Methodology for The Big Cities Health Inventory Platform 2.0
About the Data
As of October 2016, the Big Cities Health Inventory (BCHI) data platform has more than 50 indicators that look at health status, death rates, and other socio-economic and demographic factors that affect the health of a community.
The indicators encompass nine broad categories of public health importance: Behavioral Health and Substance Abuse; Cancer; Chronic Disease; Environmental Health; Food Safety; HIV/AIDs; Infectious Disease; Injury and Violence; and Maternal and Child Health. Two additional categories include demographics and life expectancy/overall death rate. These indicators were chosen based on their relationship to the leading causes of morbidity and mortality in the United States and their role in creating healthier, safer communities. The latter was largely gauged by whether an indicator was identified as a priority in the U.S. Department of Health and Human Service’s Healthy People 2020 goals, CDC’s Winnable Battles, or due to interest among Coalition members.
Much of these data come directly from cities, while some were also secured from the U.S. Census or other similar publicly available data set for which city data were available. For the most part, jurisdictions reported their three most recent years of data, which were often 2012, 2013, and 2014. Data prior to 2010 were not included, even if it meant a jurisdiction only had two years of data available. The platform will continue to be updated with more recent data as it becomes available from cities or other such sources.
The nature of the data on the platform varies considerably. When data were not provided or available, the appropriate cell was left blank. Not all health departments were able to provide data for all indicators and, in cases where dominators were too small, certain rates for subpopulations are not displayed. This is notated in two ways – either nothing is displayed or a “S” is displayed in the cell in the place of a number. As the platform is updated, the latter option will be the default, rather than leaving a cell blank. Whenever possible, “unusual” numbers were verified, either by checking with the original sources or through other means. Most data were reviewed by individual cities as well.
Some Coalition health departments were only able to provide county level data due to availability, and/or because the city accounts for most of the county population. This is footnoted where applicable.
RACE AND ETHNICITY
Where sample sizes allow, indicators are broken down into subpopulations for race and ethnicity categories. For most jurisdictions, the default options were White (Non-Hispanic), Black (Non-Hispanic), Hispanic, Asian/Pacific Islander, American Indian/Alaska Native, and Other. In areas where certain populations were too small, the various subpopulations were included in the “other” category with any additional racial/ethnic minorities. In many of the California cities, as well as Seattle, reported numbers only represent Asians; Pacific Islanders are not included. Some jurisdictions also report mixed-race numbers, and where they do, those numbers are reported as “Multi racial.”
As is customary, communicable disease indicators are reported using crude rates. Mortality rates are age-adjusted to compare relative mortality risks among cities, different demographic groups, and over time. In most cases, the 2000 standard population age was used. All mortality rates are presented per 100,000 people. Jurisdictions that used more recent population counts than the year 2010 are noted. Emergency department data rates are calculated per 10,000 people and are also age adjusted. Indicator names and/or footnotes reflect differences regarding age-adjustment and/or crude rates.
EXPLANATION OF INDICATORS
Each of the indicators and associated methodology are further explained below.
BEHAVIORAL HEALTH AND SUBSTANCE ABUSE
The percentage of adults who binge drank is based (in most cases) on the Behavioral Risk Factor Surveillance System
(BRFSS) question about how many drinks a person had on one occasion in the past 30 days. Women who answered “four” and men who answered “five” are considered binge drinkers.
Similarly, percent of high school students who binge drank is based (in most cases) on the Youth Risk Behavior Surveillance System (YRBS/YRBSS) question regarding five or more drinks within a couple hours on one or more occasions in the past 30 days. As with other indicators, if BRFSS or YRBS/YRBSS data were not available, a comparable survey was used, or the data were left out if not completely comparable.
The Opioid-Related Unintentional Overdose Mortality Rate are per 100,000 people, using 2010 U.S. Census figures, age-adjusted to the year 2000 standard population (except where noted). ICD-10 codes include: X40-X44; AND one or more of the following in any multiple cause of death field: T40.2, T40.3, T40.4. NOTE: This indicator is not exclusive of other drugs that may be included in multiple cause of death fields, such as heroin (T40.1), cocaine (T40.5), benzodiazepines (T42.4), psychostimulants with abuse potential (T43.6), other and unspecified narcotics (T40.6), or drugs not elsewhere classified (T50.9). Morphine and heroin are metabolized similarly. This may result in the over-reporting of drug poisonings associated with the effects of opioid analgesics.
The drug-abuse related hospitalization rate is based in part on how individual jurisdictions compute these numbers, as a consensus definition could not be reached in the time period necessary. For this reason, this data should not be compared across jurisdictions. While most cities used a combination of ICD codes and/or hospital databases, specific sources and methodologies are noted in the online data platform. Special attention should be paid when using these numbers.
Mortality rates for lung and female breast cancers, as well as all types of cancer (combined) are reported per 100,000 people, using 2010 U.S. Census figures, age-adjusted to the year 2000 standard population (except where noted). ICD-10 codes for all cancer include: C00-C07; for lung cancer: C33-C34; and for female breast cancer: C50.
Asthma emergency department visit rates are reported per 10,000 people, using 2010 U.S. Census figures, age-adjusted to the year 2000 standard population (except where noted). ICD-9-CM Codes include: 493.0 – Extrinsic asthma; 493.1 – Intrinsic asthma; 493.2 – Chronic obstructive asthma; 493.8 – Other forms of asthma; 493.9 – Asthma unspecified.
Heart disease and diabetes mortality rates are per 100,000 people, using 2010 Census figures, and are age-adjusted to the year 2000 standard population. ICD-10 codes: I00-I09, I11, I13, and I20-I51 for heart disease, and E10-E14 for diabetes. Comparable obesity and physical activity data are difficult to find at the city level.
Where possible, the adult obesity figure in this report is the percentage of the population 18 years or over that is considered obese, generally with a body mass index (BMI) of 30 or above, and in most cases is taken from BRFSS. Similarly, obesity rates for children are difficult to collect, though many jurisdictions know the percent of high school students that are obese, particularly in large urban school districts. In children, the definition is a BMI at or above the 95th percentile of children of the same age or sex. Physical activity data were taken from BRFSS or YRBS/YRBSS based on CDC-recommended activity levels. For adults: at least 2 hours, 30 minutes of moderate-intensity aerobic activity every week for good health; 1 hour, 15 minutes of vigorous-intensity aerobic activity; or an equivalent mix of moderate and vigorous. For high school students: physically active for a total of at least 60 minutes per day. Where possible, this publication relied on BRFSS or YRBS/YRBSS so that data were comparable. In most cases, if data were not comparable, they were excluded.
Data on cigarette use among both adults (over 18 years of age) and youth (for the most part, high school students) are also hard to obtain at the city level. The most frequently used sources of data are the BRFSS, the Youth Risk Behavioral Survey (YRBS), or the Youth Risk Behavioral Surveillance System (YRBSS). Many cities/counties oversample to have accurate data for the jurisdictions, but some do not. Sample sizes vary, as do years of data available. Youth tobacco numbers were included only if they were secured via YRBS or YRBSS or a comparable survey, both in terms of population (high school students) and question text. Readers should take note of both source and year of data availability when using the tobacco- related data in this publication.
Data for children living in poverty, median household income, foreign-born residents, high school graduates, those living below 200% of the poverty line, and the unemployment rate were calculated using the U.S. Census Bureau’s American Community Survey (ACS) 1-year estimate tables with a few exceptions. A research assistant accessed the data files and created estimates based on a micropolitan code of statistical areas defined by the Office of Management and Budget February 2013. Delineation Files may be slightly different from jurisdictional lines used for other data in this publication. County data for Fort Worth (Tarrant County), Las Vegas (Clark County), and San Diego County were secured via the American Community Survey online data tables.
Data for the following demographic indicators were also calculated using the U.S. Census Bureau’s American Community Survey (ACS) 1-year estimate tables. FIPS codes, which were confirmed by the cities, were used to isolate the data for the jurisdictions. Margins of error that were included in ACS 1-year estimates were also collected, where available.
Percent of 3 and 4 year olds currently enrolled in preschool included children in both public and private schools. The numerator was 3 and 4 year olds enrolled in school, the denominator was the population of the 3 and 4 year old age group.
Percent of households whose housing costs exceed 35% of income was calculated to capture the excessive housing cost burden for each city using selected housing characteristics from the ACS 1-year estimates to calculate the following: [SMOCAPI(35.0% or more, with mortgage) + SMOCAPI(35.0% or more, without mortgage) + GRAPI(35.0% or more)]/ Occupied Housing Units. This value was then reported as a percentage. SMOCAPI = Selected monthly owner costs as a percentage of household income, GRAPI = Gross rent as a percentage of household income
Percent of population 65 and over was an age category provided by the ACS estimates and represents those 65 years and older, divided by the total population estimate. Percent of Population under 18 was calculated by subtracting the percentage of the population 18 years and over from 100 to capture the population under the age of 18.
Percent who only speak English at home were those who reported speaking only English at home. Percent who speak Spanish at home were those who reported speaking a Spanish or Spanish Creole language other than English at home. The denominator for both values was the population five years and older.
Race/ethnicity included the percent of the total population that reported being of the following race/ethnicity categories: White (Non-Hispanic), Black (Non-Hispanic), Hispanic, Asian/Pacific Islander, American Indian/Alaska Native, Multiracial (Two or more races) and Other. Sex was reported as the percentage of the total population that was male and female. Total population was the ACS 1-year estimate of the number of individuals in the jurisdiction.
Percent of children tested under age 6 with elevated blood lead levels was calculated by cities. “Elevated” was defined by the CDC standard, children who had blood lead levels of 5 ug/dl and over. The numerator is the number of children with elevated blood levels and the Denominator is the number of children under 6 screened.
Laboratory-confirmed infections caused by salmonella or Shiga Toxin-Producing E. coli are crude rates per 100,000 people, using 2010 Census figures.
HIV/AIDS data include HIV Diagnoses Rate and AIDS Diagnoses Rate in a given year, as well as a Persons Living with HIV/AIDS Rate (which includes those who have both HIV and AIDS). Each of these indicators report the crude rate per 100,000 population, using 2010 U.S. Census figure (except where noted). The HIV-Related Mortality Rate is also reported per 100,000 people, using 2010 U.S. Census figures, age-adjusted to the year 2000 standard population. In most cases, these data are identified using ICD-10 codes B20-B24. See glossary of terms for additional definitional information.
Infectious disease indicators include flu vaccination for children and adults, pneumonia vaccination for those age 65 and older, pneumonia and influenza mortality rates, and incidence (cases diagnosed in a given year) of tuberculosis (TB). Percent vaccinated for adults over age 18 means one dose of annual flu vaccination in a given year. Similarly, percent vaccinated for children under 18 years of age means a child received at least one dose in a calendar year. Note that these data were difficult to obtain and sources vary. Where possible, Behavioral Risk Factor Surveillance System (BRFSS) data are reported for adults. Sources for childhood immunization data vary. Data on the percent over age 65 vaccinated for pneumonia were secured wherever possible, and data that were generally not comparable were left out. Pneumonia and Influenza Mortality Rate is reported per 100,000 people, using 2010 U.S. Census figures, age adjusted to the year 2000 standard population (except where noted). TB Incidence is reported as a crude rate per 100,000 people, using 2010 U.S. Census figures (except where noted).
INJURY AND VIOLENCE
Homicide, suicide, and firearm-related mortality rates are per 100,000 people, using 2010 Census figures, age-adjusted to the year 2000 standard population. ICD-10 codes are: X85-Y09, Y87.1 for homicide; X60-X84 and Y87 for suicide; and W32-W34, X72-X74, X93-X95, Y22-Y24, and Y35 for firearm-related mortality (except where noted). Motor vehicle mortality rate is per 100,000 people using 2010 Census figures, age-adjusted to the year 2000 standard population. ICD-10 Codes: V02-V04, V09.0-V09.2, V12-V14, V19.0-V19.2, V19.4-V19.6, V20-V79, V80.3-V80.5, V81.0-V81.1, V82.0-V82.1, V83-V86, V87.0-V87.8, V88.0-V88.8, V89, and V89.2 (except where noted).
Firearm Related Emergency Department Visit Rate are per 100,000 people, using 2010 Census figures, age-adjusted to the year 2000 standard population. ICD-9-CM Codes included: E922.0-E922.3, E922.8, E922.9 – Accident caused by firearm missile; E955.0-E955.4 – Suicide and self-inflicted injury by firearms; E965.0-E965.4 – Assault by firearms; E985.0-E985.4 – Injury by firearms, undetermined whether accidentally, or purposefully inflicted; E970 – Injury due to legal intervention by firearms; E979.4 – Terrorism involving firearms.
LIFE EXPECTANCY AND ALL-CAUSE MORTALITY RATES
All-cause mortality (overall death) rates and life expectancy were also requested from jurisdictions. Where data were not available, the space was left blank. A few life expectancy numbers were provided by Virginia Commonwealth University’s Center on Society and Health. Life expectancy estimates for cities often have a range of years due to the need to aggregate several years of data. For display purposes, the last year of that range is categorized in the platform. Notes on each page show the full range of data years.
MATERNAL AND CHILD HEALTH
Maternal and child health indicators include infant mortality rate, low birthweight, and percent of mothers under age 20. Infant mortality rate is the mortality rate per 1,000 live births. Percent low birthweight is defined as the percentage of babies born under 2,500 grams. Percentage of mothers under age 20 reflects the birth mothers’ age.
GLOSSARY OF TERMS
The following terms are defined by the CDC’s Principles of Epidemiology in Public Health Practice, Third Edition: An Introduction to Applied Epidemiology and Biostatistics and the United States Cancer Statistics.
Rate – A rate is a measure of the frequency with which an event occurs in a defined population over a specified period of time.
Crude rate – The total number of cases of a particular disease or condition over the total population size for a given period of time. As the crude rate is influenced by the underlying age distribution of the state’s population, cities will often report an age-adjusted rate.
Age-adjusted rate – Uses a standard population, generally the 2000 U.S. standard population, that is based on that year’s population age groups. Using direct standardization, these populations by age group serve as weights for calculating the age-adjusted rate. This ensures differences in rates are not due to different age distributions of the populations.
Stages of HIV Infection are classified according to the same standards as those laid out in the CDC’s Morbidity and Mortality Weekly Report (MMWR) Revised Surveillance Case Definition for HIV infection. HIV cases are confirmed by laboratory or clinical evidence, the definition for which is described in the MMWR previously mentioned. Once the case is confirmed, the staging is based on age-specific CD4+ T-lymphocyte count (or CD4+ T-lymphocyte percentage of total lymphocytes). To be considered an AIDS case (Stage 3), the CD4+ T-lymphocyte count is less than 200 cells/microliter when the age on date of the CD4+ T-lymphocyte test is greater than or equal to 6 years.