Appendix Methodology | 2023 Health Of Women And Children Report

Executive Brief Introduction National HighlightsFindingsHealth Outcomes Social and Economic Factors Clinical Care Behaviors State RankingsAppendixMeasures Table – Women Measures Table – Children Data Source Descriptions MethodologyState Summaries
US Summary
2023 Health of Women and Children Report – Executive Brief 2023 Health of Women and Children Report 2023 Health of Women and Children Report – State Summaries 2023 Health of Women and Children Report – Concentrated Disadvantage County-Level Maps 2023 Health of Women and Children Report – Measures Table 2023 Health of Women and Children Report – Infographics 2023 Health of Women and Children Report – Report Data (All States)
Individual Measures
This year, 122 measures were analyzed for the America’s Health Rankings 2023 Health of Women and Children Report, using the most recent data available as of August 18, 2023. Data years varied by measure due to the variety of data sources. For some measures, multiple data years were combined to ensure reliable state-level estimates. Measure definitions, sources and data years are available in the Appendix: Measures Table. Measure changes were based on input from the Advisory Committee.
Each state was ranked according to its value for each measure, with a rank of 1 assigned to the state with the healthiest value. Ties in value were assigned equal ranks. If a state value was not available for a measure in this edition, the state value was noted as unavailable or suppressed. It is important to note that rankings are a relative measure of health. Not all changes in rank translate into actual declines or improvements in health. For additional methodology information, submit an inquiry at Americashealthrankings.org/about/page/submit-an-inquiry.
Overall, Model Category and Health Topic Summations
Summations were generated overall and by model category and health topic. They show how a state compared with other states for a model category, such as social and economic factors, or for a specific health topic, such as economic resources.
Overall state rankings were based on 83 weighted measures that met the following criteria:
Represented issues that affect population health for women and children. 
Had data available at the state level. 
Used common measurement criteria across the 50 states.
Were current and updated periodically.
Were amenable to change.
Score Calculation
The state value for each measure was normalized into a z-score, hereafter referred to as score, using the following formula:
The score indicates the number of standard deviations a state value was above or below the U.S. value. Scores were capped at +/- 2.00 to prevent an extreme score from excessively influencing the state’s overall score. If a U.S. value was not available from the original data source for a measure, the mean of all states and the District of Columbia was used. If a value was not available for a state, its value from the most recent available data year was used to generate a score.
Summation scores were calculated by adding the products of the score for each measure multiplied by that measure’s assigned model weight and association with health. Measures positively associated with population health, such as volunteerism and flu vaccination, were multiplied by 1, while measures with a negative association, such as smoking and early death, were multiplied by -1. A state that ranked No. 1 had a higher summation score (e.g., 2.00), reflecting better health than a state that ranked No. 50 with a lower summation score (e.g., -2.00). Overall state rank was calculated by ranking the overall summation score, which included all 35 measures with weights in the model (see Measures, Weights and Direction for model and measure weights).
Scores and ranks are not calculated for the District of Columbia because of its uniqueness. It is an urban population and has different governing and funding mechanisms than states. While the District of Columbia is not included in the rankings, data for the District of Columbia are available in this report and on the America’s Health Rankings website.
Data Notes
Data presented in this report were aggregated at the state level and cannot be used to make inferences at the individual level. Additionally, estimates cannot be extrapolated beyond the population upon which they were created. Values and ranks from prior years were updated on the America’s Health Rankings website to reflect known errors or updates from the reporting source.
Caution is suggested when interpreting data on certain health and behavioral measures. Many were self-reported and relied on an individual’s perception of health and behaviors. Additionally, some health outcome measures were based on respondents being told by a health care professional that they had a disease and may have excluded those who have not received a diagnosis or sought or obtained treatment.
Significance is based on non-overlapping 95% confidence intervals when comparing data over time or across demographic subpopulations.
Measure Changes
Five new measures are available in this year’s report:
High school completion: Percentage of adults ages 25 and older with at least a high school diploma or equivalent.
Injury deaths - women: Deaths due to injury per 100,000 females ages 20-44.
Injury deaths - children: Deaths due to injury per 100,000 children ages 1-19.
Housing cost burden - children: Percentage of households with one or more children younger than 18 years for which housing costs are more than 30% of household income (additional measure).
Maternity care desert: Percentage of women living in an area identified as a maternity care or low maternity care area (additional measure).
Additionally, the measure climate change policies was promoted from an additional to a ranking measure, and high school graduation was demoted from a ranking to an additional measure. Four measures were retired: violent crime, publicly-funded women’s health services, risk-screening environmental indicators score and high school graduation racial disparity.
Methodology Changes
Some data availability and comparability in this year’s report were impacted by methodological changes by reporting sources, including:
Florida data were partially available for measures from the Behavioral Risk Factor Surveillance System (BRFSS), as the state was unable to collect data to meet CDC's minimum requirements for inclusion in the 2021 public-use data set. Florida data were missing from the national values for measures using 2021 BRFSS data only. For measures using 2020 BRFSS data only, Florida data were included in the national value. For measures using 2020-2021 BRFSS data, Florida estimates were based on one year of data (2020) and one year is missing (2021). As a result, we encourage readers to use caution when interpreting Florida estimates in this report. 
The National Survey on Drug Use and Health (NSDUH) retracted its 2020 data due to methodological concerns identified after release. No new 2020-2021 NSDUH data were available this year due to this, as well as a methodology change in the 2021 survey preventing comparison to prior years. All NSDUH measures in this report had repeated data from 2018-2019 data years, which were used to calculate the state rankings. 
The National Survey of Children’s Health (NSCH) implemented a new weighting methodology on its 2022 data (i.e., enhanced). The 2022 enhanced data were not comparable to prior years and cannot be combined with 2021 data in time for this data update. All NSCH measures in this report had repeated data.  
Some 2021 BRFSS survey questions were modified to improve the quality of questions. Estimates for the measures avoided care due to cost and dedicated health care provider in this report were not comparable to prior years.
The U.S. Department of Agriculture (USDA) re-estimated its 2016-2019 WIC Eligibility and Coverage data while producing 2020 estimates. WIC coverage data in this report were not comparable to prior years.
Subpopulation Group Definitions
Subpopulation analyses were performed to illuminate disparities by gender, race/ethnicity, education, income and metropolitan status. Not all subpopulations were available for all data sources and measures. Individual estimates were suppressed if they did not meet the reliability criteria laid out by the data source or by internally established criteria. Some values had wide confidence intervals, meaning the true value may be far from the estimate listed.
Gender
This report highlights data on women and includes gender stratification as female and male for youth and children’s measures as available through public data sources — even though not all people identified with these two categories. Data did not differentiate between assigned sex at birth and current gender identity. While sex and gender influence health, the current data collection practices of many national surveys limited the ability to describe the health of transgender or gender nonbinary individuals.
Age
Age data in this report were available for measures from the Centers for Disease Control and Prevention’s (CDC) Behavioral Risk Factor Surveillance System (BRFSS) and the Maternal and Child Health Bureau’s Federally Available Data (FAD), which were sourced from the National Vital Statistics System (NVSS) and the Healthcare Cost Utilization Project (HCUP). BRFSS groupings in this report were limited to females of childbearing age and included the following self-reported age ranges: 18-24, 25-34 and 35-44. FAD groupings were based on maternal age and were grouped into five age ranges: <20; 20-24; 25-29; 30-34; and ≥35. 
Education
Education data in this report were available for measures from BRFSS and FAD data sourced from NVSS. BRFSS groupings were based on responses to the question, “What is the highest grade or year of school you completed?” A response of grades 9 through 11 was classified as less than high school. A response of grade 12 or GED was classified as high school or GED. A response of college or technical school 1 year to 3 years was classified as some post-high school. A response of college 4 years or more was classified as college graduate. FAD groupings were based on the education level that best described the highest degree or level of school completed at the time of death and were grouped into four categories: less than high school (no diploma), high school graduate or GED completed, some college (no degree) and college or technical school (associate degree or higher).
Income
Income data in this report were available for measures from BRFSS and FAD data sourced from HCUP. BRFSS groupings were based on responses to the question, “[What] is your annual household income from all sources?” Responses of less than $10,000, $10,000 to less than $15,000, $15,000 to less than $20,000 and $20,000 to less than $25,000 were summed and classified as less than $25,000. Responses of $25,000 to less than $35,000 and $35,000 to less than $50,000 were summed and classified as $25,000-$49,999. Responses of $50,000 to less than $75,000 were classified as $50,000-$74,999. Responses of $75,000 or more were classified as $75,000 or more. FAD groupings were based on quartiles (poorest to wealthiest) of current year median zip code household income obtained from Claritas, a data-driven marketing company.
Metropolitan Status
Metropolitan status data in this report were available for measures from BRFSS and FAD data sourced from HCUP. BRFSS groupings were coded based on residence geography. Identification as large central metro, large fringe metro, medium metro and small metro were classified as Metro, and identification as micropolitan and noncore were classified as Non-Metro. FAD groupings were based on the 2013 National Center for Health Statistics Urban-Rural Classification Scheme for Counties. Metro was defined as metropolitan areas with at least 1 million residents. Small to medium metro was defined as metropolitan areas of fewer than 1 million residents. Non-metro was defined as micropolitan, non-metropolitan and non-micropolitan areas.
Race and Ethnicity
Data were provided where available for the following racial and ethnic groups: American Indian/Alaska Native, Asian, Black or African American (classified in this report as Black), Hispanic or Latino (classified as Hispanic), Native Hawaiian or Other Pacific Islander (classified as Hawaiian/Pacific Islander), white, multiracial and those who identify as other race. Hispanic ethnicity includes members of all racial groups. Racial/ethnic groups were defined differently across data sources (details below). In summary, BRFSS, CDC WONDER and FAD race data were presented as non-Hispanic, while the American Community Survey data were presented as Hispanic-inclusive (except for white, which is non-Hispanic). 
Race and ethnicity categories by source:
BRFSS: American Indian/Alaskan Native (non-Hispanic); Asian (non-Hispanic); Black or African American (non-Hispanic); Hispanic, Latino/a or Spanish origin (any race); Native Hawaiian or Other Pacific Islander (non-Hispanic); white (non-Hispanic); multiracial (non-Hispanic); and other race (non-Hispanic).
CDC WONDER: American Indian or Alaska Native (non-Hispanic); Asian (non-Hispanic); Black or African American (non-Hispanic); Hispanic (any race); Native Hawaiian or Other Pacific Islander (non-Hispanic); white (non-Hispanic); and more than one race (non-Hispanic).
American Community Survey: American Indian and Alaska Native; Asian; Black or African American; Hispanic or Latino Origin (any race); Native Hawaiian or Other Pacific Islander; white (non-Hispanic); two or more races; and some other race.
FAD: American Indian/Alaska Native (non-Hispanic); Asian (non-Hispanic); Black (non-Hispanic); Hispanic (any race); Native Hawaiian/Other Pacific Islander (non-Hispanic); and white (non-Hispanic). NVSS also included multiple race (non-Hispanic) while HCUP categorized multiple race and other race as Other (Hispanic inclusive).
Methodology

Individual Measures

Overall, Model Category and Health Topic Summations

Score Calculation

Data Notes

Subpopulation Group Definitions

Reports

Explore Data

News

About Us
