Low-Income Asian Americans: High Levels Of Food Insecurity And Low Participation In The CalFresh Nutrition Program

Little is known about food insecurity and the extent of Supplemental Nutrition Assistance Program (SNAP) participation in the heterogeneous Asian American population. Using California Health Interview Survey data from the period 2011–20, we examined both issues among low-income Asian American adults from six origin groups: Chinese, Filipino, Japanese, Korean, South Asian, and Vietnamese. We found high and varied levels of overall food insecurity, with the highest burden among Filipino adults (40 percent). Food insecurity by severity was also heterogenous; very low food security affected 2 percent of Chinese adults but 9 percent and 10 percent of Filipino and Japanese adults, respectively. Participation in CalFresh (California-implemented SNAP) ranged from 11 percent and 12 percent among Korean and Chinese adults, respectively, to 20 percent among Vietnamese adults. Compared with English-proficient low-income Asian American adults, those with limited English proficiency were no less likely to participate in CalFresh, possibly reflecting language assistance required by California law and provided by community-based organizations. These results underscore the importance of collecting and reporting disaggregated data by Asian origin group that could inform targeted outreach and interventions.

African American/Black respondents.For race and ethnicity, except for the Hispanic/Latino category, all remaining categories are non-Hispanic/Latino.Other Asian/Multiple Asian origin groups includes Burmese, Cambodian, Hmong, Indonesian, Laotian, Malaysian, Taiwanese, Thai, other Asians not listed here or in the major six origin groups, and those indicating that they identified with 2+ Asian origin groups.Other includes American Indian/Alaskan Native; Native Hawaiian/Pacific Islander, Other race (one race); Other race (multiple races).Family size and income are treated as continuous variables in logistic regressions.For race and ethnicity in logistic regressions, Hispanic/Latino respondents were chosen as the reference group given the small proportion of White or African American respondents identifying as limited English proficiency.
Compared to the analysis with 2011-2020 data, results did not change substantially.Results that were statistically significant in the 2011-2020 data analysis but no longer statistically significant in 2011-2018 data analysis are highlighted.We did not detect any results that were not statistically significant in the 2011-2020 data analysis but then became statistically significant in the 2011-2018 data analysis.** p < 0.05, *** p < 0.01, **** p < 0.001.Notes: CHIS=California Health Interview Survey.LEP=Limited English proficiency.For race and ethnicity, except for the Hispanic/Latino category, all remaining categories are non-Hispanic/Latino.Other Asian/Multiple Asian origin groups includes Japanese (due to insufficient sample of this group that indicates LEP), Burmese, Cambodian, Hmong, Indonesian, Laotian, Malaysian, Taiwanese, Thai, other Asians not listed here or in the major six origin groups, and those indicating that they identified with 2+ Asian origin groups.Other includes American Indian/Alaskan Native; Native Hawaiian/Pacific Islander, Other race (one race); Other race (multiple races).Family size and income are treated as continuous variables in logistic regressions.For race and ethnicity in logistic regressions, Hispanic/Latino respondents were chosen as the reference group given the small proportion of White or African American respondents identifying as LEP.The model with the interaction term explores whether the association between English proficiency and CalFresh participation differs based on race and ethnicity.

African
Compared to the analysis with 2011-2020 data, results did not change substantially.Results that were statistically significant in the 2011-2020 data analysis but no longer statistically significant in 2011-2018 data analysis are highlighted.Results that were not statistically significant in the 2011-2020 data analysis but then became statistically significant in the 2011-2018 data analysis are in red.** p < 0.05, *** p < 0.01, **** p < 0.001.

Appendix B -Low-Income CHIS respondents, 2011-2020, by Asian origin group and other racial and ethnic group
Authors' analyses of the pooled 2011-2020 California Health Interview Survey data.Notes: CHIS=California Health Interview Survey.For race and ethnicity, except for the Hispanic/Latino category, all remaining categories are non-Hispanic/Latino.

Race and ethnicity * English proficiency
Source: Authors' analyses of the pooled 2011-2020 California Health Interview Survey data.Notes: CHIS=California Health Interview Survey.LEP=Limited English proficiency.For race and ethnicity, except for the Hispanic/Latino category, all remaining categories are non-Hispanic/Latino.Other Asian/Multiple Asian origin groups includes Japanese (due to insufficient sample of this group that indicates LEP), Burmese, Cambodian, Hmong, Indonesian, Laotian, Malaysian, Taiwanese, Thai, other Asians not listed here or in the major six origin groups, and those indicating that they identified with 2+ Asian origin groups.Other includes American Indian/Alaskan Native; Native Hawaiian/Pacific Islander, Other race (one race); Other race (multiple races).Family size and income are treated as continuous variables in logistic regressions.For race and ethnicity in logistic regressions, Hispanic/Latino respondents were chosen as the reference group given the small proportion of White or African American respondents identifying as LEP.The model with the interaction term explores whether the association between English proficiency and CalFresh participation differs based on race and ethnicity.**p < 0.05, *** p < 0.01, **** p < 0.001.Source: Authors' analyses of the pooled 2011-2020 California Health Interview Survey data.Notes: CHIS=California Health Interview Survey.Models were controlled for age, sex, education, employment status, family type, family size, income, and citizenship status.We did not include separate data for Japanese, White, and African American/Black respondents due to very low proportion of those indicating both limited English proficiency and CalFresh participation in each of these groups.**p < 0.05, *** p < 0.01, **** p < 0.001.Appendix I -

Odds ratios of CalFresh participation among low-income Hispanic/Latino and Asian American CHIS respondents, stratified by race and ethnicity, 2011-2020 Models stratified by race and ethnicity
: Authors' analyses of the pooled 2011-2020 California Health Interview Survey data.Notes: CHIS=California Health Interview Survey.Models were controlled for age, sex, education, employment status, family type, family size, income, and citizenship status.We did not include separate data for Japanese, White, and African American/Black respondents due to very low proportion of those indicating both limited English proficiency and CalFresh participation in each of these groups.

Low-Income CHIS respondents, 2011-2018, by Characteristics and CalFresh Participation Characteristics Total Not participating in CalFresh Participating in CalFresh Unadjusted bivariate odds of CalFresh participation Race and ethnicity (%)
: Authors' analyses of the pooled 2011-2018 California Health Interview Survey data.Notes: CHIS=California Health Interview Survey.We defined low-income respondents for this analysis as those with household income at or less than 200% of the federal poverty level.Our unweighted sample of respondents included: 1,604 Chinese, 721 Filipino, 283 Japanese, 957 Korean, 316 South Asian, and 1,603 Vietnamese respondents.In addition, the unweighted sample included 20,952 Hispanic/Latino, 21,735 non-Hispanic White, and 3,333 non-Hispanic African American/Black respondents.These respondents represented weighted California population estimates of Source

Odds ratios of CalFresh participation among low-income CHIS respondents by race and ethnicity, English proficiency, and other sociodemographic characteristics, 2011-2018 Model 1 (no interaction term) Model 2 (with race and ethnicity*English proficiency interaction term)
: CHIS=California Health Interview Survey.For race and ethnicity, except for the Hispanic/Latino category, all remaining categories are non-Hispanic/Latino.Compared to the analysis with 2011-2020 data, results did not change substantially.: Authors' analyses of the pooled 2011-2020 California Health Interview Survey data. NotesSource