CLOSE ×
Obesity Prevalence & Comorbidity Map
BASE MAP
OVERLAY
DEMOGRAPHICS
Select up to three demographic characteristics
VARIABLE TYPE
Due to inconsistencies in race/ethnicity variables among data sets and small sample sizes at the state level, PCDS results are only present at a national level to ensure the estimates are reliable.
NOTE: Race results are only available at a national level. Breast and cervical data is female only.
SOURCE: NORC analysis of the percent of cancers detected by screening (PCDS), 2017.
The interactive tool was created in JavaScript using the Leaflet library. All estimates and their variances were created using R Version 4.2.1 and survey package (4.1-1).
This interactive mapping tool provides estimates of the prevalence of obesity and select comorbidities at the state level for U.S. adults. The tool allows users to select up to three sociodemographic characteristics (i.e., age, sex, race/ethnicity) to see estimates for specific subpopulations at the state level. This interactive map allows users to compare across subpopulations within a state as well as a comparison of subpopulations across states.
All estimates in the mapping tool use the Centers for Disease Control and Prevention’s Behavioral Risk Factor Surveillance System (BRFSS) data sets from 2013 through 2021, pooled in 3-year groups (2013-15, 2016-18, and 2019-21).
To correct for any reporting bias in the Body Mass Index (BMI) measure, based on self-reporting of height and weight, NORC adjusted the distribution of BMI scores to that of the National Health and Nutrition Examination Survey (NHANES) for corresponding time periods by age, sex, and race/ethnicity using the method described in the article Projected U.S. State-Level Prevalence of Adult Obesity and Severe Obesity.1 For example, the 2013 and 2014 BRFSS data were pooled to adjust the BMI distribution to the 2013-14 NHANES distribution. The 2021 BRFSS did not have corresponding data from the NHANES and instead use adjustment factors determined by the 2017-2020 BRFSS to 2017-March 2020 NHANES adjustment.
All variables were estimated using available responses; no imputation was performed to fill in “Don’t know”, “Refused” or other missing values. Estimates were produced using the composite sample weights provided in the BRFSS data that correct for over- (or under-) sampling of specific groups as part of the sample design and reflect the probability of selection in the overlapping landline and cellphone sample frames as adjusted to correct for survey nonresponse.
The tool produces estimates for 50 states and the District of Columbia, dropping BRFSS data from Island Territories (Puerto Rico and the Virginia Islands). The 2019 BRFSS did not have data for New Jersey and the 2021 BRFSS did not have data for Florida -- as a result, the 2019-21 estimates for these states are calculated on only two years of data.
To assure the reliability of the estimates, state estimates of subpopulations were only produced if there were at least 100 observations in the 3-year pooled data for the state/subpopulation intersection. If there are fewer than 100 observations for the selected subpopulation, the map for those states will display a gray color, which means “insufficient data”. The map will also not visualize estimates for subpopulations where less than 30 states met the threshold of 100 observations. For these subpopulations, users will not be able to select the combination.
A large body of research suggests that numerous diseases, or comorbidities, are associated with obesity.2 These individual associations may also be reflected in ecological associations between aggregations of obesity and comorbidities. As such, two criteria were used in selecting which comorbidities would be included in the mapping tool: (1) is there a comparable measure of the condition collected in the BRFSS, and (2) is there at least a moderate to strong correlation between the state prevalence of obesity and the state prevalence of the comorbidity. To determine correlation, NORC calculated the correlation coefficients between state prevalence of obesity and state prevalence of the comorbidity. A correlation coefficient measures the linear correlation between two sets of data. In general, correlation coefficients of <0.3 are considered weak correlation, 0.3-0.7 is considered a moderate correlation, and > 0.7 is considered a strong correlation.3 In consultation with Novo Nordisk, NORC included comorbidities with a positive correlation of 0.5 or greater (see Table 3 below), which resulted in including nine comorbidities in the map. Some comorbidities commonly associated with obesity are not included in the map, which are discussed in the limitations section.
Data for some comorbidities were not collected in every year; pooled multi-year estimates may reflect only those years for which data were available. For instance, hypertension and dyslipidemia information were only collected in odd years; the 2019-2021 estimate is therefore based on responses in 2019 and 2021, while the 2016-2018 estimate is based solely on data from 2017.
Note on diabetesDiabetes is included in the mapping tool and includes both Type I and Type II diabetes in one measure. Based on how the data is collected in BRFSS, it is not possible to distinguish between respondents who have Type I versus Type II diabetes.
The mapping tool allows users to select age-adjusted or crude estimates (non age-adjusted). Age-adjustment of estimates was performed using Distribution #8 from the 2000 U.S. Standard Population.4 Users should select age-adjusted estimates when comparing prevalence estimates across states or between population subgroups. Using age-adjusted estimates removes age as a confounding variable, making the comparison more accurate across different populations. Users should use crude estimates when looking at estimates as a standalone statistic (such as, the percentage of adults in Ohio living with obesity), or any time a user has included an age group in the selected demographic subpopulation.
Several comorbidities that are regularly associated with obesity are not included in this mapping tool.
Pre-diabetes:Using BRFSS data, the prevalence of pre-diabetes has a negative correlation (correlation coefficient of -0.315) with prevalence of obesity. Because obesity may be associated with both transitions into pre-diabetes (from a non-diabetes condition state) and transitions out (into a diabetes condition state), the percent of the population with this pre-clinical condition state may not appear correlated with obesity despite the commonly associated relationship between obesity and poor glycemic control.
Asthma:While asthma is commonly associated with obesity, the BRFSS data showed a negative correlation (correlation coefficient of -0.233) with prevalence of obesity and is not included as a comorbidity in this tool. While obesity is commonly included as a factor that may lead to developing asthma, there are several other factors including family history, allergies, viral respiratory infections during infancy and childhood, smoking, and air pollution that also impact development of asthma. The large number of other factors may explain the negative correlation between obesity and asthma in the BRFSS data.5
Cancer (other than skin):): BRFSS data has two categories of cancer: skin cancer and all other cancers. Although prior research suggests certain cancers are associated with obesity,6 this relationship is not clear through BRFSS due to the aggregation of all other cancers in data collection. The correlation coefficient of cancer (other than skin) with prevalence of obesity in BRFSS data is 0.481.
Depression:Research identifies mental health and depression as regularly associated with higher rates of obesity.7 The relationship in the BRFSS data was relatively weak (correlation coefficient of 0.398) and is not included in this mapping tool. This relatively weak correlation may be due to self-reporting bias and underreporting of mental health conditions.8
The tables below describe each of the data sources and definitions for the variables included in the tool.
| Variable | Data Source | Definition |
|---|---|---|
| % of Population Living with Obesity – Category I & above (BMI 30+) | Behavioral Risk Factor Surveillance System (BRFSS), adjusted by National Health and Nutrition Examination Survey (NHANES) | All non-pregnant (all males and those females who did not report being pregnant at the time of the survey) respondents with BMI greater than or equal to 30. |
| % of Population Living with Obesity – Category II & above (BMI 35+) | All non-pregnant (all males and those females who did not report being pregnant at the time of the survey) respondents with BMI greater than or equal to 35. | |
| % of Population Living with Obesity – Category III (BMI 40+) | All non-pregnant (all males and those females who did not report being pregnant at the time of the survey) respondents with BMI greater than or equal to 40. |
| Variable | Data Source | Definition |
|---|---|---|
| Sex |
Prevalence of obesity by sex:
|
Respondents asked Are you male or female? Respondents required to answer to continue survey. |
| Race/Ethnicity |
Prevalence of obesity by race/ethnicity:
|
Other races includes respondents who answered: American Indian or Alaskan Native, Native Hawaiian or other Pacific Islander, Multiracial, or Other race. Respondents able to answerDon’t Know, Not sure, or Refuse |
| Urbanicity |
Prevalence of obesity by urbanicity:
|
OMB defines a Metro area at the county level based on the population of a principal city or urban cluster of at least 50,000 people and the adjacent counties that share social and economic ties via commuting patterns.9 People not in a metro area may be in either a micropolitan area (similar definitions as Metro except that the principal city/urban cluster has a population between 10,000 and 49,999) or in a county that is not part of a metropolitan or micropolitan area. |
| Education |
Prevalence of obesity by education level:
|
Respondents asked What is the highest grade or year of school completed? Respondents allowed to refuse to answer. |
| Age Group |
Prevalence of obesity by age group:
|
Respondents first asked if they are over the age of 18 to determine eligibility to continue the survey. Respondents later asked What is your age? Respondents allowed to refuse to answer second age question. |
Data source for all comorbidities characteristics is BRFSS.
| Comorbidity | BRFSS Question | Correlation Coefficient NORC calculations based on 2019-2021 BRFSS data |
|---|---|---|
| Hypertension | Have you ever been told by a doctor, nurse, or other health professional that you have high blood pressure? | 0.774 |
| Arthritis | Has a doctor, nurse, or other health professional ever told you that you had some form of arthritis, rheumatoid arthritis, gout, lupus, or fibromyalgia? | 0.680 |
| Stroke | Has a doctor, nurse, or other health professional ever told you that you had a stroke? | 0.682 |
| Cardiovascular Disease | This is an overarching measure that includes any respondent that answered “Yes” to having had a heart attack, angina/coronary heart disease, or a stroke. | 0.726 |
| Angina/Coronary Heart Disease | Has a doctor, nurse, or other health professional ever told you that you had angina or coronary heart disease? | 0.696 |
| Heart Attack | Has a doctor, nurse, or other health professional ever told you that you had a heart attack also called a myocardial infarction? | 0.683 |
| Diabetes | Has a doctor, nurse, or other health professional ever told you that you had diabetes? | 0.694 |
| High Cholesterol (Dyslipidemia) | Have you ever been told by a doctor, nurse, or other health professional that your cholesterol is high? | 0.600 |
| Kidney Disease | Not including kidney stones, bladder infection or incontinence, were you ever told you had kidney disease? | 0.529 |
1Zachary J. Ward, Sara N. Bleich, Angie L. Cradock, Jessica L. Barrett, Catherin M. Giles, Chasmin Flax, Michael W. Long, and Steven L. Gortmaker. “Projected U.S. State-Level Prevalence of Adult Obesity and Severe Obesity.” The New England Journal of Medicine, vol. 381, no. 25 (December 2019): 2440-2450. https://www.nejm.org/doi/full/10.1056/NEJMsa1909301
2Centers for Disease Control and Prevention. 2022. “Health Effects of Overweight and Obesity.” Accessed November 8, 2022.https://www.cdc.gov/healthyweight/effects/index.html
3Sarah Boslaugh and Paul Andrew Watters. Statistics in a Nutshell: A Desktop Quick Reference, (Sebastopol, CA: O'Reilly Media, 2008), Ch. 7
4Richard J. Klein and Charlotte A. Schoenborn, “Age adjustment using the 2000 projected U.S. population,” National Center for Health Statistic’s Healthy People Statistical Notes, no. 20 (January 2001): https://www.cdc.gov/nchs/data/statnt/statnt20.pdf
5American Lung Association. 2022. “Asthma Causes & Risk Factors.” Accessed November 9, 2022.https://www.lung.org/lung-health-diseases/lung-disease-lookup/asthma/learn-about-asthma/what-causes-asthma
6National Cancer Institute. 2022. “Obesity and Cancer.” Accessed November 8, 2022.https://www.cancer.gov/about-cancer/causes-prevention/risk/obesity/obesity-fact-sheet#what-is-known-about-the-relationship-between-obesity-and-cancer-
7Laura A. Pratt and Debra J. Brody. “Depression and Obesity in the U.S. Adult Household Population, 2005–2010.” NCHS Data Brief, no. 167 (October 2014), https://www.cdc.gov/nchs/data/databriefs/db167.pdf
8Melissa G. Hunt, Joseph Auriemma, and Ashara C A Cashaw. “Self-Report Bias and Underreporting of Depression on the BDI-II.” Journal of Personality Assessment, 80(1):26-30,https://www.researchgate.net/publication/10902923_Self-Report_Bias_and_Underreporting_of_Depression_on_the_BDI-II
9United States Census Bureau. “Metropolitan and Micropolitan.” 2022.https://www.census.gov/programs-surveys/metro-micro.html Accessed November 8, 2022.
Millions of children and adults in the United States are living with the chronic, complex, and treatable disease of obesity. Adults with a body mass index (BMI) of 30 or higher are considered living with obesity. The prevalence of adults living with obesity in the United States has increased over time. Pooled data from the Behavioral Risk Factor Surveillance System (BRFSS), adjusted using National Health and Nutrition Examination Survey (NHANES) found that the prevalence of U.S. adults living with obesity between 2019-2021 was 42%. This is a four-percentage point increase from 2013-2015 and means over 100 million adults in the U.S. are currently living with obesity. Notably, the prevalence of obesity varies considerably based on certain demographic and socioeconomic factors, including age, race, and college education.1
The disease of obesity has a significant economic impact on health care costs in the United States. Multiple analyses have researched the cost of obesity. The CDC reports that, in 2019, obesity-related medical care accounted for $172.74 billion in annual medical spending, and on average, adults living with obesity incur an additional $1,861 in annual medical costs compared to adults without obesity.2
Obesity is associated with other serious and chronic conditions that are among the leading causes of preventable, premature death – including cardiovascular disease, stroke, heart attack, and type 2 diabetes.3 These comorbidities are also significant drivers of the cost burden associated with obesity.
This interactive mapping tool allows users to view estimates of the prevalence of obesity and select comorbidities in U.S. adults at the state level. Users can filter by various sociodemographic characteristics to understand how characteristics including gender, age, race/ethnicity, urbanicity, and education level impact the prevalence of obesity and select comorbidities. This map serves as a tool for researchers, policymakers, and the general public, providing new insights into state-level variation in the prevalence of obesity and related diseases.
1Centers for Disease Control and Prevention. 2022. “Adult Obesity Facts.” Accessed November 1, 2022.https://www.cdc.gov/obesity/data/adult.html.
2Zachary J. Ward, Sara N. Bleich, Michael W. Long, Steven L. Gortmaker. “Association of body mass index with health care expenditures in the United States by age and sex.” PLoS ONE, 16(3) (2021): e0247307. https://doi.org/10.1371/journal.pone.0247307.
3Centers for Disease Control and Prevention. 2022. “Adult Obesity Facts.” Accessed November 1, 2022.https://www.cdc.gov/obesity/data/adult.html
This project was conducted with funding support from Novo Nordisk.
Novo Nordisk is a leading global healthcare company that's been making innovative medicines to help people with diabetes lead longer, healthier lives for more than 95 years. This heritage has given us experience and capabilities that also enable us to drive change to help people defeat other serious chronic diseases such as obesity and rare blood and endocrine disorders. We remain steadfast in our conviction that the formula for lasting success is to stay focused, think long-term and do business in a financially, socially and environmentally responsible way. With U.S. headquarters in New Jersey and production and research facilities in seven states, Novo Nordisk employs nearly 6,000 people throughout the country. For more information, visit novonordisk.us, Facebook, Instagram, and Twitter.
NORC at the University of Chicagois an objective, non-partisan research institution that delivers reliable data and rigorous analysis to guide critical programmatic, business, and policy decisions. Since 1941, NORC has conducted groundbreaking studies, created and applied innovative methods and tools, and advanced principles of scientific integrity and collaboration. Today, government, corporate, and nonprofit clients around the world partner with NORC to transform increasingly complex information into useful knowledge.
NORC’s Health Care Strategygroup combines decades of industry expertise with quantitative and qualitative evidence to provide health care leaders with a clear way forward. The Health Care Strategy group explores health care from all angles – from access to care and new payment and delivery models, to health outcomes and the socioeconomic, policy, and environmental factors that influence health – and translates complicated information into clear, usable knowledge that informs critical decision-making.
www.norc.org/healthcarestrategy
For more information please contact:
Eric Young
NORC Senior External Affairs Manager
young-eric@norc.org
(301) 634-9536
This interactive mapping tool allows users to view estimates of the prevalence of obesity and select comorbidities in U.S. adults at the state level, using the CDC’s Behavioral Risk Factor Surveillance System (BRFSS) data. Users can filter by various sociodemographic characteristics to understand how characteristics including gender, age, race/ethnicity, urbanicity, and education level impact the prevalence of obesity and select comorbidities. This map serves as a tool for researchers, policymakers, and the general public, providing new insights into state-level variation in the prevalence of obesity and related diseases.
The base layer of the map visualizes the percent of a state’s adult population living with obesity. The user has the option to select between three options:
Use the “Obesity Measure” dropdown to select which category of obesity is visualized on the map. For the obesity measure, the darker blue means higher prevalence of obesity while lighter blue means lower prevalence. A legend in the top right indicates the percent ranges for each color.
This mapping tool contains data from 2013 – 2021, pooled into three-year groups: 2013-2015, 2016-2018, and 2019-2021. Users can select which time period is visualized on the map by selecting from the dropdown “Time Period”.
Users also have the option to select between age-adjusted estimates and crude estimates, by selecting from the dropdown labeled “Estimate Type”. Users should select age-adjusted estimates when comparing prevalence estimates across states or between population subgroups. Using age-adjusted estimates removes age as a confounding variable, making the comparison more accurate across different populations. Users should use crude estimates when looking at estimates as a standalone statistic (such as, the percentage of adults in Ohio living with obesity), or any time a user has included an age group in the selected demographic subpopulation.
The user can select from a list of eight comorbidities to overlay additional data on top of the obesity measure. Use the “Comorbidity” dropdown to select which comorbidity is visualized on the map. The comorbidities overlay is displayed as a red circle on top of each state. The size of the red circle demonstrates the prevalence of the selected comorbidity in the adult population of a state. The larger the circle, the higher the prevalence of the selected comorbidity in the state. A legend above the map indicates the percent ranges for each size circle. Users can only select one comorbidity at a time to visualize on the map.
This mapping tool allows users to select demographics to obtain prevalence estimates by subpopulations. As a user selects demographic characteristics, the base layer (obesity measure) and the overlay (comorbidity, if selected) will update on the map to visualize the prevalence for the selected subpopulation.
There are five demographic categories: Sex, Race/Ethnicity, Urbanicity, Education, and Age Group. Users can select up to three demographic characteristics to identify a subpopulation of interest. The tool will only allow users to select up to three demographic characteristics:
| 3 demographic characteristics | Available? | 4 demographic characteristics | Available? |
|---|---|---|---|
| Female, In a metro area, 30-39 years | YES | Female, In a metro, 30-39 years, Bachelor’s degree or higher | NO |
Due to small sample sizes, some subpopulations may not be available in certain states, or not available at all. If there is insufficient data for a state, the map will show a gray color. If there is insufficient data in more than 20 states for a specific subpopulation, the tool will not allow a user to select that combination of demographics.
Reminder:any time a user has included an age group as one of the selected demographics, the user should select “Crude Estimate” from the “Estimate Type” drop down. Age group demographics will only display when “Crude Estimate” is selected.
Users can view details for a specific state by clicking on a state and clicking “View Details”. A data table will open containing the prevalence estimates for the subpopulation, time period, and estimate type the user selected in the map. This data table includes prevalence estimates for the three obesity measures and all eight comorbidities. For example, if a user has selected “Males, In a Metro Area, and Less than High School” and selects Texas, the data table will show the prevalence of obesity and each comorbidity in Texas males living in a metro area with an education level of less than high school. The data table will also show a national comparison for the same subpopulation.
Within the data table, users can add one additional subpopulation for comparison. Click the “Add Comparison Demographic”. A small window will popup where users can select the new subpopulation to add to the data table. In this example, a user may want to understand effects of education, so they could select “Male, In a Metro Area, Bachelor’s degree or higher” to add to the data table. The prevalence estimates for the comparison demographic will be added to the data table, along with another national estimate for the comparison demographic. Users can print these data tables.