Research That Matters (January 17 - 20, 2008)

Purpose: In studies focused on drinking behaviors, Asian American and Pacific Islanders (AAPI) are less likely than Blacks, Hispanics, or Whites to drink alcohol. Although considerable variation exists among subgroups of the AAPI population, most of these studies examine AAPI as a homogeneous group by aggregating them into an “Asian Pacific American” group for comparison with other racial groups. The heterogeneity of the AAPI populations accordingly challenges the general notion of homogeneity of the drinking behaviors in these groups. Most of the existing literature, however, describes the drinking behaviors of the population as an aggregated group rather than comparing individual ethnic AAPI subgroups. Using the 2001 and 2003 California Health Interview Study (CHIS), which is the largest population-based health survey in the U.S. conducted in California, this study aims to address these gaps by focusing on the six Asian subgroups and Pacific Islander group in California.

Methods: This study utilizes adult data (aged 18 or older) from both the 2001 (N=56,270) and 2003 CHIS (N=42,044), which over-sampled for Asian subpopulations. For the current study, the 2001 and 2003 CHIS adult data were combined to increase sample size, especially for the AAPIs. For socio-demographic variables, age, gender, education, employment, household income, and marriage were used in the analysis. With regard to immigration background, the level of English spoken at home and the length of years in the U.S. were measured. Current drinking was measured by asking “Have you drunk at least one time in the last 30 days preceding the survey?” Since CHIS data were weighted to Census 2000 California population estimates, SUDAAN, a software for analyzing weighted data, was utilized for data analysis. We constructed a multiple logistic regression model to examine whether self-reported English proficiency and length of years in the U.S. were associated with current drinking among AAPIs after controlling for demographic characteristics.

Results: Logistic regression analysis suggested that for aggregated AAPIs, those who were men, young (aged less than 31), not married, employed, and with higher education, income, and longer residency in the U.S. were more likely to drink than their counterparts. When examining Asian subpopulations, as seen in this aggregated AAPI group, Asian men were more likely to drink than women across all Asian subgroups. However, when it comes to other factors, inconsistent results can be found. Middle-aged South Asians were more likely to drink than those who were in younger and older groups. Marital status only accounted for Koreans and South Asians; the married were less likely to drink. Employment, education, and income were associated with drinking behaviors only for Chinese and Pacific Islanders. With regard to immigration factors, years of residency in the U.S. was only related to current drinking for the Filipino and Japanese groups.

Implications: The data provides findings that there is a wide range of drinking behaviors across AAPI groups. These findings will be discussed in relation to the development of an intervention strategy as well as implications for social work practice, policy, and research.