Obesity and severe obesity prevalence among youth aged 2–17 in Minnesota, 2020–2025

Abstract

Objective

Existing sources for monitoring pediatric obesity prevalence are limited by delayed data availability and insufficient granularity for small-area estimates. We used data from the Minnesota Electronic Health Record Consortium's Health Trends Across Communities (HTAC) project to estimate obesity and severe obesity prevalence among Minnesota youth.

Methods

In serial cross-sectional analyses, HTAC data were obtained from 11 health systems in Minnesota. Patients aged 2–17 years with a valid body mass index (BMI) measurement at a participating health system from 2020 to 2025 were included. Obesity was defined as BMI ≥95th percentile for age and sex, and severe obesity as BMI ≥120% of the 95th percentile. Results are reported as prevalence estimates stratified by sociodemographic characteristics.

Results

Among 486,744 youth with BMI measured in 2025, 80,360 (16.5%, 95% CI: 16.4, 16.6) had obesity and 25,776 (5.3%, 95% CI: 5.2, 5.4) had severe obesity. Prevalence was higher among youth 12–17 years, males, and youth who were American Indian, Black, or Hispanic, and among those living in small towns or rural communities and in areas of high social vulnerability.

Conclusions

HTAC data provide timely pediatric obesity and severe obesity prevalence estimates in Minnesota and may serve as a model for other regions.

1. Introduction

Prevalence of obesity and severe obesity in children and adolescents has increased dramatically in recent decades (Noiman et al., 2026). Obesity that develops during childhood or adolescence is associated with the co-occurrence of hypertension, type 2 diabetes, and other chronic health conditions (Hoe et al., 2024; Munte et al., 2025; Parker et al., 2016). Longitudinal cohort studies have demonstrated that childhood obesity is associated with long-term sequelae, conferring increased risk for cardiovascular morbidity and mortality during adulthood (Jacobs Jr. et al., 2022).

In the United States, pediatric obesity prevalence estimates come primarily from the National Health and Nutrition Examination Survey (NHANES). From 2021–2023, based on NHANES data, 21.1% of children and adolescents ages 2–19 years in the United States were obese and 7% had severe obesity (Emmerich et al., 2025). Further, recent studies have found that the prevalence of pediatric obesity increased during the COVID-19 pandemic (Emmerich et al., 2025; Munte et al., 2025; Hu et al., 2025). While NHANES applies rigorous methods for sampling and measurement of height and weight among those consenting to participate, limitations should also be noted including declining participation rates, delayed data availability, and insufficient sample size for local-area estimates (Emmerich et al., 2025).

State or regional health data networks may offer a pragmatic approach for surveillance of pediatric obesity and other health conditions, providing population-level estimates and numbers sufficient for stratification of prevalence estimates by selected sociodemographic variables and for smaller geographic areas. In contrast, single or multi-site health system collaborations describe the prevalence of pediatric obesity in the population receiving care in the health system(s) and may differ from the health status of the underlying population.

Using data from the Minnesota Electronic Health Record Consortium's (MNEHRC) Health Trends Across Communities (HTAC) project, (Minnesota Electronic Health Record Consortium, 2026) we describe obesity and severe obesity prevalence among youth aged 2–17 years with at least one body mass index (BMI) measurement from 2020 through 2025, and explore prevalence estimates by race/ethnicity, age group, and by social vulnerability and rurality at the census tract level.

2. Methods

2.1. Study design and population

Formed in 2020, the MNEHRC is a federated network of the 11 largest health systems in Minnesota and includes data from over 90% of residents in the state (Johnson et al., 2026). All health systems participating in the MNEHRC have mapped their electronic health record (EHR) data to the Observational Medical Outcomes Partnership (OMOP) common data model. The MNEHRC utilizes a privacy-preserving, distributed approach, with only summary data shared between sites and reported through HTAC. HTAC includes a public-facing dashboard that leverages MNEHRC data to provide prevalence estimates for an increasing number of physical and mental health conditions (Zellmer et al., 2025). The pediatric obesity and severe obesity measures were developed and added to the HTAC dashboard in March 2025. The data we present were updated in March 2026. The HTAC project was reviewed by the Institutional Review Boards of the data-contributing sites and was determined to be exempt from further review and approval in accordance with regulations governing public health surveillance.

Youth 2–17 years with Minnesota residence, a visit to a participating MNEHRC health system, and at least one BMI measurement during a care visit and recorded in the EHR were included in the pediatric obesity and severe obesity HTAC measures. Youth were included in the HTAC dashboard for each year, 2020–2025, in which they had a valid BMI measurement. For those with multiple valid BMI measurements in a single calendar year, the last measurement for that year was included. Data from participating health systems were de-duplicated for the study period using a privacy-preserving hashing algorithm to ensure BMI observations reflected unique patients (Johnson et al., 2026). Patients who received care in multiple MNEHRC health systems during the study period were attributed to the health system where they received the greatest volume of care in the prior 12 to 36 months.

2.2. Measures

Following the American Academy of Pediatrics guidelines (Hampl et al., 2023), pediatric obesity was defined as a BMI ≥95th percentile and severe obesity was defined as a BMI ≥120% of the 95th percentile (Emmerich et al., 2025; Liu et al., 2025). BMI percentile was calculated based on the United States Centers for Disease Control and Prevention (CDC) Extended BMI-for-age growth charts (Centers for Disease Control and Prevention, 2025). The CDC Extended BMI-for-age growth charts were also used to identify and exclude biologically implausible values (BMI z scores < −4 or > 4).

EHR-derived covariates included age, sex, and race/ethnicity. Race/ethnicity categories were harmonized across participating health systems. Patient home address was used to assign each patient to a census tract for social vulnerability index (SVI) and rurality classification. An overall percentile ranking of SVI from 0 to 1 was assigned to each census tract, with 0 representing the lowest vulnerability and 1 representing the highest vulnerability. Low vulnerability was defined as the lowest quartile; mid-low vulnerability as the second quartile; mid-high vulnerability as the third quartile; and high vulnerability as the highest quartile (Centers for Disease Control and Prevention/Agency for Toxic Substances and Disease Registry, 2020). Rurality was defined using 2020 rural–urban commuting area (RUCA) codes (United Stated Department of Agriculture, 2025) and the percent of individuals living in rural areas from the United States Census American Community Survey (Winkelman et al., 2022; United States Census Bureau, 2023).

2.3. Statistical analysis

All analyses were descriptive; no hypothesis testing or regression modeling was performed. We calculated pediatric obesity and severe obesity prevalence estimates by year and stratified by age group. For youth with BMI measured in 2025, we present crude prevalence estimates and 95% confidence intervals (CI) by sex, race/ethnicity, SVI quartile, and RUCA categorization. All analyses were conducted using RStudio, version 4.4.1 (R Project for Statistical Computing).

3. Results

Annual cross-sectional cohorts of youth 2–17 years with BMI measurements available in HTAC ranged in size from 376,157 in 2020 to 486,744 in 2025, and the demographic and geographic characteristics of the cohorts remained broadly stable throughout the study period (Supplementary Tables 1 and 2). Among 486,744 youth with a BMI measurement in 2025, 80,360 (16.5%, 95% CI: 16.4, 16.6) met criteria for obesity and 25,776 (5.3%, 95% CI: 5.2, 5.4) met criteria for severe obesity (Table 1). Pediatric obesity and severe obesity prevalence, respectively, was higher in youth who were male (17.6%, 95% CI: 17.4, 17.7; and 5.8%, 95% CI: 5.7, 5.9), American Indian (32.9%, 95% CI: 31.6, 34.3; and 14.2%, 95% CI: 13.2, 15.2), Black (20.3%, 95% CI: 20.0, 20.6; and 6.6%, 95% CI: 6.4, 6.8), or Hispanic (25.7%, 95% CI: 25.3, 26.1; and 9.1%, 95% CI: 8.9, 9.4). Obesity and severe obesity prevalence was higher among youth living in small towns (20.5%, 95% CI: 20.1, 20.9; and 7.0%, 95% CI: 6.8, 7.3) or rural communities (18.6%, 95% CI: 18.3, 18.9; and 6.3%, 95% CI: 6.1, 6.5), while obesity and severe obesity prevalence was lower among youth living in exurban (13.6%, 95% CI: 13.3, 13.9; and 4.0%, 95% CI: 3.9, 4.2) and urban communities (16.0%, 95% CI: 15.9, 16.2; and 5.1%, 95% CI: 5.0, 5.2). Obesity and severe obesity prevalence was higher among those living in mid-high vulnerability (18.0%, 95% CI: 17.7, 18.2; and 6.0%, 95% CI: 5.8, 6.1) or high vulnerability communities (21.2%, 95% CI: 21.0, 21.5; and 7.4%, 95% CI: 7.2, 7.5) and lower among those living in low vulnerability (12.4%, 95% CI: 12.2, 12.6; and 3.5%, 95% CI: 3.4, 3.6) and mid-low vulnerability communities (15.1%, 94% CI: 14.9, 15.3; and 4.6%, 95% CI: 4.5, 4.7) (Table 1).

Table 1.

Obesity and severe obesity prevalence among Minnesota youth by sociodemographic characteristics, 2025.

		BMI available (n)	Obesity¶ (n)	Obesity prevalence, % (95% CI)	Severe obesity¶¶ (n)	Severe obesity prevalence, %¶(95% CI)
	Overall	486,744	80,360	16.5 (16.4, 16.6)	25,776	5.3 (5.2, 5.4)
Age (years)	2–5	118,156	13,197	11.2 (11.0, 11.3)	1724	1.5 (1.4, 1.5)
	6–11	177,895	30,504	17.1 (17.0, 17.3)	9906	5.6 (5.5, 5.7)
	12–17	190,693	36,659	19.2 (19.0, 19.4)	14,146	7.4 (7.3, 7.5)
Sex	Female	239,500	36,901	15.4 (15.3, 15.6)	11,368	4.7 (4.7, 4.8)
	Male	247,244	43,459	17.6 (17.4, 17.7)	14,408	5.8 (5.7, 5.9)
Race/Ethnicity	American Indian/Alaska Native	4804	1581	32.9 (31.6, 34.3)	682	14.2 (13.2, 15.2)
	Asian/Pacific Islander	36,594	6803	18.6 (18.2, 19.0)	2181	6.0 (5.7, 6.2)
	Black/African American	67,850	13,780	20.3 (20.0, 20.6)	4489	6.6 (6.4, 6.8)
	Hispanic/Latino	44,214	11,347	25.7 (25.3, 26.1)	4041	9.1 (8.9, 9.4)
	Multiracial	14,586	2603	17.8 (17.2, 18.5)	886	6.1 (5.7, 6.5)
	Other/Unknown/Missing	19,191	3103	16.2 (15.7, 16.7)	998	5.2 (4.9, 5.5)
	White	299,505	41,143	13.7 (13.6, 13.9)	12,499	4.2 (4.1, 4.2)
Rurality (RUCA category) §	Urban	317,938	50,975	16.0 (15.9, 16.2)	16,136	5.1 (5.0, 5.2)
	Exurban	50,949	6938	13.6 (13.3, 13.9)	2063	4.0 (3.9, 4.2)
	Small town	39,401	8061	20.5 (20.1, 20.9)	2772	7.0 (6.8, 7.3)
	Rural	65,221	12,105	18.6 (18.3, 18.9)	4093	6.3 (6.1, 6.5)
Social Vulnerability Index§	Low vulnerability	135,766	16,804	12.4 (12.2, 12.6)	4815	3.5 (3.4, 3.6)
	Mid-low vulnerability	114,199	17,286	15.1 (14.9, 15.3)	5245	4.6 (4.5, 4.7)
	Mid-high vulnerability	106,871	19,203	18.0 (17.7, 18.2)	6388	6.0 (5.8, 6.1)
	High vulnerability	116,668	24,786	21.2 (21.0, 21.5)	8616	7.4 (7.2, 7.5)

BMI = body mass index, RUCA = Rural-Urban Commuting Area.

Race and ethnicity are mutually exclusive and defined based on self- or guardian-reported electronic health record data. Hispanic/Latino patients were classified as Hispanic/Latino regardless of race; non-Hispanic/Latino patients were classified by race.

^¶Obesity defined as BMI ≥95th percentile based on the Centers for Disease Control and Prevention (CDC) Extended BMI-for-age growth chart.

^¶¶Severe obesity defined as ≥120% of the 95th percentile based on CDC Extended BMI for age growth chart.

^§Totals across rurality and social vulnerability index are lower than the overall total due to missing census tract data in about 3% of the study population.

In 2025, obesity and severe obesity prevalence increased by age group, with prevalence lowest among youth aged 2–5 years (11.2%, 95% CI: 11.0, 11.3; and 1.5%, 95% CI: 1.4, 1.5) and highest among youth aged 12–17 years (19.2%, 95% CI: 19.0, 19.4; and 7.4%, 95% CI: 7.3, 7.5). These age-based trends in obesity and severe obesity prevalence were consistent across the years 2020 through 2025 (Fig. 1). Across all age groups from 2020 to 2025, obesity and severe obesity prevalence were highest in 2021 (18.0%, 95% CI: 17.8, 18.1; and 5.8%, 95% CI: 5.7, 5.8) (Fig. 1 and Supplementary Table 3). Increases in obesity and severe obesity prevalence in 2021 were most prominent among youth aged 6–11 years (19.3%, 95% CI: 19.1, 19.5; and 6.2%, 95% CI: 6.1, 6.3) and 12–17 years (21.2%, 95% CI: 21.0, 21.4; and 8.2%, 95% CI: 8.1, 8.4).

Fig. 1.

Obesity and severe obesity prevalence trends among Minnesota youth overall and by age group, 2020–2025. (A) Overall obesity and severe obesity prevalence among youth aged 2–17 years, 2020–2025. (B) Obesity and severe obesity prevalence among youth aged 2–5 years, 2020–2025. (C) Obesity and severe obesity prevalence among youth aged 6–11 years, 2020–2025. (D) Obesity and severe obesity prevalence among youth aged 12–17 years, 2020–2025.

4. Discussion

This report demonstrates the feasibility of leveraging a statewide federated EHR data network for conducting near population-level surveillance of pediatric obesity and severe obesity. Using BMI data recorded during health care visits in Minnesota, we found in 2025 that 16.5% (95% CI: 16.4, 16.6) of youth 2–17 years with a BMI measured had obesity while 5.3% (95% CI: 5.2, 5.4) had severe obesity. Similar to other data sources, these prevalence estimates for pediatric obesity in Minnesota are lower than the national pediatric obesity prevalence (Robert Wood Johnson Foundation, 2026). Furthermore, the trends observed by age group, sex, and race/ethnicity are consistent with those described by researchers using NHANES data (Hu et al., 2025; Messiah et al., 2026; Emmerich et al., 2025; Liu et al., 2025).

The HTAC dashboard offers unique advantages over other approaches to pediatric obesity surveillance. First, the large cohorts allow for stratification of prevalence by sociodemographic characteristics, including the availability of data at the census tract level supporting linkages to social vulnerability and rurality categorization. At least one prior study has demonstrated that compared to urban children, those living in rural communities are more likely to be overweight or obese (Crouch et al., 2023). Nevertheless, the increased prevalence of obesity and severe obesity among youth residing in small towns and the lower prevalence of these conditions among youth in exurban areas has not been well described in other surveillance systems. Second, the pediatric obesity data available through HTAC are timely and granular, allowing for the calculation of single year prevalence. While other groups have noted increases in obesity prevalence during the pandemic period (Liu et al., 2025; Hu et al., 2025), the HTAC data demonstrate increases were most pronounced in 2021 in school-aged children and adolescents.

Several limitations to the HTAC data should also be noted. First, while the data is comprehensive, prevalence estimates are limited to youth with at least one health care encounter in which a BMI was measured. We are missing youth who received care at sites that do not contribute data to the MNEHRC or who did not have a care visit with BMI measured in a calendar year. Annual cohort sizes were lowest in 2020 and 2021, likely reflecting disruptions in care related to COVID-19 as well as features of the MNEHRC de-duplication process, which prioritizes more recent encounters and may exclude earlier records for individuals who transitioned between health systems. Despite these differences in cohort sizes, demographic characteristics of youth with BMI available were largely comparable throughout the study period (Supplementary Table 1). Second, we are unable to assess the characteristics or extent of records excluded due to having BMI z scores < −4 or > 4, because those exclusions were applied at participating health systems before the data were aggregated for reporting. Third, we do not provide data on extremely severe obesity (class 3, class 4 or class 5) given sparse data and concerns that the most extreme measures may reflect data errors. Finally, the data presented is cross-sectional and cannot be used for longitudinal assessments, although this is consistent with other surveillance systems.

5. Conclusions

The pediatric obesity and severe obesity prevalence data from the HTAC dashboard provide timely prevalence estimates for youth in Minnesota with a BMI recorded. These data can be used for measuring the burden of obesity across communities, evaluating the effects of population-based interventions, and informing health policy. Similar federated EHR data infrastructures could support routine pediatric obesity surveillance in other regions.

CRediT authorship contribution statement

Samuel T. Patnoe: Writing – original draft, Methodology, Formal analysis, Data curation, Conceptualization. Gabriela Vazquez-Benitez: Writing – review & editing, Supervision, Methodology, Conceptualization. Kelly R. Bergmann: Writing – review & editing. Gretchen J. Buchanan: Writing – review & editing. Michael Arenson: Writing – review & editing. Summer L. Martins: Writing – review & editing. Tyler N.A. Winkelman: Writing – review & editing, Supervision, Methodology, Conceptualization. Elyse O. Kharbanda: Writing – review & editing, Supervision, Methodology, Conceptualization.

Funding

This work was supported by a Minnesota Department of Health State Infrastructure Fund grant. The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Dr. Gabriela Vazquez-Benitez reports research support from AbbVie, unrelated to the current study. The remaining authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary material.