Executive BriefIntroductionDesignNational FindingsKey FindingsSocial and Economic FactorsPhysical EnvironmentClinical CareBehaviorsHealth OutcomesState SummariesAlabamaAlaskaArizonaArkansasCaliforniaColoradoConnecticutDelawareDistrict of ColumbiaFloridaGeorgiaHawaiiIdahoIllinoisIndianaIowaKansasKentuckyLouisianaMaineMarylandMassachusettsMichiganMinnesotaMississippiMissouriMontanaNebraskaNevadaNew HampshireNew JerseyNew MexicoNew YorkNorth CarolinaNorth DakotaOhioOklahomaOregonPennsylvaniaRhode IslandSouth CarolinaSouth DakotaTennesseeTexasUtahVermontVirginiaWashingtonWest VirginiaWisconsinWyomingU.S. SummaryAppendixMeasuresData SourcesMethodologyNational Advisory CommitteeThe Team
Depending on the source, three to five years of data were pooled across three time periods between 2003-2019 to produce reliable estimates. For all indicators, state and nationally-representative population estimates were produced, along with their confidence intervals using methods appropriate for each data source.
Estimates were created for the nation, 50 states and District of Columbia as a whole and for subpopulation groups. Subpopulation analyses were conducted by race and ethnicity, gender, and education, and if available by metropolitan status.
Estimates were compared within subpopulation groups and over time to ascertain whether differences were statistically significant at the 95% confidence interval. Disparity gaps within states were also calculated by identifying the groups with the smallest and largest estimates within each subpopulation, and calculating absolute differences between those estimates. This allowed for the identification of states with the highest and lowest disparities by subpopulation group. Finally, ratios were calculated to compare groups and further highlight the magnitude of disparities.
Individual estimates were suppressed if they did not meet the reliability criteria laid out by the data source or, failing specification by the data source, established internally for doing so.
Years of data were combined for some data sources and measures to increase sample size. Because data availability varied by source, the years of data combined for each of the three periods varied by source and measure.
Subpopulation Group Definitions
This report highlights disparities by four key subpopulation groups as summarized below. Not all subpopulations are available for all data sources and measures. In addition, where they are reported, the definitions may vary.
Race and Ethnicity. Data are provided for seven racial and ethnic groups:
- American Indian or Alaska Native
- Asian or Pacific Islander
- Black or African American
- Hispanic or Latino
Racial and ethnic groups are mutually exclusive. Specifically, “Hispanics” includes members of all races, while all racial groups include only non-Hispanic members.
Gender. This report includes data for a binary measure of gender: females and males.
Educational Attainment. Data by educational attainment are categorized as:
- Less than a high school education
- High school graduate
- Some college
- College graduate
Geography. Where available, differences were also assessed by geography in terms of metropolitan and non-metropolitan regions. The definition of “metropolitan” varied by data source. For indicators from the American Community Survey, the designation is based on the Public Use Microdata Area (PUMA) of residence. PUMAs that are majority metropolitan area are defined as metropolitan; all others as non-metropolitan. For data from the National Vital Statistics System, metropolitan status was defined according to the 2013 National Center for Health Statistics Urban-Rural Scheme for Counties. Metropolitan included counties coded as large central, large fringe, medium and small metropolitan areas. Non-metropolitan counties included those coded as micropolitan and non-core.
Given the methods used to ensure adequate sample size by pooling multiple years of data where necessary, estimates produced are reliable. Further, the sampling methods used by the surveys ensure estimates are representative at the state and national levels. With that said, there are several limitations that users should be aware of. First, subpopulations are defined differently, in some cases, across data sources. For one, the “other” race category used by the source of the Severe Housing Problems data includes both respondents who report as being of some other race and those who report being multiracial. In addition, the metropolitan and non-metropolitan categories are defined differently for variables created using national vital statistics data (infant mortality, maternal mortality, low birthweight and premature mortality) and those created using data from the American Community Survey, as described previously.
Second, despite combining three to five years of data, minority groups that make up less than 5 percent of a state’s population in smaller states frequently have suppressed estimates for some variables. Where estimates are not suppressed, confidence intervals can be wide, meaning that the true rate may be far from the point estimate listed.
Third, estimates cannot be extrapolated beyond the population upon which they were created. Users should refer to the variable definitions to confirm the population used. This is particularly true for the Able-Bodied indicator which includes a broader age range than the indicator used in other America’s Health Rankings reports.
Finally, caution should also be taken when interpreting data on specific health measures. Of note, many health outcome measures indicate whether a respondent has been told by a health care professional that they have a disease, excluding those who may not have received a diagnosis or not have sought or obtained treatment.