Obesity is a leading modifiable risk factor for cardiovascular diseases globally. It is projected that by 2030, ∼50% of American adults will be obese, and there will be significant disparities in the prevalence of obesity among demographic subgroups. Obesity disproportionately affects individuals from low socioeconomic backgrounds as well as racial and ethnic minority populations, and the prevalence of obesity rises as the burden of adverse social determinants of health (SDOH) increases.1 However, in contrast to the above projections, a recent data brief published by the U.S. Centers for Disease Control and Prevention showed no significant change in the overall prevalence rate of obesity among adults from 2013 to 2023.2 Therefore, we aim to assess for subpopulation level changes, focusing on adults aged 20 to 59 years, given that this group had the highest prevalence of obesity and that cardiovascular disease mortality is increasing most rapidly in this population.
Methods
The National Health and Nutrition Examination Survey (NHANES) is a biennial survey of children and adults in the United States, which comprises interviews, physical examination, and laboratory measurements. It is designed to generate nationally representative samples.3 For this study, we analyzed data from the 2017-2020 and 2021-2023 cycles. We included all nonpregnant adults aged 20 to 59 years. This subsample was divided into 2 groups by age, 20 to 39 and 40 to 59 years. We defined obesity as body mass index ≥30 kg/m2 and computed age-adjusted prevalence rates of obesity in the overall cohort as well as by sex, race, educational status, health insurance status, household income, and marital status. We computed a modified composite SDOH score from 4 SDOH domains, including income, marital, insurance, and educational status.1 Family income to poverty ratio (PIR) <130% of the federal poverty level, marital status of divorced/separated/widowed for individuals aged 20 to 39 years, divorced/separated/widowed or single for those aged 40 to 59 years, and having less than a college degree were each scored one, and otherwise a score of 0 was assigned for each domain. Scores in all 4 domains were summed up as composite SDOH score. Those with SDOH score 0 were those who screened negative in all 4 domains and score of 4 were those who screened positive in all 4 domains. We compared rates in the 2021 to 2023 period to the 2017 to 2020 period across sociodemographic and SDOH strata by computing the absolute percentage-point difference in prevalence rates between both periods. We defined a significant change as any increase or decrease of 5%.4 The Mobile Examination Center examination weights were applied to all analyses. Missing data were <5%. Data analysis was performed using SAS version 9.4 (SAS Institute). This study followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guideline for cross-sectional studies. Study protocols for NHANES were approved by the National Center for Health Statistics Ethics Review Board. All adult participants provided signed informed consent.
Results
In the 2017-2020 cycle, there were 5,403 respondents, corresponding to a weighted frequency of 168,792,435 and 3,292 respondents in 2021-2023 cycle with a weighted frequency of 168,837,704.
Adults aged 20 to 39 years
The prevalence rate of obesity was 39.7% in 2017 to 2020 and decreased to 35.5% in 2021 to 2023. Prevalence rates by sex, race, education, income, insurance, and marital status are presented in Figure 1. Compared to the 2017 to 2020 prevalence rate, there was a 4.2% absolute percentage-point difference decrease in obesity, though this change was not statistically significant. However, specific groups experienced significant changes. Among White individuals, the prevalence of obesity decreased by 7.4%, among males by 5.6%, and among college graduates by 7.3%. Those with private insurance had an 8.2% reduction in the prevalence of obesity. Reductions were also noted across income groups, with individuals whose household income was 130 to 350% above the federal poverty line experiencing an 8.1% decrease, and those earning more than 350% above the poverty line demonstrating a 5.1% decrease. Changes by marital status were also observed. Married individuals had a decrease of 5.3%, while those who were divorced, separated, or widowed had a 13.3% increase. Although there was no statistically significant difference in prevalence among individuals screening positive for any suboptimal SDOH domain, those without any suboptimal SDOH experienced an 8.9% decrease in obesity prevalence in 2021 to 2023 relative to 2017 to 2020.
Figure 1.
Age-Adjusted Prevalence Rates of Obesity in the Pre- and Post-Pandemic Period in the United States
Adults aged 40 to 59 years
The prevalence rate of obesity was 44.3% in 2017 to 2020 and increased to 46.4% in 2021 to 2023. Prevalence rates by sex, race, education, income, insurance, and marital status are presented in Figure 1. Relative to the 2017 to 2020 prevalence rates, there was no significant change in the 2021 to 2023 overall prevalence rate or among sociodemographic groups, except among the uninsured, where the prevalence rate increased by 11.2%. Across composite SDOH score strata, while the prevalence rate was lowest among those with no SDOH burden during both periods, we did not see any significant change in the 2021 to 2023 rates relative to 2017 to 2020 rates in all the SDOH strata.
Discussion
Overall, we found heterogeneous changes in obesity prevalence from 2021 to 2023. While there was a decrease in obesity prevalence in adults aged 20 to 39 years, this was mainly observed in those who are male, White, college educated, married, have private health insurance, or have a household income ≥130% PIR. In contrast, among adults aged 40 to 59 years, the only significant change in obesity was an increase in those who were uninsured.
While nationally aggregated statistics reported by the U.S. Centers for Disease Control and Prevention showed no significant change in the overall prevalence of obesity over the 2017 to 2023 timeframe, other data have projected an upward trend in obesity over this time in middle-aged individuals 20 to 59 years.2 Therefore, it is crucial to realize that certain subpopulations have seen marked declines in obesity prevalence while others have seen stability or an increase in rates. This analysis is important in understanding this heterogeneity to help inform targeted public health interventions.
There are several potential explanations for this heterogeneous decline in obesity rates. One such factor is the exponential increase in glucagon-like peptide-1 receptor agonist use for weight loss among young adults between 2020 and 2023, where there was an increase in prescription of these medications to males by 481%.5 However, this increase is mainly among non-Hispanic White adults and those with higher socioeconomic status. Moreover, these medications were more likely to be prescribed primarily for weight loss among young adults than older adults. Another potential explanation is the varying impacts of the COVID-19 pandemic on different subgroups regarding nutrition security and physical activity likely played a role.
This study has some limitations. First, NHANES consists of cross-sectional survey data of relatively small numbers of people; therefore, causality regarding the change in obesity rate cannot be determined. Second, the lack of data regarding physical activity, nutrition, and medication prescriptions prevents us from assessing their association with obesity prevalence. Third, the NHANES lacks data on glucagon-like peptide-1 receptor agonist prescription, which could have been beneficial in exploring its role in reducing the prevalence of obesity. Fourth, it is unclear how the effect of the COVID-19 pandemic on cardiometabolic conditions such as obesity may have contributed to this study’s findings.
Our results have important implications from a health policy and equity standpoint. This disaggregated data help understand which populations have seen a decline in obesity rates and which have not, setting the stage for a more nuanced investigation into underlying causes. While this analysis does not identify the cause, identifying groups with unchanged or worsening prevalence rates of obesity allows for a strategic and targeted deployment of proven interventions to combat obesity, including antiobesity medications, as a tool to achieve cardiovascular health equity.
Funding support and author disclosures
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Footnotes
The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the Author Center.
References
- 1.Javed Z., Haisum Maqsood M., Yahya T., et al. Race, racism, and cardiovascular health: applying a social determinants of health framework to racial/ethnic disparities in cardiovascular disease. Circ Cardiovasc Qual Outcomes. 2022;15(1) doi: 10.1161/CIRCOUTCOMES.121.007917. [DOI] [PubMed] [Google Scholar]
- 2.National Center for health S, editor national health and nutrition examination survey. 2024. Obesity and severe obesity prevalence in adults: United States, August 2021–August 2023. Hyattsville, MD. [DOI] [Google Scholar]
- 3.NHANES survey methods and analytic guidelines [Internet] wwwn.cdc.govhttps://wwwn.cdc.gov/nchs/nhanes/analyticguidelines.aspx
- 4.Tatem K.S., Romo M.L., McVeigh K.H., et al. Comparing prevalence estimates from population-based surveys to inform surveillance using electronic health records. Prev Chronic Dis. 2017;14:E44. doi: 10.5888/pcd14.160516. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lee J.M., Sharifi M., Oshman L., Griauzde D.H., Chua K.P. Dispensing of glucagon-like peptide-1 receptor agonists to adolescents and young adults, 2020-2023. JAMA. 2024;331:2041–2043. doi: 10.1001/jama.2024.7112. [DOI] [PMC free article] [PubMed] [Google Scholar]