoholic beverages including beer, wine and hard liquor. Data from the completed surveys were analyzed using SPSS.

Cross-tabular analyses were used to compare alcohol consumption differences across gender and generational groups. Where appropriate, categories were collapsed in order to increase cell size. Measures used for the analysis of alcohol consumption included “current use,” defined as at least one drink in the last 30 days, and “binge drinking,” defined as five or more drinks on the same occasion at least once in the past 30 days. As drinking has been linked to smoking (Moore and Lee, in review; Room, 1994), we wished to assess the degree to which our sample strata, particularly sampling by smoking status, may have biased the findings on drinking. A pseudo-mean was generated for average number of drinks per day by taking the average number of drinks per day by number of drinking days in the past 30 days.

Qualitative data were collected in semi-structured interviews which were digitally recorded. Interview recordings were transcribed and, if necessary, translated into English. The transcripts were uploaded into ATLAS/ti qualitative data management software (Muhr, 1991) and subsequently coded using category codes (such as “male,” first generation,” or “smoker”) as well as thematic codes suggested by review of the literature, the research questions and emergent themes (such as “alcohol,” “family,” or “social influences”). Analyses for this paper were conducted by extracting and reviewing all coded texts related to alcohol use, comparing the data by category codes, and sorting the texts by overarching and subsidiary themes.

RESULTS

Problematic drinking patterns

Analysis of the quantitative data indicated problematic patterns of alcohol use among the Southeast Asians in the study, compared to national and local trends. 64% of all respondents reported drinking in the past 30 days, and 20% reported binge drinking. Drinking rates compared by gender were closer than expected: 63% of women compared to 65% of men reported current drinking, and 15% of women compared to 24% of men reported binge drinking (Table 2).

Table 2.

Patterns of alcohol use

Variable	Total sample (N = 164)	Male (N = 82)	Female (N = 82)	Adult (N = 113)	Underage (N = 51)
Any drinking in past 30 days	64%	65%	63%	68%	55%
Any heavy episodic drinking in past 30 days	20%	24%	15%	18%	23%

Alcohol consumption rates were compared to reports of drinking prevalence in the overall U.S. population and among populations described as “Asian” from the 2004 National Survey on Drugs and Health (NSDUH). For respondents who were 21 years and older, the current drinking rate was 68%. This exceeded national results for adult Asians of 37.4% as well as the U.S. adult population rate of 54.5%. The rate of binge drinking among the sampled adults was 18%. This rate, while lower than the U.S. total of 22.8%, still exceeded national rates for Asians at 12.4%. 55% of underage youths in the study reported current drinking-- nearly twice the rate of underage drinking nationwide, at 28.7%, and more than three times the rate reported for underage Asian youths at 16.4%. 23.5% of underage youth in the study reported binge drinking. This was more than reported by U.S. youths overall at 19.6% and nearly three times higher than that reported by underage Asians nationally at 8.0%.

Based on the pseudo-mean drinks measure, overall, males consumed more drinks than females in nearly all categories, smokers consumed more than non-smokers overall, and older respondents more than younger respondents. For non-smoking younger respondents (second generation), however, females consumed slightly more alcohol than males. Older respondents (first generation) consumed more drinks than younger respondents, although for females the amounts were somewhat closer between generations. Smokers consumed approximately twice as many alcoholic drinks as non-smokers in most gender and generation categories, although there was less difference between younger females who did or did not smoke (Table 3).

Table 3.

Sample strata compared by mean number of drinks in the past 30 days

	Men	Women
First generation	19.8	13.5
Smokers	28.6	19.3
Non-smokers	11.4	7.6
Second generation	14.0	11.6
Smokers	19.2	13.0
Non-smokers	8.5	10.1

Traditional contexts of alcohol use

Analysis of the qualitative data indicated that drinking alcohol was highly normative for both generations of Southeast Asians. Alcohol was typically provided and consumed at family and community gatherings such as New Year’s celebrations, religious holidays and birthday parties as well as casual house parties, which could be quite large. Drinking among older people was generally associated with such events, and many respondents said they only ever drank in these situations. One man noted, “In Laos we do drink a lot but we drink on occasions.” Respondents described alcohol and cigarettes as traditionally expected and ritually necessary offerings to house visitors, family members, religious leaders, traditional healers and ancestral spirits, and drinking was expected at most social occasions, although women might drink less than men.

Social drinking and drinking alone

Many older Southeast Asians, however, appeared to be frequently drinking alone and/or drinking to be drunk. Many elders in particular lived in relative isolation and with a good deal of depression and anxiety from traumatic pasts and unhappy presents. An older Cambodian woman reported that many evenings “I would just sit there, and drink and smoke.” An older Laotian woman stated, “If you don’t drink, then you have things to think about, and if I think about them, I can’t fall asleep a lot of the times. And so I drink alcohol.” When asked why he thought Laotian people drank, one youth said it was probably “to forget the past; because they got a bad past; they just don’t want to remember it.”

Social influences on alcohol use

Although drinking was described as quite common in their home countries, some older respondents described quitting or cutting back on drinking since arriving in the U.S. Usually this was for health reasons, sometimes on the advice of a doctor. Some reported receiving drunk driving citations, or worrying about this, as a reason to cut back. Many did not cut back, however. One respondent observed a friend drinking even after suffering severe health problems. The respondent blamed peer pressure for his friend’s continued drinking: “The other guys around him would say, ‘You gotta drink, that’s good for you, it helps, heals’.” One older woman described her husband’s social drinking: “If they drink alcohol one day, and his friends come over, five bottles is still not enough! They keep pouring and pouring it to drink, and then they get very drunk.” During field observations among young adult Southeast Asians, researchers also noted the tendency to drink to get drunk, and typical social gatherings for young adults involved drinking shots of hard liquor in multiple rounds.

Alcohol and gender identities

Drinking and smoking were considered by both younger and older respondents to be attributes of masculinity. Male drinking was identified with being “hard and tough.” An older man stated, “If [you are] a man, you have to drink, you have to smoke!” Men who were “not good at drinking alcohol” were said to have a “weak neck,” while those with a “strong neck” liked to get intoxicated. Drinking and smoking cigarettes among women were closely linked for most respondents. Attitudes towards alcohol consumption by women showed a marked difference by generation in the U.S. Several older respondents stated that women do not drink as much as men; some noted that custom and tradition prohibited female drinking and smoking. Women who did drink and smoke were said to be considered “loose,” “daring,” “immoral” or “not good.” These attitudes showed signs of shifting in the U.S. Some respondents described female drinking and smoking as indicating a woman was “modernized.” One respondent noted that living in the U.S. might have changed her mother’s habits, allowing the older woman to add drinking to the repertoire of social activities for herself and her circle of friends. Female drinking was less stigmatized among younger compared to older respondents. Some younger respondents did, however, note a correlation between girls who drank heavily and a “bad reputation” for sexual promiscuity.

Families and alcohol use

Women of both generations described the traditional use of the herbal-alcohol infusions during and after pregnancy, often insisted upon by the older women. Some young women described quitting or cutting back on their drinking and drug use because of having children. Others who had quit or cut back attributed the use of the herbal-alcohol mixtures with their subsequent relapse into regular or heavy drinking. One youth described his mother drinking every day. Some youths described drinking regularly with their mothers.

Normative drinking among youths

Younger respondents described drinking as so common among their peers as to be considered “no big deal.” Some noted using alcohol to relieve stress or depression, or to “be cool,” “fit in.” Many described drinking and drug use as a way to pass time in what they considered the stressful or boring circumstances of their lives: “Just trying to kill the day ’cause you ain’t got nothing else better to do.” Most younger respondents described drinking in order to “get buzzed” or feel drunk. Many stated alcohol was not addictive, but noted that the amount needed to get drunk increased as they “got used to it.” Drinking was in general considered a social behavior by younger respondents. Solitary drinking and daily drinking, but not heavy drinking, were associated with being an “alcoholic” by many youths. Heavy drinking was not uncommon, even at early ages: “They got kids that’s 13, 14, they’re like professional drinkers,” observed one youth.

Alcohol availability and drink preferences

Interview and observational data indicated that hard liquor and beer were the most common types of alcohol. At family gatherings, offering “top shelf” liquors was noted as a status symbol among adults. While some youths reported not drinking at these events for fear of being caught and punished, others described sneaking or being given their first sips of alcohol and learning to drink by watching their elders there, and younger respondents reported a preference for “Remy” and “Hennessy.” Alcohol was described as readily available, even for minors, in their neighborhoods. The most common strategy for underage drinkers was to ask either drug addicts loitering near a store or an older friend to buy liquor for them, although some described buying alcohol for themselves. A few noted the use of traditional herbal infusions; one respondent stated that these infusions were particularly potent, and “if we really want to get drunk, we drink our herbs.” Local service providers noted that traditionally these herbal tonics were based on home-brewed rice wine, a “white” alcohol, and that in lieu of this Southeast Asians in the U.S. were substituting other “white” liquors such as vodka and gin, with much higher alcohol content, without modifying drinking amounts. Women in their 30s and 40s were said to have taken up wine drinking as a sign of class advancement, hosting wine tasting parties, and serving wine and liquor at many social occasions, including children’s parties.

Drinking, partying and violence

Youths discussed drinking alcohol as an integral aspect of socializing with friends, either in small groups at their own or a friend’s home, or at weekend parties. Many described skipping school to drink, or drinking at school. Youths’ most frequently cited reasons to drink were “to have fun and party.” Some noted that alcohol allowed for deeper levels of social contact: “When you drink, you express your feelings, the inner spirit comes out,” one said. Some respondents noted using alcohol as an aid in approaching potential romantic or sexual partners at parties. Violence and drinking were frequently linked in descriptions of youth parties: “If there’s alcohol involved, there’s a fight,” one respondent reported. Some respondents included violence as a reason to drink: “They wanna have a fight so they drink,” one youth noted. Some respondents described complicated typologies of alcohol types related to modalities of being: “The brown calm you down, the white make you fight. People with white [liquors], they like to get real violent. People with brown [liquors], you see ‘em, they mellow, they just kickin’ it.” Beatings, shootings and stabbings were frequently described in relation to drinking parties, as were property crimes such as car theft and vandalism. Violence and crime was also related to alcohol procurement; youths reported that a common strategy for obtaining cash to buy alcohol was to rob someone.

Abstinence among Southeast Asians

Some respondents described deliberate moderate drinking, having limited their drinking or abstaining altogether. A few respondents stated that they didn’t drink very much or at all because they were “allergic” to alcohol, noting that it made them feel hot, made their faces turn red, or caused “tiny bumps” or “dots” on their skin. On the other hand, some respondents observed that friends who had this reaction “just kept drinking.” Abstainers said that they did not like the taste of alcohol, that they never acquired the habit, or that their friends did not drink. One young abstainer observed that when her friends drank they would “act stupid,” as well as argue and fight; others described observing negative effects of family members’ drinking.

DISCUSSION

Current drinking and binge drinking among the Southeast Asians in this study were found to be highly prevalent—higher in some aspects than national rates and many times higher than rates reported for “Asians” in the U.S. Especially troubling was the high prevalence of drinking among underage youth, and the apparent associations of drinking with violent and criminal behavior. Ethnographic studies have indicated that although use of alcohol and intoxicating substances may vary by ethnic group in Southeast Asia, in general alcohol has traditionally been consumed under specific social circumstances, such as ritual occasions and lifecycle events (Westermeyer, 1979; Westermeyer, 1985) whereby problematic or stigmatized behaviors such as excessive drunkenness and drinking by children and women may be monitored and controlled. In the new context of life in the U.S., traditional social controls on substance use may be more difficult to maintain (Lee and Kirkpatrick 2005), opening the possibilities of heavier and more frequent drinking, alcohol misuse and alcohol dependence.

Environmental and cultural influences specific to the Southeast Asian refugee experience may encourage drinking. Since arriving in California, many Southeast Asians have experienced racism and inter-racial tensions as well as community violence. In response, many Southeast Asians joined street gangs, which involvement has been found to contribute to drug and alcohol use and abuse (Dembo et al., 1985; Hunt et al., 1996). Social and linguistic isolation may contribute to drinking problems among adults, particularly elders. As in other studies of Southeast Asian refugees (D’Avanzo, 1997; Yee and Thu, 1987), many adults in our study reported using alcohol, tobacco and other drugs to self-medicate for depression and anxiety. The role of alcohol availability should not be underestimated; off-the-shelf drinks make alcohol more accessible and in more varieties of alcohol content for adults as well as youths.

Our finding of 68% current drinking among adults is somewhat higher than rates reported by Wong and colleagues for a sample of Southeast Asians in the Washington D.C. area, with 52.6% for 18–24 year olds and 38.3% for those 25 and older (Wong et al., 2007). Our analyses indicate that sample bias may be attributed with some but not all higher than expected rates of drinking in our study. Studies generally show low rates of smoking among Asian women in the U.S. and in Asia (WHO, 1997), and among the first generation respondents women who smoke are most likely oversampled in our study, although among younger women our analyses showed little difference between smokers and nonsmokers in terms of drinking. Conversely, as, smoking among Southeast Asian men has been reported at rates of 65% (Rumbaut, 1989) to 72% (Levin et al., 1988), and our analyses showed drinking and smoking to co-occur among males of younger and older males, drinking among Southeast Asian men of both generations may be underreported here.

Our findings are in contrast to the recent study by D’Amico and colleagues based on data gathered among Cambodians in Long Beach, CA, who found current drinking among only 26% of respondents and heavy drinking among only 2% (D’Amico et al., 2007). We should note that while the Long Beach included only first generation immigrants, with a mean age of 52.2, our bi-generational comparison included second generation youths in addition to older adults, with a mean age of 35.4 for the entire sample. Previous research has indicated an association between increased drinking among Asians and longer time in the U.S. as well as with those born in the U.S. compared to those born overseas (Zane and Sasao, 1992; Zane and Huh-Kim, 1998). Nevertheless, even among older drinkers our study showed higher rates of alcohol use than in the Long Beach study.

A further consideration is the environmental context within which the two studies were situated. The Long Beach Cambodian population constitutes a large ethnic enclave, which may provide some protection from stressors related to substance misuse, as well as increased social control and monitoring of individuals’ behaviors. Residence in ethnic enclaves has been linked with positive health outcomes for immigrants (Cagney et al., 2007; Peak and Weeks, 2002). Possibly more like the Southeast Asians in the Washington D.C. study, the Cambodians and Laotians in our study resided in neighborhoods which were not only ethnically mixed, but poor and with high rates of violence as well. At the time of the 2000 census, Oakland was 35% African American, 22% Hispanic and 15% Asian, with 19% of residents below poverty level, while the Richmond/San Pablo area was 31% African American, 31% Hispanic and 13% Asian, with 17% of residents below poverty level (U.S. Census 2000). According to 2006 FBI crime reports, the per capita violent crime rate for Richmond was twice and for Oakland nearly fourteen times the national average, while the murder rates in each of these cities were seven times the national rate (www.areaconnect.com). Such environmental factors have been identified as increasing the risk of problematic behaviors, including underage drinking and alcohol abuse.

Finally, D’Amico and colleagues raise the question of when is drinking a problem for a given population. We suggest that drinking problems for Southeast Asians – or any group, whether defined ethnically, culturally, or geographically – may be considered to be two-fold. On the one hand are problems at the individual level such intoxication, alcohol dependence, and attendant health issues such as liver cirrhosis and heart conditions. On the other hand are issues at the community level. Community alcohol problems include alcohol-related violence and aggression (Fagan, 1990; Haggard-Grann et al., 2006; Lipton and Gruenewald, 2002; Young et al., 2008); underage drinking, which in addition to being an illegal behavior and linked with other high risk behaviors such as delinquency and school drop-out, has also been linked with higher rates of problem drinking in adulthood (Grant and Dawson, 1997; Hingson and Kenkel, 2004) and with poor health outcomes in later life, such as cardiovascular risk (Fan et al., 2008); and women’s drinking, which for pregnant women has been correlated with fetal alcohol spectrum disorders (Caetano et al., 2006; Tsai et al., 2007) and has been implicated in conditions from breast cancer and osteoporosis to depression, suicide and domestic violence (Hill, 2002). In light of this, our findings on the destigmatization and normalization of alcohol use among prenatal and post-partum women in this population, which echo similar findings on drinking among pregnant Cambodian women (D’Avanzo and Barab, 2000), are troubling. Similarly, our findings on the high prevalence of youthful drinking and heavy drinking, particularly when combined with the indications a relationship between youthful drinking and crime and violence, are indications of another area of problematic alcohol use for Southeast Asians.

Deterrents to problematic and underage drinking among our respondents appear to stem primarily from the experience of negative consequences, such as criticism or punishment of underage drinkers by adults and negative health, legal and social consequences for adults. Some respondents reported quitting or reducing due to the advice of doctors, but these cases may be limited to persons in chronic poor health. The so-called “flushing response,” caused by the inactive ALDH2 gene and thought to contribute to lower rates of drinking and alcohol use disorders among Asians (Wall et al., 2001), appears to be of negligible influence among these Southeast Asians, since determined drinkers reportedly ignored the discomfort and continued drinking. These findings support previous research indicating that cultural norms and alcohol expectancies may greatly mediate the efficacy of the flushing response in preventing alcohol misuse (Hendershot et al., 2005; Johnson and Nagoshi, 1990; Johnson et al., 1990). However, data on the flushing response were not systematically collected in this study, and no genetic material was collected for analysis.

The data presented here should be interpreted with caution. As the study is based on a community sample and thus is not random, the findings cannot be said to represent the entirety of Southeast Asians. For example, while the researchers made many efforts to diversify the sample within the community, it may be that we unintentionally sampled pockets of heavier drinkers. However, as Wong and colleagues have noted (Wong et al., 2007), targeted snowball sampling such as reported on here may be the only way to access subgroups like Southeast Asians who may be not only underrepresented but also at risk for substance misuse. The findings may indicate the utility of disaggregating by subethnic groups in considering problematic drinking among racial/ethnic minorities.