Socioeconomic Status and Obesity

Abstract

Obesity is a pervasive public health problem that causes debilitating complications across the life course. One opportunity for preventing the onset of obesity is to focus on its social determinants. Socioeconomic status (SES), which includes factors such as income, educational attainment, occupational prestige, and access to resources, is a key determinant of obesity. In this scoping mini-review, we summarized review articles and meta-analyses of the SES-obesity association. From the 1980s to the present, cross-sectional studies have demonstrated a persistent socioeconomic gradient in obesity in which the association is negative in developed countries and positive in developing countries. Longitudinal studies have revealed the bidirectionality of the SES-obesity association; some studies demonstrate that socioeconomic adversity precedes the onset of obesity, while others provide evidence of reverse causality. While earlier studies relied on anthropometric assessments of weight and height to define obesity, the use of modern technologies like dual-energy x-ray absorptiometry and bioelectrical impedance have demonstrated that the socioeconomic gradient in obesity is robust across multiple indicators of body composition, including direct measures of lean and fat mass. More recently, examination of mediators and moderators of the SES-obesity association have highlighted causal pathways and potential intervention targets, with a focus on health behaviors, environmental conditions, psychological factors, and biological processes. We describe current gaps in knowledge and propose opportunities for future innovation to reduce the burden of obesity and related socioeconomic disparities.

Rates of obesity and its associated conditions, such as type 2 diabetes, metabolic syndrome, and nonalcoholic fatty liver disease, are increasing globally [1-3]. Obesity predicts premature death and causes debilitating complications across the life course, including cardiovascular disease, certain types of cancers, higher rates of depression, and worse health-related quality of life [4, 5]. In addition to these health effects, the consequences can extend beyond the individual with obesity. The economic effect of obesity, both in health care expenditures and lost productivity, is high [6-8]. Furthermore, excess adiposity among adults can propagate disease in future generations via the intergenerational transmission to offspring [9, 10].

One opportunity for reducing the burden of obesity and preventing its onset is to focus on its social determinants [11, 12]. The social determinants of health (SDoH) have been described as the conditions in which people are born, grow, live, work, and age that affect their health [13]. This includes protective or harmful conditions that are often disproportionately allocated in the population in ways that systematically benefit or disadvantage some groups over others [14]. One prominent example of an SDoH in the health sciences literature is socioeconomic status (SES) because it is thought to be a key determinant of a wide range of health outcomes, including obesity, type 2 diabetes, and cardiovascular disease [15-20]. SES, broadly defined, encompasses income, educational attainment, occupational prestige, perceptions of social status, and access to opportunities and resources. Socioeconomic conditions can influence health via multiple pathways: health behaviors, access to health care, environmental exposures, physiological processes, cultural ideals of health, and psychological factors [21]. The ongoing and projected increases in income inequality, at least in some countries, may intensify the activation of these pathways, and exacerbate health inequities [22].

In a seminal review paper published in the 1980s on the topic of SES and obesity, Sobal and Stunkard summarized the direction of the SES-obesity association worldwide [23]. SES and obesity are positively associated in countries with low-income economies and inversely associated in countries with high-income economies [23]. Since then, decades of studies showed that these gradients have persisted, although most studies focused on cross-sectional associations among adult women [24-31]. Despite the persistence of this patterning, the factors that mediate and modify these associations likely have evolved over time due to major shifts in urbanization, globalization, technologies of convenience, global nutrition transitions, and cultural ideals about food, exercise, and body type [32-34]. Socioeconomic conditions are difficult to change, so understanding modifiable pathways and resilience factors that may buffer the effects of SES on obesity can help reduce health inequities. Furthermore, the methodologies to examine body fat composition and distribution have evolved over time, and early studies of SES and obesity that relied on anthropometric measures of weight and height have been repeated using more rigorous measures of adiposity that can distinguish between lean and fat mass, such as dual-energy x-ray absorptiometry, bioelectrical impedance, air-displacement plethysmography, and 3-dimensional body scanning [27, 35].

The purpose of this scoping review was to summarize previous work on the SES-obesity association from the last two decades and highlight potentially modifiable pathways and resilience factors that could stimulate the design of randomized controlled obesity-prevention trials. We focused on individual- and household-level SES and future opportunities to advance the science on obesity prevention among those in households with lower SES.

Materials and Methods

Search Strategy

To identify relevant studies examining the relationship between SES and obesity, we conducted a literature search from March to May 2024 using PubMed and EMBASE. The search strategy employed multiple combinations of the following search terms: “socioeconomic status,” “income,” “education,” “overweight,” “obesity,” “adiposity,” “body mass index,” “body composition,” and “body fat distribution” in the title or abstract. Articles were prioritized for review if they met the following criteria: (1) published in English, (2) published in the last 20 years (since 2004), (3) SES was the predictor variable and obesity or some other measure of adiposity was the outcome variable, (4) review articles or meta-analyses published in a peer-reviewed journal that followed PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), or similar, guidelines, and (5) multicountry papers, although we included country-specific reviews when they provided new insights. When no recent review papers were found, we searched for single-study original research articles. We excluded reviews of qualitative studies and experimental study designs. The “Results” section describes the findings from 19 review articles and meta-analyses published between 2007 and 2024 (Supplemental Table 1) [36]. Although a few reviews incorporated individual studies published prior to 2000, the majority of reviews included studies published after 2000, ensuring that the synthesis presented here is based primarily on recent data. We compared the bibliographies across all 19 articles to ensure that overlap was minimal and that we did not overemphasize the results from any given study.

The screening process was conducted in 2 stages. In the first stage, titles and abstracts of all identified studies were independently screened by 2 reviewers (K.A. and T.A.B.). In the second stage, the full texts of the selected articles were reviewed. Additionally, reference lists of the included studies were manually searched to identify any additional relevant articles missed during the initial database search.

Definitions of Socioeconomic Status and Obesity

The American Psychological Association defines SES as encompassing not only income but also educational attainment, occupational prestige, and subjective perceptions of social status and social class. According to the American Psychological Association definition, SES also reflects quality-of-life attributes and opportunities afforded to people within society (eg, access to health care) [37]. This definition is further expanded to encompass material capital (eg, property), human capital (eg, skills, abilities), and social capital (eg, social networks, power) as a reflection of the status of individuals in the social structure [38]. While the construct of SES is holistic and complex, it is often operationalized in the health sciences literature using concrete, one-dimensional indicators such as annual household income or years of formal education. Thus, many of the papers included in this review used the variables income and educational attainment to operationalize SES.

The World Health Organization defines obesity as a complex chronic disease characterized by excessive fat accumulation that can impair health [39]. Among adults, the most commonly used indicator is body mass index (BMI), calculated as body weight in kilograms divided by height in meters squared (kg/m²). The cutoff for adult obesity is a BMI of 30 or greater [39]. Among children, multiple weight- and height-based indicators that are sex and age specific have been used to define obesity [40]. While BMI is correlated with total body fatness, BMI is limited in that it does not distinguish between lean and fat mass or describe how fat is distributed around the body [41, 42]. Another limitation of BMI is that the association between BMI and percentage of body fat can vary by age, sex, and ethnic group [43]. There are many indicators of fat mass, fat-free mass, and body fat distribution, but these measurements are used less frequently than BMI, possibly because they are relatively expensive and burdensome to collect compared to weight and height. Thus, the majority of papers included in this review used BMI to categorize individuals as having or not having obesity. Nevertheless, the search strategy included terms to capture studies that examined more rigorous measures of body composition and body fat distribution, given their association with disease risk independent of BMI [44].

Results

Cross-sectional Studies of Socioeconomic Status and Obesity

We evaluated 10 review articles published in the last 20 years focused on cross-sectional associations between SES and obesity [15, 24, 25, 31, 45-50]. The review articles collectively incorporated multiple indicators for obesity (eg, continuous BMI or BMI category) or SES (education, income, occupation, composite index, area-level deprivation, employment status, property ownership). The study samples included children, adolescents, and adults living in countries with low-, middle-, and high-income economies, and data were collected over 3 decades from the late 1980s to 2021. Findings were generally consistent across all 10 review articles. After stratifying the findings by national income, SES and obesity were positively associated in countries with low-income economies and inversely associated in countries with high-income economies. The SES-obesity association tended to be stronger among females compared to males. In countries with middle-income economies, the SES-obesity gradient was inconsistent across articles or flat, possibly because the positive association between SES and obesity was transitioning to an inverse one as countries were experiencing economic development [47].

Longitudinal Studies of Socioeconomic Status and Obesity

The present review article conceptualized SES as the exposure variable. Cross-sectional studies have provided valuable insights, but their findings should be interpreted with caution because these study designs cannot establish a temporal relationship between SES and obesity. There are several plausible pathways by which obesity could lead to lower SES among some individuals. Examples include the possibility that individuals with obesity may enter into lower-income jobs due to labor-market discrimination or obesity contributing to high health-care expenditures that pull individuals below the poverty line, although this is certainly not the case for all individuals with obesity [51]. Longitudinal studies can establish temporality by following individuals over time and measuring SES and obesity at multiple time points. These types of studies can help identify windows of susceptibility and specific life stages (eg, in utero, early adulthood) when SES may have the greatest effect on obesity.

We identified 2 systematic reviews and meta-analyses that summarized the findings from longitudinal studies of the association between SES and obesity [51, 52]. All studies included in the reviews were conducted in high-income countries, although this was not an inclusion criterion. A 2017 meta-analysis by Newton and colleagues [52] (14 studies) focused on studies among adults that included life course SES as the exposure, operationalized with variables such as childhood family income or father's occupation. Among females, there was a consistent association between lower life course SES and higher BMI, obesity prevalence, and waist circumference. Among males, the pattern differed, with lower life course SES not associated with obesity but showing a positive association with waist circumference. A 2018 meta-analysis by Kim and von dem Knesebeck [51] focused on prospective studies of the association between income and obesity. Of the 21 studies included this review, 14 showed an inverse association between SES and subsequent obesity, while 7 provided evidence for reverse causality .

Associations Between Socioeconomic Status and Direct Measures of Body Composition

Among adults and children, there is some evidence that the association between SES and obesity may vary depending on which indicator of adiposity is used [53, 54]. Specifically, the use of BMI may overestimate or underestimate the socioeconomic gradient in obesity compared with studies that examine fat distribution or body composition [53, 54]. Two systematic reviews published in 2021 examined the association between socioeconomic status (eg, income, education, overcrowding, area-level deprivation) and direct measures of body composition (eg, using dual-energy x-ray absorptiometry or bioelectrical impedance, and not anthropometric measurements of weight and height to calculate BMI) [27, 28]. Among adults and children in countries with high-income economies, there was an inverse association between SES and fat mass, especially among females. In countries with middle-income economies, the direction of the association was reversed, so that lower SES was associated with lower fat mass. Regardless of national income or life stage, higher SES was generally associated with greater lean mass [27, 28].

Mediators of the Socioeconomic Status–Obesity Association

Two systematic reviews summarized studies examining mediators of the SES-obesity association among children in high-income countries [55, 56]. Both reviews identified evidence for a mediating role of television viewing, parent BMI, breastfeeding duration, breakfast omission, childcare attendance, and some dietary behaviors. Other potential mediators were explored (birth weight, physical activity, fruit and vegetable intake, smoking during pregnancy), but findings were inconsistent across studies, so the mediating role of these variables is inconclusive [55, 56]. Another systematic review of mediators among youth in Ireland and the United Kingdom confirmed many of these findings, and also identified evidence for the mediating role of access to green space and favorable neighborhood conditions [57]. Among adults, a 2019 systematic review of psychosocial pathways between SES and obesity in high- and middle-income countries indicated a mediating role for neighborhood quality, stress, self-esteem, expectations for the future, and early-life adversity (eg, low parental support), although the pathways differed by sex. For example, stress and social support were more important mediators among women compared to men. Job stress, social support, attitudes and motivation toward health behaviors, self-efficacy, and personality characteristics were examined as potential mediators in only a small number of studies, or findings were mixed; thus, the mediating role of these factors merits further study [58]. The aforementioned reviews focused on the mediating role of health behaviors and social/psychosocial factors. No reviews were identified for biological mediators of the SES-obesity association, but individual studies have highlighted areas where additional research may be warranted. This includes a mediating role for epigenetics (eg, DNA methylation) [59], inflammatory markers (eg, C-reactive protein) [60], markers of chronic stress [61], sex-specific biological factors (eg, sex hormones) [62], and immune system dysregulation and adipokine secretion [61], any of which could increase adipose tissue accumulation, central fat deposition, or insulin resistance [59-61, 63-65]. Testing hypotheses about biological mediation requires longitudinal studies that collect biospecimens and measures of SES and obesity at multiple time points.

Moderators of the Socioeconomic Status–Obesity Association

Given that SES itself is difficult to change, one promising strategy for reducing socioeconomic inequities in obesity is to identify modifiable resilience factors that attenuate the effects of SES on obesity. However, relatively few studies have examined moderators of the association between SES and obesity. A 2023 systematic review examined the moderating role of built environmental features on the association between residence in a low-SES neighborhood and overweight or obesity [16]. The authors concluded that there was no evidence of moderation by street connectivity, population density, the food environment, access to physical activity facilities, or several perceived environmental attributes (eg, perceived neighborhood safety). Findings were mixed for the moderating role of walkability. Advancements in assessing built environmental features, such as satellite images, have become more widely available since many of these studies were published. Future research could continue to provide insights using more precise and comprehensive approaches to measuring the built environment. Additionally, many studies have shown that the association between SES and obesity differs by sex, with steeper gradients among females vs males [15, 16, 23]. While these findings do not provide a modifiable target for intervention, these sex differences can generate hypotheses about sex-specific moderators. For example, greater access to opportunities for physical activity among males vs females may help buffer the effects of low-SES on obesity [66].

Discussion

In this paper, we summarized 19 review articles and meta-analyses about the association between SES and obesity published in the last two decades. While our review is nonexhaustive, we have summarized a range of study designs and highlighted key findings. We concluded that the patterns first identified in the 1980s have persisted worldwide, from childhood to adulthood, and across sexes, although the association tends to be stronger in females. In many, but not all cases, the findings are mostly robust regardless of which SES or obesity indicator was used [15, 54].

Opportunities for methodological advances include the need for prospective study designs in which SES, adiposity, and potential mediators and moderators are measured at multiple time points. This study design is critical for understanding the directionality of these associations by establishing temporality. Longitudinal designs that employ a life course perspective can reveal [1] life stages when actions to prevent obesity have the potential to be most effective [2], whether there are long-term effects of exposure to adverse socioeconomic conditions (eg, whether SES in early life can affect obesity in adulthood independent of adult SES), and [3] if a longer duration of exposure to adverse socioeconomic conditions is associated with a higher risk or greater severity of obesity [14, 64, 67]. One recent example is a 2022 study by Aris and colleagues [68] using data from the Environmental influences on Child Health Outcomes (ECHO) Consortium that showed neighborhood-level indices of socioeconomic adversity at birth were associated with a lower risk of obesity from childhood to adolescence . Studies that assess obesity outcomes at multiple time points are particularly valuable because growth trajectories have been linked to adverse cardiometabolic outcomes; and transitional periods from one life stage to another may be critical periods for the development of obesity [69, 70].

In addition to longitudinal studies that employ a life course perspective, research is needed with rigorous measures of body composition and body fat distribution that can overcome the limitations of BMI [27, 28]. Distinguishing between lean and fat mass, or between central and peripheral adiposity, using methods such as skinfold measurements and air displacement plethysmography, can enhance our understanding of the health effects of socioeconomic conditions and also provide insight into mechanistic pathways. Objective biomarkers of cardiovascular risk, such as serum lipids, and their association with SES also merit further study. Among the few studies that have been conducted, some show a positive association, whereas others show no association [71, 72]. The use of these types of finer-tuned measures of adiposity and metabolic health can inform the development of clinical and public health interventions through the identification of 1) individuals with central adiposity and a high body percentage whose health status may be masked by a normal BMI, and 2) metabolically healthy individuals with a BMI in the overweight range who may be a lower priority for some interventions [73].

Another opportunity for scientific innovation is to examine biological mediators of the SES-obesity association [61, 74]. In 2023, scientists at the National Institutes of Health highlighted specific biological pathways that merit further research. First, low SES may be a source of chronic stress, which activates the sympathetic nervous system and hypothalamic-pituitary-adrenal axis leading to higher levels of catecholamines and cortisol [61]. Yet, catecholamine resistance due to the stress of socioeconomic adversity, and its effect on fat accumulation over time, is not well understood. Second, little is known about the mediating role of adipokines in the SES-obesity association. Specifically, more observational studies can examine how the SDoH, including low SES, alter leptin resistance and the secretion and function of adiponectin [61]. These types of studies could build on previous work suggesting a mediating role for adipokines on the association between other environmental exposures (ie, air pollution) and risk factors for cardiovascular disease [75]. Another opportunity for scientific innovation is to identify novel environmental mediators and moderators of the SES-obesity association that are increasingly pervasive in some populations. Examples include the expanded use of social media and mobile technologies in developed economies, and increased exposure to highly processed foods, chemicals in household products, and air pollutants in developing economies due to globalization and industrialization [76-78]. Finally, many SDoH that are intertwined with SES have been linked to obesity. For example, neighborhood environments, including perceptions of safety and social cohesion, are correlated with socioeconomic status and obesity [79]. Another example is the association between the social constructs of self-identified race and ethnicity and obesity, which may be partially explained by structural racism, residential segregation, acculturation, or socioeconomic conditions [12, 80-83]. Future studies that measure multiple SDoH in the same population can help establish the independent effects of these factors on obesity. Many large-scale studies, such as the National Health and Nutrition Examination Survey (NHANES) and the National Institutes of Health ECHO Consortium are already collecting these types of data, and deidentified data sets are available for public download [84, 85]. Other opportunities for innovation include linking individual-level data on obesity-related health outcomes to publicly available data sources of upstream and midstream SDoH and related indices, many of which have been constructed by the US Census Bureau and the Centers for Disease Control and Prevention's PLACES database [86, 87]. Examples include the Structural Racism Effect Index, the Index of Concentration at the Extremes, the Social Vulnerability Index, the Child Opportunity Index, The Social Deprivation Index, the Area Deprivation Index, and the CDC's Environmental Justice Index. While many of these indices have been linked to health, their relationship with obesity and other dimensions of cardiovascular health are not well understood [68, 80, 88-97].

While intervention studies were not the focus of this review paper, we note that systematic reviews and meta-analyses published in 2023 and 2024 summarized targeted efforts to reduce obesity among youth and adults from low-SES households [98, 99]. Most intervention studies among individuals with low SES show small or nonsignificant decreases in BMI in the short term, or unsustained benefits in the long term [98, 99]. Examples of interventions that have been less effective among individuals with low SES include social support for weight loss and health education to encourage individual behavior change. In contrast, there is evidence that financial incentives, interactive feedback (ie, receiving real-time feedback on progress toward behavior change goals via mobile application or text message), tailored weight loss programs delivered by primary care providers, community-based strategies, and policies that include structural changes to the environment may be promising approaches [98-101]. In developing economies worldwide, obesity-prevention efforts are less common, likely due to the concentration of obesity among individuals with high SES, as well as the continuing need to prioritize efforts to reduce undernutrition [102]. While efforts to improve SES itself are not within the scope of public health, it should be noted that higher SES attenuates the association between higher genetic risk for obesity and obesity [103, 104].

We conclude that the steady worldwide increase in obesity (Supplemental Fig. 1) [36, 105] since the 1970s and the persistence of the socioeconomic gradient over decades reinforce the need for continued research on the association between SES and obesity [39]. Efforts to mitigate these inequities are incomplete, with few interventions demonstrating significant and sustained benefits for obesity prevention among socioeconomic groups with the highest obesity burden. Existing gaps in the literature, including understanding causal pathways and identifying factors that buffer the effects of socioeconomic conditions on obesity, provide novel opportunities for innovation.

Abbreviations

BMI

body mass index

SDoH

social determinants of health

SES

socioeconomic status

Funding

This project was unfunded.

Disclosures

The authors have no conflicts of interest to disclose.

Data Availability

Data sharing is not applicable to this article as no data sets were generated or analyzed during this study.