Food Insecurity among Lesbian, Gay, Bisexual, and Transgender Adults During the COVID-19 Pandemic in the United States, July-October 2021

Data

Data on food insecurity and selected socioeconomic and demographic characteristics for LGBT and straight/heterosexual populations were derived from the six consecutive rounds of the US Census Bureau's Household Pulse Survey (HPS) conducted from July 21 through October 11, 2021.^6,21 The six independent HPS samples from Phase 3.2 included: Week 34, July 21-August 2; Week 35, August 4-16; Week 36, August 18-30; Week 37, September 1-13; Week 38, September 15-27; and Week 39, September 29-October 11.^6,21 Information on sexual orientation and gender identity was not available in the previous rounds of HPS conducted before July 2021. The HPS is a national sample household survey in which data on socioeconomic, demographic, self-assessed health, mental health, food insecurity, and health care characteristics during the COVID-19 pandemic are collected in near real-time via email and the internet. All information collected in the survey is based on self-reports. The HPS uses a systematic sample design and is representative of the civilian non- institutionalized population of the United States.^1,2,6,22

The HPS was developed as a rapid response survey in order to track the social and economic impacts of the COVID-19 pandemic on American households on a weekly or a bi-weekly basis in partnership with several federal statistical agencies: Bureau of Labor Statistics, National Center for Health Statistics, Centers for Disease Control and Prevention, Maternal and Child Health Bureau, US Department of Health and Human Services (HHS/ASPE), National Institute for Occupation Safety and Health, US Department of Agriculture's Economic Research Service, National Center for Education Statistics, Department of Housing and Urban Development, Social Security Administration, Bureau of Transportation Statistics, Energy Information Administration, and Consumer Financial Protection Bureau.^1,2,6,22 Substantive and methodological details of the HPS are available elsewhere.^1,2,6,22

Measurement of Food Insecurity and Related Indicators

We pooled 6 rounds of HPS data during July- October 2021 to ensure sufficient sample sizes for analyzing food insecurity patterns by LGBT status and other sociodemographic characteristics. Food insecurity was derived from the question "Getting enough food can also be a problem for some people. In the last 7 days, which of these statements best describes the food eaten in your household? 1) Enough of the kinds of food (Uwe) wanted to eat; 2) Enough, but not always the kinds of food (Uwe) wanted to eat; 3) Sometimes not enough to eat; 4) Often not enough to eat.” Adults with responses 3 or 4 (i.e., sometimes or often not enough to eat) were defined as those experiencing food insecurity. Adults with responses 1 or 2 (i.e., having enough to eat) were considered to be food secure. Additionally, we considered several indicators of or reasons for food insecurity (e.g., not being able to afford food, neighborhood safety concerns, lack of transportation, and health limitations), and participation in food assistance programs such as the Supplemental Nutrition Assistance Program (SNAP) and receipt of free groceries or meals.

Definition of Sexual Orientation and Gender Identity, Primary Covariates of Interest

Starting with Phase 3.2 of the HPS, respondents were asked questions about their sexual orientation and current gender identity, the primary covariates of interest in this study. Regarding their sexual orientation, respondents were asked: "Which of the following best represents how you think of yourself (1) gay or lesbian; (2) straight, that is not gay; (3) bisexual; (4) something else; (5) I don't know?” The current gender identity variable was based on the question: "Do you currently describe yourself as male, female, transgender, or none of these?” For this study and consistent with past research,²³ we combined the variables of sexual orientation and gender identity to create a composite variable, LGBT status, which consisted of 5 mutually exclusive categories: 1=lesbian or gay, 2=bisexual, 3=transgender, 4=straight/heterosexual, 5=other. In order to present aggregate data for all LGBT adults, we also defined LGBT status as a three-category variable: 1=gay or lesbian, bisexual, and transgender, 2=straight/heterosexual, 3=other. Those with missing data on LGBT status and food insecurity were excluded from the analysis, resulting in an effective/final sample size of 339,317 for analysis.

Socioeconomic and Demographic Covariates

Based on previous research and the social determinants of health framework, we selected the following covariates: age, race/ethnicity, marital status, region of residence, educational attainment, household income, housing tenure, recent household job loss, and health insurance status.^5,11-13,24 These covariates were measured as shown in Figures 1 and 2. Additionally, the prevalence of food insecurity was analyzed for the 15 largest Metropolitan Statistical Areas (MSAs) of the US that were available in the HPS (Figure 1). The other metropolitan areas were not identified in the HPS public-use file.²¹

Close

Figure 1:

Prevalence (%) of Food Insecurity among LGBT Adults ≥18 Years in 15 Largest Metropolitan Statistical Areas (MSAs), United States, July 21-October 11, 2021. Notes: Differences in prevalence of food insecurity by LGBT status were statistically significant at P<0.05 for Atlanta, Boston, Los Angeles, San Francisco Seattle, and Washington DC. Differences in food insecurity prevalence among straight adults across all MSAs were statistically significant at P<0.001. Differences in food insecurity among the “other” LGBT status category across all MSAs were not statistically significant at P<0.05. Source: Data derived from July 21-October 11, 2021 Household Pulse Survey

Export to PPT

Statistical Methods

Multivariate logistic regression was used to model disparities in food insecurity after controlling for socioeconomic, demographic, and health care characteristics. Logistic models of food insecurity were fitted for the total population, the LGBT population only, and the population consisting of the 15 largest metropolitan areas. The Chi-square statistic was used to test the overall association between each covariate and food insecurity prevalence, whereas the two-sample t test was used to test the difference in prevalence between any two groups or geographic areas. To account for the complex sample design of the HPS, SUDAAN software was used to conduct all statistical analyses, including the logistic modeling procedure RLOGIST.²⁵ Adjusted food insecurity prevalence estimates were derived by the logistic model at the mean values of the covariates.

Sociodemographic Characteristics of LGBT and Straight/Heterosexual Populations

The LGBT population represented 7.90% of the total adult population (Table 1). LGBT and straight/heterosexual populations differed significantly in their sociodemographic characteristics. Compared with the straight population, LGBT adults, as a combined group, were more likely to be younger, non-Hispanic White, single, without health insurance, and had lower education, income, and home ownership rates, and higher unemployment and job losses (Table 1). There is substantial socioeconomic and demographic heterogeneity within the LGBT population. While Lesbian/gay adults had similar education and income levels as the straight population, transgender adults were particularly disadvantaged with respect to socioeconomic status, job loss, and lack of health insurance. About 24.7% of transgender adults were without a high school diploma, compared with 6.9% of the straight population, 7.4% of bisexual adults, and 8.6% of lesbian/gay adults. About 19.2% of transgender adults had annual household incomes <$25,000, compared with 10.1% of the straight population, 12.8% of lesbian/gay adults, and 15.4% of bisexual adults. Transgender adults were approximately two times more likely to experience recent household job losses or lack health insurance, compared with the straight population and lesbian/gay adults.

Disparities in Prevalence and Odds of Food Insecurity During the Pandemic

The prevalence of food insecurity was 25.6% for transgender adults aged ≥18, significantly higher than that for lesbian/gay adults (11.6%), bisexual adults (13.5%), and straight/heterosexual adults (7.8%), and the total adult population (8.7%) (Table 2). After controlling for sociodemographic characteristics (i.e., age, race/ethnicity, marital status, region of residence, education, household income, housing tenure, household job/income loss, and health insurance status), lesbian/gay, bisexual, and transgender adults had, respectively, 36%, 35%, and 157% higher odds of food insecurity than their straight counterparts (Table 2). Other sociodemographic characteristics were associated with food insecurity. Those in younger ages, Black/African Americans, Hispanics, other and mixed-races, with lower SES, job losses, and without health insurance were significantly more likely to experience food insecurity (Table 2).

Predictors of Food Insecurity among LGBT Adults During the Pandemic

Black/AfricanAmerican, Hispanic, or mixed race/ethnicity, lower education and income, lack of home ownership, divorce/separation, lack of health insurance, and recent job loss were significant predictors of food insecurity among LGBT adults (Table 3). About 21.7% of Black/ African American LGBT adults reported experiencing food insecurity, significantly higher than non-Hispanic White LGBTs (10.1%) and Asian LGBTs (8.1%). After adjusting for covariates, LGBT adults who were Black/ African American, other and mixed races, or Hispanic had, respectively, 45%, 41%, and 31% higher adjusted odds of food insecurity than their non-Hispanic White counterparts.

LGBT adults who experienced divorce or separation had 58% higher adjusted odds of food insecurity than their married counterparts. Education and income levels were strongly and consistently linked to food insecurity in LGBT adults. Prevalence of food insecurity varied from 32.6% in LGBT adults with less than a high school education to 4.6% among LGBT adults with at least a college degree. Prevalence of food insecurity varied from 30.8% in LGBT adults with household income <$25,000 to 2.8% among LGBT adults with household income ≥$200,000 (Table 3). After adjusting for covariates, LGBT adults with less than a high education or with a high school diploma had had 4.0 and 3.1 times higher odds of food insecurity respectively than those with at least a college degree. LGBT adults with household income <$25,000 or $25,000-$34,999 had 7.4 and 4.5 times higher adjusted odds of food insecurity respectively than those with household income ≥$200,000.

About 29.2% of LGBT adults who experienced recent household job loss reported food insecurity, compared with 8.7% of LGBT adults who did not experience job loss. Controlling for covariates, LGBT adults with recent job losses had 2.8 higher odds of experiencing food insecurity than their counterparts without job loss. LGBT adults who did not have health insurance had 57% higher adjusted odds of food insecurity than those with health insurance (Table 3).

Prevalence and Odds of Food Insecurity During the Pandemic by LGBT Status in 15 Largest US Metropolitan Areas

Inequalities in food insecurity by LGBT status in the 15 largest US metropolitan areas as a whole were similar to those for the total population. The prevalence of food insecurity in metropolitan areas was 28.5% for transgender adults aged ≥18, significantly higher than that for lesbian/gay adults (10.6%), bisexual adults (10.2%), and straight/heterosexual adults (7.9%) (Table 4). After adjusting for covariates, lesbian/gay and transgender adults in large metropolitan areas had, respectively, 36% and 192% higher odds of food insecurity than their straight counterparts.

LGBT adults experienced higher levels of food insecurity compared to the straight population in several US metropolitan areas, including Atlanta, Georgia; Boston, Massachusetts; Los Angeles, California; San Francisco, California; Seattle, Washington; and Washington, DC (Figure 1). The prevalence of food insecurity among the LGBT population varied from a low of 6.9% in Washington, DC to 15.6% in San Francisco, CA. However, differences in food insecurity prevalence across the 15 US metropolitan areas were not statistically significant owing to small sample numbers of those with food insecurity in each metropolitan area. The prevalence of food insecurity among the straight population ranged from a low of 4.9% in Seattle, Washington to 11.9% in Miami, Florida, with differences across the metro areas being statistically significant (Figure 1).

Reasons for Food Insecurity and Receipt of Food Assistance During the Pandemic According to LGBT Status

Not being able to afford to buy more food was the primary reason for both LGBT and straight populations reporting food insecurity. Compared to the straight population (58.1%), 69.1% of lesbian/gay adults and 67.4% of bisexual adults reported this specific reason (Figure 2). The second important reason specified for food security was not being able to get to the store to buy food because the respondent did not have transportation, had mobility, or health limitations. Transgender (20.1%) and bisexual adults (17.7%) were about twice as likely to report this specific condition as straight adults (9.4%) as the reason for food insecurity. The third important reason given for food insecurity was not being able to go to the store due to safety concerns and delivery of groceries not available to the neighborhood of residence. This was a significantly greater concern among transgender and bisexual adults compared to straight adults.

Close

Figure 2:

Reasons for Food Insecurity, and Food Assistance among LGBT and Straight/Heterosexual Adults ≥18 Years, United States, July 21-October 11, 202. Notes: Differences in each reported reason for food insecurity by LGBT status were statistically significant at P<0.001. Differences in the percentage of households receiving free groceries or meals or SNAP by LGBT status were statistically significant at P<0.001. Source: Data derived from July 21-October 11, 2021 Household Pulse Survey.

Export to PPT

Transgender and bisexual adults with food insecurity were more likely to report receiving free groceries or free meals than lesbian/gay and straight adults. About 33.3% of the LGBT population that experienced food insecurity reported their households receiving SNAP benefits, compared with 13.6% of the food-secure LGBT population (Figure 2). Reported participation in the SNAP program varied within the LGBT population, from a high of 36.3% of food-insecure bisexual adults to a low of 15.6% of food-insecure transgender adults. Participation in the SNAP program was lower among food-insecure and food-secure straight adults compared with their LGBT counterparts.

Limitations

This study has several limitations. Estimates of food insecurity prevalence from the HPS are based on self-reports, which may not accurately estimate the actual prevalence of food insecurity among various sociodemographic groups, including LGBT adults, when compared with direct measures such as food calorie intake and food expenditure.³⁰ Singleitem measures of food insecurity used in our study may also underestimate the prevalence based on more comprehensive, multi-item measures used by USDA.^5,11,31

Second, the LGBT status, defined in our study, includes a residual “other” category that represents 3.9% of the total adult population and that has high levels of social disadvantage and food insecurity - almost similar to those observed for transgender adults. Specific sexual orientation or gender identity of this “other” group (such as queer, or questioning (LGBTQ+) individuals) remains unknown. Third, immigrant/citizenship status, one of the important factors associated with food insecurity in the US, is not available in HPS.²⁹ Fourth, our definition of food insecurity is dichotomous and does not consider a broader, more detailed definition of food insecurity (such as low food security and very low food security) due to lack of such data in HPS. Future research may consider the four-level food insecurity variable in HPS to determine how various LGBT groups might differ from their straight/heterosexual counterparts across the food-security/insecurity continuum.

Fifth, the reference period for defining food insecurity during the past 7 days, used in HPS, may underestimate the level of food insecurity in comparison to the reference periods (such as the past month or past year) used to define food insecurity in other federal surveys such as the Current Population Survey (CPS) or the Behavioral Risk Factor Surveillance System (BRFSS), although the temporary or variable nature of food insecurity may persist over time.⁵ For example, in 2020, the USDA prevalence estimates for household food insecurity based on annual and 30-day periods in the CPS were 10.5% and 5.7%.¹¹ However, HPS, being the first federal survey to collect current food insecurity using the 7-day reference period prior to the survey,⁵ may have an advantage over other surveys for tracking fluctuations in current food insecurity during the pandemic using weekly or biweekly surveys.

Sixth, because of the small sample size for the LGBT population, we were unable to examine if patterns of food insecurity among LGBT varied across racial/ethnic and socioeconomic groups. This needs to be examined in future studies. Finally, exclusion of the institutionalized population or homeless individuals from HPS may have underestimated the level of food insecurity reported in our study.

Conflicts of Interest:

The authors declare that they have no conflict of interest.

Financial Disclosure:

None to report.

Ethical approval:

No IRB approval was required for this study, which is based on the secondary analysis of a public- use federal database.

Disclaimer:

The views expressed are the authors' and not necessarily those of their institutions.