The Most Critical Unresolved Issues Associated with Race, Ethnicity, Culture, and Substance Use

Abstract

This paper discusses the limitations of previous research on race, ethnicity, culture, and substance use. The study offers the following recommendations for future research in this area: (1) move beyond simple comparisons of mutually exclusive groups, (2) focus on the meaning of an ethnic label to the individual, (3) consider the complex interactions between an individual’s cultural identity and the cultural context, (4) understand and acknowledge the researcher’s inherent biases, and (5) translate research findings into practice and policy change.

Over the past century, our world has transformed from a collection of distinct, isolated cultures to a global culture, where most people have numerous opportunities to experience diverse cultural norms, values, and practices through interaction with others around the world and their exported media culture (Arnett, 2002). This is perhaps most true among the young, who take for granted that the world is literally at their fingertips. The Millennial Generation, those born since the mid-1980s, has never known a world in which obtaining information, opinions, news, and video about virtually any topic was any more difficult than typing a few words into an internet browser (Greenberg & Weber, 2008). The youngest members of this generation have never known a world in which it was not possible to communicate with unlimited numbers of reallife and virtual friends instantly, from nearly anywhere, on a device that fits in a pocket. At this point in history, it would be naïve to assume that young people’s decisions, including decisions about whether or not to experiment with alcohol, tobacco, and other drugs, are influenced only by their local culture and not by the global culture.

Has research on substance use and misuse kept pace with this dramatic transformation? Sadly, it has not. The field still conceptualizes racial, ethnic, and cultural groups as being distinct, isolated clans that can be merely compared and contrasted with one another. Indeed, this is definitely an advance over earlier research that did not consider racial, ethnic, and cultural differences at all. A small scientific revolution occurred when researchers began to focus on differing patterns of substance use across groups and realized that what was true among middleclass Whites in the United States or Europe was not necessarily true among minority groups or in other cultural contexts. This led to the recognition of racial and ethnic disparities in substance use, which led researchers to conduct research among underserved groups. This body of research grew exponentially: numerous substance use behaviors were compared across numerous groups and subgroups; posited risk and protective factors identified in one group were put into regression models to predict substance use in other groups; and evidence-based prevention programs were evaluated across populations and cultural contexts (for reviews, see Chen & Yin, 2008; Edwards, Giroux, & Okamoto, 2010; Giovino, 1999; Harachi, Catalano, Kim, & Choi, 2001; Hawkins, Cummins, &Marlatt, 2004; Horigian, Lage, & Szapocznik, 2006; Ivers, 2003; Kerr-Corrêa, Igami, Hiroce, & Tucci, 2007; Kim, Withy, Jackson, & Sekaguchi, 2007; Makimoto, 1998; Mermelstein, 1999; Moolchan et al., 2007; Payne & Diefenbach, 2003; Sussman, 2005; Szapocznik, Prado, Burlew, Williams, & Santisteban, 2007; Tauras, 2007; Zhang & Wang, 2008). A PubMed search for (ethnic OR ethnicity OR race OR racial) AND (drug OR alcohol OR tobacco OR smoking OR cigarette) AND (adolescent OR adolescence OR adolescents) revealed 176 articles before 1970, 635 articles between 1971 and 1980, 1160 articles between 1981 and 1990, 3121 articles between 1991 and 2000, and 6731 articles between 2001 and 2010.

Research about racial, ethnic, and cultural factors in substance use and misuse has become increasingly sophisticated over the past few decades. The earliest studies documented simple differences across ethnic groups in the prevalence of use of specific substances. This allows researchers and policy-makers to label certain groups as being “high risk” and “low risk,” which is useful for allocating resources for prevention and treatment but is also highly stigmatizing. This labeling, rarely if ever, delineated between imminent risk, long(er)-term risk and nonrisk (temporal-based posited protective mechanisms) or the necessary conditions (endogenous as well as exogenous ones, microlevel to macrolevel) for the processes to operate (begin, continue, change as realities changed, cease, begin again, etc.) or not to operate. Not surprisingly, the groups that are typically identified as having a high risk for substance use-related problems also tend to be the groups that have experienced sociopolitical and economic strain, blocking their access to educational and employment opportunities that could serve as more adaptive vehicles to attain universal goals such as feelings of belonging, comfort, and pleasure. Ethnic differences in substance use often become nonsignificant after statistically controlling for conflated variables such as socioeconomic status (Lillie-Blanton, Anthony, & Schuster, 1993; Maddahian, Newcomb, & Bentler, 1986). Rather, factors associated with being amember of a disadvantaged minority group are likely responsible, including limited access to resources and opportunities, discrimination, and exposure to stressful neighborhood environments.

Another stream of research has investigated the effects of living in different cultural contexts on the risk of substance use among immigrants. Numerous studies have documented that the risk of substance use increases as minorities acculturate to the “mainstream” culture of the United States or other Westernized, industrialized countries (Bethel & Schenker, 2005; Choi, Rankin, Stewart, & Oka, 2008; De La Rosa, 2002). This is especially true among women, who typically have very low prevalence of substance use in their home cultures but sometimes embrace the freedom and relative gender equality in Westernized nations by experimenting with behaviors that were taboo for women in their home cultures, including substance use. Although early research in this area made the assumption that exposure to Westernized cultures made immigrants more susceptible to substance use, more recent research has clarified that it is not exposure toWesternized cultures per se that increases the risk of substance use, but the culprit is actually the loss of or alienation from protective traditional collectivist cultural values (Schwartz, Unger, Zamboanga, & Szapocznik, 2010).

The next step in this line of research has been to identify the risk and protective factors that are responsible for disparities in substance use. Many studies have moved beyond simple descriptions of intergroup variation in prevalence and have attempted to uncover the underlying cause of these disparities. For example, some researchers have speculated that family influences are stronger among some ethnic groups than among others (e.g., Gillmore et al., 1990) or that tobacco and alcohol companies target their advertising to certain groups (e.g., Grier & Kumanyika, 2010). Many of these studies actually end up finding more similarities than differences in the predictors of substance use (Barrera, Castro, & Biglan, 1999; Flint, Yamada, & Novotny, 1998; Griesler et al., 2002; Kandel, Kiros, Schaffran, & Hu, 2004; Skinner, Haggerty, & Catalano, 2009; Warheit et al., 1995). For example, friends’ substance use tends to be the strongest predictor of substance use across groups (Flannery, Vazsonyi, Torquati, & Fridrich, 1994), although some studies have found that it tends to be stronger among Whites and weaker among African Americans (Newcomb & Bentler, 1986; Unger et al., 2001). Other studies have found racial and ethnic differences in the strength of the associations between posited risk/protective factors and substance use (Chartier, Hesselbrock, & Hesselbrock, 2009; Tobler, Livingston, & Komro, 2011; Tanski, Stoolmiller, Gerrard, & Sargent, in press; ; Pampel, 2008). Overall, the findings from these studies suggest that the risk and protective factors for substance use are similar across groups (e.g., peer influences, parental monitoring, media exposure, etc.), but their relative strength may vary across groups.

If certain ethnic groups are indeed at greater risk for substance use and its adverse consequences, prevention and treatment programs are needed to reduce their risk factors and/or increase their resilience to cope with risk factors without resorting to substance use. Many evidence-based substance use prevention programs exist, and most have been evaluated in culturally diverse and/or minority samples. Many of these studies have concluded that general-audience programs are effective for minority youth or that programs developed for use in one racial or ethnic group can be exported successfully to other groups (Botvin, Griffin, Diaz, & Ifill-Williams, 2001; Faggiano et al., 2008; Kumpfer, Pinyuchon, Teixeira de Melo, & Whiteside, 2008). However, it has been argued that minorities could benefit even more from interventions that are targeted to their unique cultural characteristics. Cultural targeting includes any adaptation of a program that makes it more relevant to a specific population. This is often limited to surface targeting (e.g., changing the skin colors and names of the characters in booklets), but sometimes includes deep targeting (invoking ingrained cultural values, using culturally meaningful role models, symbols, and modes of communication) (Resnicow, Solar, Braithwaite, Ahluwalia, & Butler, 2000). A few culturally targeted programs have been evaluated rigorously and have been shown to be effective in preventing or reducing substance use in the short term, although long-term effectiveness is not known (De Heer Koehly, Pederson, & Morera, 2011; Hecht et al., 2003; Kim, Ziedonis, & Chen, 2007; Marsiglia, Yabiku, Kulis, Nieri, & Lewin, 2010; Nollen et al., 2007; Ringwalt & Bliss, 2006; Wong et al., 2008; Wu et al., 2009). However, many additional programs are being developed, implemented, and disseminated without an evidence base. Moreover, there is a paucity of research comparing culturally targeted programs with similar general-audience programs to determine whether cultural targeting increases the effectiveness of programs over and above that of nontargeted programs.

LIMITATIONS OF PREVIOUS RESEARCH

Although great progress has been made in understanding the role of race, ethnicity, and culture in substance use, researchers need to shift their paradigms to keep pace with the rapidly changing world. Nearly all research about substance use and ethnicity has made the tacit assumption that ethnicity is a stable construct that is present at birth (based on skin color and national origin) and remains constant over the life course and across contexts. However, this view of ethnicity is an oversimplification.

Much of the research on ethnic differences in substance use has compared a small number of discrete ethnic categories (e.g., White, Black, Hispanic, and Asian), without considering the vast heterogeneity of people within these categories (Cheung, 1993; Trimble, 1995). Indeed, when these groups are divided into more specific categories based on national origin, there are typically large differences in substance use patterns across subgroups (An, Cochran, Mays, & McCarthy, 2008; Caraballo, Yee, Gfroerer, Pechacek, & Henson, 2006; Chen, Unger, Cruz, & Johnson, 1999; Lum, Corliss, Mays, Cochran, & Lui, 2009; Ma, Tan, Toubbeh, & Su, 2003; Siqueira & Crandall, 2008). More studies with sufficient statistical power are needed to take the additional step to subdivide the large categories (e.g., Asian or Hispanic subgroups based on national origin, or even more specific categorizations that take into account the interactions among culture of ancestry, culture of birth, and culture of residence).

In addition, few studies have focused on individuals who identify with two or more groups. Since the United States Census began allowing people to choose multiple racial and ethnic categories in 2000, there has been increasing recognition that many people belong to and identify with multiple groups when given the opportunity to note their ethnicity. The sheer number of possible combinations makes it difficult to put each individual into a convenient box and make comparisons with other individuals who fit neatly into other boxes. Research on multiethnic individuals is especially important because adolescents who identify as multiethnic, regardless of the specific ethnic groups they choose, are at increased risk of experimenting with substance use, relative to their mono-ethnic/mono-racial peers (Choi, Harachi, Gillmore, & Catalano, 2006; Unger, Palmer, Dent, Rohrbach, & Johnson, 2000; Whaley & Francis, 2006). However, it is not clear why multiethnic individuals are at higher risk for substance use. Are they attempting to fit into a peer group of substance users because they feel excluded from other peer groups? Are they attempting to self-medicate feelings of discrimination and stigma? Do they have more access to alcohol, tobacco, and other drugs because they have friends and family members in more diverse neighborhoods? Do they live in families that are more tolerant of substance use? Research is needed to understand this phenomenon.

Especially among multiethnic individuals, but also among all individuals, ethnic identity is a choice rather than an assigned category. People cannot change their ancestry or their skin color, but they can decide which aspects of their cultural identity they wish to incorporate most closely into their personal identity, their inner self, and their more external social selves as they function and adapt, daily, in a range of roles, networks, contexts, and environments. Phinney’s classic work (1996) demonstrated that ethnic identity develops over time, through a process of exploring the meanings of various identities and incorporating one or more of them into the self-concept. Constructing a stable ethnic identity is a complex process that involves multiple social influences, including peers and family, as well as their intersections with structural conditions, individual biographies, and psychologies. Having an examined ethnic identity can be a source of resilience against substance use and other harmful behaviors because ethnic pride can protect against the damaging effects of negative stereotypes and discrimination (Phinney, 1996).

The process of ethnic identity formation typically occurs during adolescence and young adulthood, during the same developmental stage when individuals are making key decisions about substance use that may change the course of their lives. When making decisions about substance use, individuals may think about whether the use of various substances is consistent with their stereotypes of the various cultural identities that they are exploring. They may use substances because they believe that substance use is normative for their cultural group (regardless of whether their perceptions are accurate), or they may challenge society’s view of their ethnic group by engaging in substance use behaviors that they perceive as nonnormative for that group (e.g., Asians attempting to distance themselves from “model minority” stereotypes). Research is needed to understand the complex processes by which ethnic identity exploration and formation influences decisions about substance use. Of course, it is important to remember that the concept of adolescence (and young adulthood) as a life stage when individuals have the luxury of experimenting with various identities before making major life decisions is primarily a phenomenon of industrialized countries. The process may be cut short if youth assume adult roles and responsibilities at a young age.

MOVING FORWARD

Some scholars have argued that we now live in a “postethnic” world, a world in which the boundaries between ethnic groups have become so blurred that they are arbitrary and irrelevant. The Millennial Generation in the United States and many other countries is the most culturally diverse generation in history. Nearly 30% of Millennials are children of immigrants, and they have grown up connected with other cultures via local diversity and the global media (Greenberg &Weber, 2008). After all the effort that has been devoted to documenting differences and similarities in substance use across ethnic groups, how will the field adjust as ethnicity is increasingly recognized as being far more complicated than putting people into a finite number of boxes? What would substance use and misuse research look like in a postethnic world?

First, it would need to move beyond simple comparisons of mutually exclusive groups. Researchers need to realize that racial, ethnic, and cultural groups overlap, and many individuals identify with multiple groups. The traditional either/or mind set needs to be changed. Conducting statistical analyses with people who do not fit neatly into mutually exclusive categories is inconvenient, but it is necessary. Instead of deleting multicultural people from the analytic sample or grouping them into a meaningless “other/multiethnic” category, researchers need to develop defensible statistical methods to analyze complex cultural data. Just as Berry (1980) revolutionized the field of acculturation research by positing that nativeculture orientation and receiving-culture orientation are independent constructs (and therefore, people can be acculturated to one, both, or neither culture), perhaps it is time to include degrees of identification with multiple racial/ethnic groups as independent constructs. For example, instead of asking respondents to put themselves into one or more categories, perhaps we could ask questions such as, “How strongly do you identify with each of the following groups?”, with a separate rating scale for each relevant group. Then, each respondent’s “Black-ness,” “Mexican-ness,” “Chinese-ness,” “Jewishness,” etc. could be included as an independent dimension predicting substance use attitudes and behaviors, and interaction terms among these identities could be included as well.

Second, it would need to focus on the meaning of an ethnic label to the individual. The meaning of identifying with a cultural group is complex. It includes feelings of pride in the group’s strengths and accomplishments, anger or embarrassment about negative stereotypes associated with the group, and expectations to behave in a certain way. So far, most research on culture and substance use has asked individuals which culture they identify with, but not what it means to the individual to be a member of that culture. Perhaps it is time to abandon the practice of assessing people’s cultural labels and assigning stereotypic traits to those labels. Instead, perhaps we should be directly assessing the phenomena that accompany ethnic identity: feelings of belongingness, cultural pride, accessible role models, family-based and community-based norms and values, experiences of discrimination, acculturative stress, etc.

Third, we need to take into account the complex interactions between an individual’s racial/ethnic/cultural identity and the cultural context in which he or she lives. Being African American in a predominantly African American neighborhood is a very different from being African American in a predominantly White neighborhood or a predominantly Hispanic neighborhood, which is very different from being an immigrant Hispanic in a traditionally African American neighborhood that is transforming into a Hispanic enclave. Research integrating microlevel and macrolevel cultural factors is needed to understand these complex influences on behaviors including substance use. Future research should determine which cultural environments increase or decrease the likelihood of substance use, misuse, and nonuse for which individuals, in what contexts roles, and during what phases in their lives.

Fourth, researchers should understand and acknowledge their own inherent biases. It is virtually impossible to conduct a study about ethnicity and substance use without coming to the task with an inherent bias. Researchers may be aiming to highlight differences across ethnic groups, highlight commonalities across ethnic groups, demonstrate that a particular group is at high risk, demonstrate that a particular group is at low risk, or document a practice that is unique to a particular group. These a priori hypotheses, goals, and philosophies will color the way the researcher approaches the research design, analyzes the data, and interprets the findings. We, as researchers, need to be aware of our own personal biases and challenge them whenever necessary.

Fifth, there is a need to translate research findings into practice or policy change that can actually help individuals. Transforming data into knowledge is only useful to the extent that the knowledge is applied to real-world problems. Work is needed at all stages of the continuum from basic research to dissemination.

Research about race, ethnicity, culture, and substance use has proliferated over the past few decades. We now have the opportunity to advance the field by moving beyond oversimplifications and developing models that more accurately represent the reality of our rapidly changing, diverse, and interconnected world.

Biography

Jennifer B. Unger, Ph.D., social psychologist, is a Professor of Preventive Medicine at the Keck School of Medicine, University of Southern California, Los Angeles, California. Her research focuses on psychological, social, and cultural risk and protective factors for substance use among adolescents. She is a member of the editorial board of Substance Use and Misuse and has published many articles.