Abstract
Food insecurity has been identified as an understudied risk factor for weight-based stigma and discrimination. This study leverages data from 1,904 working-age adults to examine the food insecurity–weight-based discrimination relationship, as well as tests gender differences for this association. Logistic regression models revealed that food insecurity was significantly associated with higher odds of reporting weight-based discrimination in both unadjusted and adjusted models. Importantly, this association was only significant in the female sample. Contrary to expectations, body mass index (BMI) did not moderate this relationship among women. These findings highlight the gendered nature of the food insecurity–weight discrimination link, underscoring the need for targeted interventions and policies addressing food insecurity and weight stigma, particularly among women. Specifically, public health practitioners should consider screening for both food insecurity and experiences of weight discrimination, especially when working with female patients. Furthermore, policymakers should consider the interconnected nature of food insecurity and weight discrimination when designing programs to address either issue. Finally, future research should explore the mechanisms underlying these associations and investigate potential interventions to mitigate the negative impacts of food insecurity and weight discrimination on health and well-being.
Introduction
Food insecurity, defined as “limited or uncertain availability of nutritionally adequate or safe foods, or limited or uncertain ability to acquire foods in socially acceptable ways” (Bickel et al., 2000, p. 6), has been recognized as a critical component of economic hardship with important health implications. It has been estimated that about 12% (or 17 million) of American households experienced food insecurity in 2022 (Rabbitt et al., 2023); however, due to economic and supply chain disruptions, this proportion reached about 35% during the early months of COVID-19 pandemic in 2020 (Kim-Mozeleski et al., 2023).
A substantial body of evidence illustrates the extensive contributions of food insecurity to poor physical, mental, and behavioral health outcomes (Graham & Ciciurkaite, 2023; Holben & Pheley, 2006; Laraia, 2013; Myers, 2020; Olson, 1999; Seligman et al., 2010; Vozoris & Tarasuk, 2003). A less well-studied and understood health harm associated with food insecurity is weight stigma and discrimination, which have been shown to have detrimental effects to individual health and well-being (Becker et al., 2017, 2021; Gastón-Panthaki et al., 2023). Specifically, weight stigma and discrimination are associated with greater distress, anxiety, depressive symptoms, body dissatisfaction, poor self-esteem, suicidality, reduced academic performance, and lower health care utilization (Pearl, 2018; Puhl & King, 2013; Robinson et al., 2017; Sutin et al., 2016).
Although the extant scholarship is scant, evidence does suggest that individuals who experience food insecurity also report greater internalized weight stigma and weight-based discrimination (Becker et al., 2017, 2021; Gastón-Panthaki et al., 2023). This research, however, lacks theoretical explanations for the observed association and has only been conducted in small samples of specific populations (e.g., food pantry clients or bariatric surgery patients) without controlling for individual body mass index (BMI) or using the validated USDA instrument to measure food insecurity. While the association between food insecurity and weight discrimination may be bidirectional (i.e., weight discrimination may lead to food insecurity via a downward economic pathway (Fikkan & Rothblum, 2012), we further discuss several plausible mechanisms that explain why food insecurity may increase the risk of weight-based discrimination.
First, researchers have demonstrated that individuals who experience food insecurity often rely on low-cost, energy-dense, and highly processed foods, and there is ample evidence to show that poor dietary quality is related to higher body weight and risk of chronic diseases (Jardim et al., 2021; Leung et al., 2014). Furthermore, research indicates that greater food insecurity is associated with cycles of “feast-and-famine” due to unreliable access to adequate food supplies, as well as higher levels of overeating, binge-eating, and overall eating disorder (ED) pathology (Becker et al., 2017, 2019; Rasmusson et al., 2019). As an illustration, participants in the classic Minnesota Starvation Study (Keys et al., 1950) did exhibit disordered eating behaviors, that is, binge-eating, sense of loss of control, overeating, and preoccupation with food during the starvation and refeeding phases. In a separate strand of research, fluctuations in food availability and unstable eating patterns have been found to disrupt metabolism and contribute to weight gain over time (Hazzard et al., 2021). In addition, food insecurity has been recognized as a significant chronic strain, and chronic stress is associated with increased cortisol levels, which may also lead to weight gain (Ciciurkaite & Brown, 2021; Myers, 2020; Scott et al., 2012). Overall, these findings suggest that individuals who experience food insecurity may also experience weight-based discrimination due to their increased risk for weight gain and obesity, although the association between food insecurity and obesity has been found to be more robust among women relative to men (see Carvajal-Aldaz et al., 2022; Dinour et al., 2007 for review). Finally, anti-obesity messaging and campaigns largely target behaviors such as unhealthy eating and physical inactivity that disproportionately affect low-income and marginalized populations, which may unintentionally reinforce stereotypes linking poverty with poor health habits and obesity, leading to greater stigmatization (Swami et al., 2023).
Furthermore, considering the unequivocal association between food insecurity and obesity among women (Carvajal-Aldaz et al., 2022; Dinour et al., 2007), we anticipate that gender might also shape the association between food insecurity and weight discrimination, which has not been considered in extant scholarship. This is an important limitation for several reasons. First, rates of food insecurity are generally higher among women relative to men. Women, particularly those who are heads of the households and are caring for children, are particularly vulnerable to food insecurity potentially due to lower wages and limited access to economic and social resources (Rabbitt et al., 2023). In addition, food insecurity is a more significant stressor for women relative to men due to gender roles and expectations regarding meal planning and provision (Adams et al., 2003; Lyons et al., 2008; Wilde & Peterman, 2006). Specifically, researchers have found that women in households with limited economic resources tend to prioritize their family members’ nutritional needs over their own by skipping meals, waiting to eat, or replacing their meals with highly processed foods, all of which may lead to unintentional weight gain (Martin & Lippert, 2012; Stevens, 2010). Finally, women’s appearance and bodies are scrutinized more than men’s based on consistent evidence that women experience a greater degree of weight discrimination and have lower levels of BMI, on average (Dutton et al., 2014). As such, food insecurity may be particularly salient for women with respect to reports of weight discrimination, and this association may be further moderated by individual body weight.
Given these considerations, the current study uses data from a community sample of working-age adults in the Intermountain West (N = 1,904) to test the following hypotheses:
Method
Data
This study drew data from a community survey of residents in the Intermountain West region (Colorado, Utah, Idaho, and Wyoming), focusing on health outcomes related to food insecurity during the COVID-19 pandemic. The research specifically examined experiences of individuals with and without self-reported disabilities. Qualtrics, a leading online research platform, managed participant selection, recruitment, and data collection. Qualtrics employs a matching process that utilizes detailed participant profiles to identify and randomly select appropriate respondents. This approach ensures diverse representation while preventing duplicate survey participation. The sampling methodology maintained a balanced representation, recruiting nearly equal numbers of participants with and without disabilities. To be eligible, participants had to be at least 18 years old and reside in the Intermountain West. The disability screening drew from comprehensive categories adapted from the Individuals with Disabilities Education Act, including a wide range of conditions such as autism, developmental disabilities, psychiatric conditions, intellectual disabilities, learning disabilities, sensory impairments, physical mobility challenges, chronic illnesses, and traumatic brain injuries. Participants completed an online survey consisting primarily of multiple-choice questions, with an average completion time of 20 minutes. All research procedures were conducted in accordance with the Utah State University Institutional Review Board’s requirements, ensuring ethical standards of informed consent and participant protection. Data collection occurred in July 2020.
Measures
Summary statistics for key study variables are presented in Table 1. Our analyses consider the main outcome of self-reported weight discrimination. Our focal independent variable is food insecurity.
Descriptive Statistics of the Study Sample.
Note. p-values established based on chi-square test for categorical variables and two-sample t test for continuous variables. GED = general educational development
p < .05. **p < .01. ***p < .001.
Weight discrimination is measured as a binary variable based on responses to the everyday discrimination scale (Williams et al., 2008). Survey respondents were asked about the frequency of relatively minor forms of discrimination (10 items) in their everyday life. Specifically, respondents were asked how often they are (1) called names or insulted; (2) threatened or harassed; that, compared with other people, they are (3) treated with less courtesy; (4) treated with less respect; and (5) received poorer service at restaurants or stores; and that people act as if (6) they are afraid of you; (7) they think they are better than you are; (8) they think you are not smart; (9) they think you are dishonest; and (10) you are followed around in stores. Respondents who reported any discriminatory experiences were also asked about the reasons for these instances. Those individuals who marked “body weight” as the reason were counted as having experienced self-reported weight discrimination.
Food Security
The 10-item U.S. Adult Food Security Survey Module was used to assess food insecurity status (Bickel et al., 2000). As per established criteria, respondents were categorized into those with high food security (0 affirmative responses); marginal food security (1–2 affirmative responses), low food security (3–5 affirmative responses), and very low food security (6 or more affirmative responses). For reporting consistency and ease of interpretation, the variable was then recoded into a binary measure, where 1 = food insecurity and 0 = food security.
All analyses controlled for age, gender, race/ethnicity, marital status, education, household income, and disability status. Age was measured as a continuous variable ranging from 18 to 64. Gender was measured as a categorical variable, and includes females, nonbinary individuals, and males (reference category). Race/ethnicity included non-Hispanic White, non-Hispanic Black, Hispanic, Asian, and other adults. Due to small cell issues, when conducting bivariate and multivariate analyses, this variable was binarized into White and Other (reference category). Marital status was indexed with three categories of married or living with a partner; divorced, widowed, or separated; and never married (reference category). Education was measured as: less than high school (reference category), high school or general educational development (GED), some college, and college or more. Household income was measured with five categories: less than US$25,000 (reference category), US$25,000 to US$44,999, US$45,000 to US$64,999, US$65,000 to US$85,000, and more than US$85,000. Individual BMI was calculated based on self-reported weight and height. Finally, we considered disability status as a control variable because individuals with disabilities were oversampled in this study. Specifically, to measure disability status, respondents were asked whether they had any of the following diagnoses: autism; developmental disability; psychiatric or emotional disability; intellectual disability; learning disability; speech/language disability; hard of hearing or deaf; blindness or low vision; physical disability requiring a mobility assistive device; chronic/long-term illness; traumatic brain injury; or other. Those with affirmative responses were coded as 1 and those with no diagnoses were coded as 0 (reference category).
Analyses
We first estimated descriptive statistics for all study variables for the overall sample and for subsamples of men and women. Between-group significance was established using a two-sample t test for continuous variables and a chi-square test for categorical variables. These estimates help to contextualize study findings. Then, we estimated the association between food insecurity and self-reported weight discrimination using logistic regression models. First, we examined the bivariate relationship between food insecurity and weight discrimination in the full sample (Model 1). Then, in Model 2, we included the previously mentioned control variable that may confound the association. Furthermore, due to established gender-based group differences in weight-based discrimination, Models 3 and 4 estimated the adjusted among men and women separately. Finally, we tested whether BMI moderated the association between food insecurity and weight discrimination in the female sample (Model 5). However, because the results were not significant, they are not presented in this article.
Results
Descriptive statistics of the main study variables for the full sample as well as subsamples by gender are presented in Table 1. Notably, female relative to male respondents report significantly more experiences of weight-based discrimination (23.65% vs. 14.04%, respectively, p < .001). Furthermore, the rates of food insecurity do not significantly vary between the two subsamples (40.85% vs. 38.79%).
Table 2 presents logistic regression models predicting the odds of experiencing weight-based discrimination as a function of food insecurity. As presented in Model 1, the unadjusted model, experiencing food insecurity is associated with 1.57 times higher odds of reporting weight discrimination (p < 001). In Model 2, the adjusted model, the association between food insecurity and weight discrimination remains significant, although the coefficient size decreases to 1.39, suggesting that part of the observed association is accounted for by the demographic and socioeconomic characteristics, as well as BMI. Furthermore, Models 3 and 4 present results from the adjusted logistic regression analyses, stratified by gender. Notably, the significant association between food insecurity and weight discrimination was only observed in the female sample (odds ratio [OR] = 1.59, p < .001). We also conducted sensitivity analyses testing for an interaction effect between gender and food insecurity in the full sample. Because p values are not reliable when testing nonlinear interaction effects, we computed average marginal effects of food insecurity on weight discrimination by gender and tested their equality (Mize, 2019). We established a significant difference in the predicted probability of experiencing weight-based discrimination associated with food insecurity between men and women (Average Marginal Effect(male) = .16, p < .001, Average Marginal Effect(female) = .27, p < .001, contrast = .10, p < .001). The results from moderation analysis are visually presented in Figure 1. Finally, in Model 5 (results not presented), we tested whether BMI moderated the association between food security and weight discrimination in the female sample; however, the results were not significant.
Binary Logistic Regression of Weight-Based Discrimination on Food Insecurity, Sociodemographic and Socioeconomic Characteristics, and BMI.
Note. Reference categories are food security; non-White; male; never married; less than high school; less than US$25,000; no self-reported disability. BMI = body mass index; OR = odds ratio.
p < .05. **p < .01. ***p < .001.
Predicted Probability of Experiencing Weight Discrimination Associated With Food Insecurity by Gender.
Discussion
Extensive research has explored the association between food insecurity and body weight while also examining how anti-obesity messaging can potentially stigmatize low-income communities. Building on this scholarship, our study, first, aimed to investigate the connection between food insecurity and self-reported weight discrimination. Moreover, given the well-established link between food insecurity and obesity among women, and the pervasive societal scrutiny of women’s bodies, we specifically examined the role of gender and BMI for the food insecurity–weight discrimination association. Utilizing data from a community survey collected in 2020, we first documented a statistically significant association between experience of food insecurity and self-reported weight discrimination in the unadjusted and adjusted models for the full study sample (Hypothesis 1). Second, after examining this association in gender-stratified subsamples, we found that food insecurity was associated with greater odds of experiencing weight-based discrimination in the women sample only (Hypothesis 2). Our finding that food insecurity may be a risk factor for weight-based stigma or discrimination is largely corroborated by extant scholarship; however, previous research has been conducted primarily in clinical or small homogeneous samples of socioeconomically marginalized individuals where data on BMI were not collected (Becker et al., 2017, 2021; Gastón-Panthaki et al., 2023), limiting their ability to model variation and properly adjust for confounders.
To our knowledge, our study is the first to demonstrate that this association is particularly significant for women. Prior research has offered possible explanations for the gender differences in the effects of food insecurity, specifically pointing to the greater prevalence of obesity among food-insecure mothers (Martin & Lippert, 2012). The gendered nature of this relationship could also be driven by the higher prevalence of eating disorders and disordered eating behaviors among food-insecure women relative to men (Hazzard et al., 2020). This is significant as such behaviors are associated with increased depressive symptoms and suicidality (Kim & Kim, 2009; Siegel, 2002). However, further research is needed to understand the effects of eating disorders and disordered eating behaviors on reports of weight-based discrimination among food-insecure populations.
Regarding the lack of significance for the moderating role of BMI in the food insecurity–weight-based discrimination association, we would like to consider several possible explanations. First, there is evidence that experiences of weight-based discrimination and stigma are prevalent among individuals of diverse body weights (Becker et al., 2021; Prunty et al., 2020). Second, the association between food insecurity severity and individual body weight does not seem to be linear, such that individuals in mild to moderately food-insecure groups generally report the highest body weight (Nettle et al., 2017). This could potentially be attributed to the cyclical food consumption patterns that have been documented among individuals who experience food insecurity, such that when food is available, individuals tend to consume calorie-dense diets or exhibit binge-eating behavior (Goode et al., 2021). Because food insecurity tends to be episodic rather than chronic, the relative effect of worrying about food running out or having to consume a less desirable diet vis-à-vis having to skip meals on individual body weight and one’s perceptions about body weight remains unclear. These nuances highlight the need for additional investigations regarding the intricacies of the effect that food experiences of anxieties about food running out or actual food shortage may have on individual body weight.
Several study limitations merit comment. First, due to the cross-sectional nature of our data, we could not determine causality. Future studies should collect longitudinal data to more clearly establish the causal relationship between food insecurity and weight-based discrimination. In addition, our use of an online quota-based sample that overrepresented individuals with disabilities limits generalizability of our findings to the general U.S. population. Likewise, the subsample of nondisabled participants does not reflect the demographic characteristics, such as the racial/ethnic composition, of the broader population either. Moreover, as all data were collected exclusively in the Intermountain West, our findings may not be applicable to other regions of the United States.
Limitations notwithstanding, research results have far-reaching implications for both research and practice. These results open new avenues for inquiry while also suggesting practical strategies for addressing these interconnected issues. With respect to research, our findings underscore the critical need for more gender-focused studies exploring the complex relationship between food insecurity and weight discrimination. Future investigations should delve deeper into the underlying mechanisms that make women more susceptible to experiencing weight discrimination in the context of food insecurity. Researchers should explore the psychosocial processes linking food insecurity to weight discrimination, such as internalized weight stigma, body image concerns, and disordered eating behaviors. The development and evaluation of interventions aimed at reducing both food insecurity and weight discrimination, particularly among women, should also be a priority for future research.
Turning to practice, our findings have significant implications for health care providers and public health practitioners. There is a clear need to incorporate screening for both food insecurity and experiences of weight discrimination into routine patient assessments, especially for female patients. On a broader scale, our findings call for the development and implementation of community-based programs that promote food security while also fostering body positivity and challenging weight stigma. There is also a need for policies that not only address food insecurity but also protect individuals from weight-based discrimination, particularly in employment and health care settings. Public health professionals and researchers should work together to advocate for these comprehensive policies. Finally, our findings highlight the importance of interdisciplinary collaboration. Nutrition experts, mental health professionals, and public health practitioners should work together to develop comprehensive approaches to addressing food insecurity and weight discrimination. This collaborative approach can lead to more effective strategies that address the complex, interrelated nature of these issues.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by funding from the Utah Agricultural Experiment Station, Utah State University.
