FAQ

The City Health Dashboard

Here you will find a list of questions and answers that may come up as you explore the City Health Dashboard. If your question is not listed here, please email us at [email protected].

Why was the Dashboard created?

Communities may find it difficult to access rigorous data at the census tract or city level. The City Health Dashboard simplifies and streamlines this process.  The Department of Population Health at NYU Langone Health and the Robert F. Wagner School of Public Service at NYU created the Dashboard as a health improvement planning resource for U.S. cities. The Dashboard continues to grow, and is a one-stop resource for viewing and comparing data from multiple sources on health and the factors that shape health to guide local solutions.

What is health?

Health is more than health care. A wide range of factors influence how long and how well we live from education and income to the quality of our housing and the safety of our neighborhoods. Similar to the City Health Dashboard, the County Health Rankings & Roadmaps, a partnership between the Robert Wood Johnson Foundation and the University of Wisconsin’s Population Health Institute, helps communities understand what influences how healthy residents are and how long they will live.  The Rankings are based on a model of population health that emphasizes the many factors that, if improved, can help make communities healthier places to live, learn, work and play. The Rankings measure two sets of data for counties:

• Health outcomes- the county’s current health status, including physical and mental health and premature mortality.

• Health factors- the elements that drive a county’s health outcomes, including the physical environment (such as air quality and the availability of parks and grocery stores); social and economic factors (such as education, income, and employment); health behaviors (such as tobacco use, diet, and exercise); and access to and quality of clinical care.

The City Health Dashboard uses a similar model to categorize over 40 measures of health and drivers of health in cities to help community leaders and residents pinpoint and take action on gaps in health and opportunity.

Who should use the Dashboard?

The Dashboard is geared toward policymakers, city government staff, community organizations and advocates, research institutions, and media. We welcome all users! How should I cite City Health Dashboard?

City Health Dashboard should be cited when the data or graphics are used, including in published presentations, articles, research, blogs, policy documents, and other print or digital media.

We encourage use of Dashboard data and visualizations, and suggest the following citation: Department of Population Health, NYU Langone Health. City Health Dashboard. https://www.cityhealthdashboard.com/. Accessed [INSERT DATE OF ACCESS].

Can I download the Dashboard data to analyze on my own?

Yes, our data are available for download as a .CSV here. Note that the .CSV files have the most recent year of data for all metrics and all geographies on the website. If you need help, please email us at [email protected]. Our data is not currently available to access through API at this time. Due to website infrastructure updates that allow for more efficient data releases and a faster user experience, the API is being redeveloped and is unavailable starting on July 26, 2023. The Dashboard team and our website development partners, Forum One, will be working thoughtfully to resume access to the API soon. If you need help attaining data, please email us at [email protected]. Who are the Dashboard’s sponsors?

The Department of Population Health at NYU Langone Health and the Robert F. Wagner School of Public Service at NYU created the Dashboard in partnership with the National Resource Network. The project launched in 2017 and has received generous ongoing support from the Robert Wood Johnson Foundation.

How is the Dashboard different from other data tools, like the County Health Rankings or the PLACES Project (formerly 500 Cities Project) data? The City Health Dashboard is one of several tools supported by the Robert Wood Johnson Foundation to leverage data to increase awareness and drive action to improve the nation's health. Our focus is city and neighborhood level data, and making it easily accessible across health factors and drivers. Click here to learn more about the various data projects and to explore how they are similar to and different from the Dashboard.

The Dashboard Process

How were the Dashboard metrics and data sources chosen?

The following metric inclusion criteria were used to compile accurate, consistent, and comparable data across the 5 domains:

• Rigorous and equitable methods underlying the original data collection

• Feasible data acquisition by the NYU Grossman School of Medicine analytic team

• Evidence of importance and validity in academic literature

• Metrics that are amenable to city-level intervention

• Time lag between the Dashboard release and data collection ≤ 5 years (and preferably less)

• Updated regularly, preferably at least every 2 years

• Balanced across the 5 domains (clinical care, health behaviors, health outcomes, physical environment, and social and economic factors)

When possible:

• Aligned with other population health reporting frameworks (e.g., County Health Rankings & Roadmaps, Vital Signs, Culture of Health)

• Disaggregated by census tracts or demographics

• Available for all Dashboard cities

• Aligned with city preferences based on input from the Dashboard pilot cities and City Advisory Board

There are many metrics that would be valuable to include on the Dashboard. However, often for logistical reasons around the underlying data sources, we are unable to include a proposed metric. We welcome suggestions and ideas.

How were the Dashboard cities selected?

The Dashboard launched in 2018 with data for the 497 most populous cities in the nation and the largest cities in Vermont (Burlington), West Virginia (Charleston), and Wyoming (Cheyenne). At least one city from every state was represented in this initial set of cities, mirroring the Centers for Disease Control and Prevention’s 500 Cities Project. In April 2020, the Dashboard added 256 small and midsize U.S. cities with populations above 50,000 (based on 2017 census estimates). These cities were selected from an analysis completed by the City Health Dashboard team, in partnership with the Robert Wood Johnson Foundation.  In 2022 and 2023, we added 207 additional cities to the Dashboard through a few different avenues: population growth from the 2020 Census, a Dashboard-led challenge to add smaller cities, and the expansion of our definition of cities to include a designation that captures towns, townships, villages, and boroughs. The Dashboard team also occasionally works with partners to add new, smaller cities for specific projects. To learn more about partnering with the Dashboard, email us. Why is the population listed under the city header different than I expected?

The population number listed under the city header is based off of ACS 2021 5-year estimate data (updated from 2020 data in the 2023 summer release). This value is updated every year after ACS data are released annually, though there is usually some lag. Other websites or sources may use data from the 2020 census to determine population values, which may differ slightly from the ACS estimate. This may be due to several reasons, including that the ACS estimate is a 5-year estimate pooling data from multiple years, while the census estimate is from a single year, ACS provides a survey-based estimate while the census provides a direct count of all people in a city, and ACS is updated each year while the census is updated every 10-years. We chose to present ACS 5-year estimate values for city populations because these values are updated annually and are available for even very small cities, allowing us to provide a more up-to-date estimate for population for all cities on the Dashboard.

How were the metrics calculated?

More information on metric calculations is available in the City Health Dashboard Technical Document. How are the city and census tract boundaries determined?

The city and census tracts boundaries are created from the Census block-level using US Census TIGER shapefiles. Census tract boundaries represent the portion of the tract within the city, not the entire census tract. Depending on the underlying data source, the city and tract boundaries are based on 2010 or 2020 decennial Census. Please refer to the City Health Dashboard Technical Document for more detail.

How did the Dashboard validate the data?

The Dashboard implements a multi-step data validation process to ensure the accuracy of metric value calculation and data uploaded to the website display. If you’d like to learn more, please email us. All analyses conducted by the City Health Dashboard analysis team were first conducted by a primary analyst (some data are posted as received from the underlying data source - see below). The analysis was then replicated independently by a secondary analyst within the group. Results were compared; any discrepancies were investigated and internally discussed until the two analyses generated identical values. Additionally, analyses of data presented online were performed to confirm that the calculated values matched the values presented on the website. More information on this process and on which metric values were posted as received from the data source is available in the City Health Dashboard Technical Document.

Does the Dashboard rank cities?

The Dashboard does not provide a weighted composite score to rank cities. Instead, it offers a comparison function that allows users to evaluate a city’s performance on certain metrics against self-selected peer cities, thus providing a more detailed perspective of a city’s health performance.

Does the Dashboard provide multi-year data?

Yes, the Dashboard provides multi-year data for over 30 of our metrics. More information on how to use and interpret our multi-year data is available here.

What is the Dashboard’s update schedule?

The Dashboard strives to refresh the data in as near real time as possible. However, the update schedule depends on the update schedules for the underlying data sources. All the data sources are updated regularly, generally at least every 2 years.

The Dashboard Definitions

What is a census tract?

Census tracts are subdivisions of a county or an equivalent entity with a population of 1200 to 8000 people. Census tract boundaries facilitate statistical comparison from census to census; however, census tract boundaries may change between decennial censuses. A census tract roughly corresponds to a neighborhood, but it does not always align with neighborhood boundaries. Learn more about census tracts here. How does the Dashboard define race/ethnicity?

The Dashboard categorizes race/ethnicity into 5 groups:

• Asian*

• Black/African American

• Hispanic/Latino

• Other**

• White

*In general, “Asian” represents a combination of Asian and Native Hawaiian/Other Pacific Islander. The exact definition of “Native Hawaiian/Other Pacific Islander” varies by data source. See City Health Dashboard Technical Document for more detail by data source.

**In general, “Other” represents a combination of American Indian and Alaska Native, “other race”, and “two or more races”. Because some data sources do not release data on “other race” and/or “two or more races”, the definition of “Other” varies by data source and associated metrics. See City Health Dashboard Technical Document for more detail by data source.

More information on the Dashboard’s racial/ethnic categorization is available in the City Health Dashboard Technical Document.

Why are American Indian/Alaska Native and Hawaiian/Other Pacific Islander presented as part of Other racial groups?

Small population counts for American Indian/Alaska Native and Native Hawaiian/Other Pacific Islander prevent the City Health Dashboard from breaking out measures on these demographic subgroups. American Indian/Alaska Native populations are combined with "Two or More Races" and "Some Other Race" to ensure adequate population size. For the same reason, NHOPI estimates are combined with "Asian.”

The Dashboard reports the populations of AIAN, NHOPI, and Two or More Races population groups separately on the Demographics Overview and Demographics by Census Tracts tabs on each city's City overview page.

The Dashboard reports metrics as rates, percentages, or indices. What do these values mean and why does the Dashboard provide them?

The Dashboard provides these values to be consistent with standard scientific data reporting practices and to maintain a transparent methodology.

• A percent is a proportion expressed per 100 parts. A proportion is the frequency with which a given event occurs in a given population.

• A rate is the frequency with which a given event occurs in a given population within a given period of time.

• An index is a statistic that summarizes and ranks multiple indicators to produce a single composite value.

The type of measure is determined by the data analyzed to derive each estimate. More information on these values is available in the City Health Dashboard Technical Document.

For select metrics, the Dashboard provides error margins. What do these values mean and why does the Dashboard provide them?

The Dashboard provides these values to be consistent with standard scientific data reporting practices and to maintain a transparent methodology.

• An error margin provides a measure of the variation around a given value. With a few exceptions, the Dashboard reports 90% CIs for its metrics. These values can be found in downloadable data and in the tabular view of the city data.

More information on the calculation of error margins for specific metrics is available in the City Health Dashboard Technical Document.

Data Usage for the Dashboard

What are appropriate uses of the Dashboard data?

The data may be used to:

• Identify health issues in a city or neighborhood(s)

• Identify emerging health problems

• Prioritize health issues

• Establish key health objectives

• Explore multi-year data for select metrics

• Develop and implement targeted prevention-focused programs and policy changes

• Compare health status and drivers of health across neighborhoods and in similar cities

• Target issues for cross-sector or cross-jurisdiction (city, county, state) partnership and action

What are inappropriate uses of the Dashboard data?

There are global limitations to the Dashboard data, as well as limitations unique to certain data sources. More information on the nuances of specific data source limitations is available in the City Health Dashboard Technical Document. Generally, the data should not be used to:

• Rank the overall health of a city in comparison to other cities. The Dashboard does not provide a weighted composite score across metrics for cities. Therefore, the data should not be used to rank the overall health of a city in comparison to one or more other cities. 

• Evaluate the impact of local public health programs or policies. Local intervention effects will not be reliably detected in the next year’s refreshed data because of how the underlying data sources collect and, in some cases, the use of modeled data for specific metrics.    

• Copy data from website for statistical testing. Users should download the underlying data from the Data Downloads page for projects requiring statistical tests.

• Users should review our multi-year data best practices, including the limitations of individual data sources. More information is available here, in the City Health Dashboard Technical Document, and under Tips and Cautions.

How can I use the Compare Cities feature?

With the Dashboard’s Compare Cities feature, you can compare a city’s performance on specific metric with that of other cities. You may select comparator cities using criteria such as: similar population size, region, state, adjacent state, income level, race/ethnicity distribution, and similar performance levels on other Dashboard metrics. In addition, it is important to understand the nuances of your city comparison. Please use caution when interpreting these results, paying attention to changes in years of data and underlying geography. Refer to the Tips and Cautions for more details.

How can I use the Compare Metrics feature?

With the Dashboard’s Compare Metrics feature, you can select and compare two metrics. Use informative visuals, such as metric comparison maps, scatterplots, and bar charts, to better understand the relationship between almost any two metrics.

In addition, it is important to understand the nuances of your metric comparison. Please use caution when interpreting these results, paying attention to changes in years of data and underlying geography. Caution is important when examining metric comparison maps and scatterplots: just because two metrics are lower or higher in the same neighborhood, or are correlated does not mean that one causes the other. Learn more about correlation and causation here. Refer to the Tips and Cautions for more details.

Why are some metrics stratified by demographics?

Not all metrics have demographic data, depending on the demographic groups reported by underlying data sources.

Why are some metrics only reported at the city level?

All metrics are reported at the city level; however, census tract-level data availability depends on the geographic granularity provided by underlying data.

Why are data missing from my city’s page?

Unfortunately, data are not available for every city and census tract. For some metrics, data that are missing for one year might be are available for another year. Data may be missing for a variety of reasons, including:

1. Sparse data. Smaller populations, especially those located in smaller cities, do not produce as much data as large populations. This may make estimates unreliable, preventing their release.

2. Data not released. Data sources may not release values to protect individual privacy or because of quality concerns.

3. Internal censorship policies. Per the terms of a data source’s use agreement the Dashboard may not release values that do not meet certain criteria. More information on data censorship criteria is available in the City Health Dashboard Technical Document.

We might add the data you are looking for in the future. Sign up here to be notified when new data are released.

Why don’t I see error margins for city values?

Error margins are not displayed on the Dashboard. However, they are available in the downloadable data files and in the tabular view of the city data. Why might the Dashboard’s estimates vary from estimates of similar metrics on other sites?

Estimates of similar metrics may differ between the Dashboard and other resources due to different data sources and analytic techniques. When comparing the Dashboard’s estimates to other similar metrics on similar sites, please use caution and refer to the City Health Dashboard Technical Document to understand all potential sources of difference.

Why do population breakdowns on the Demographic Detail page vary from metric to metric?

On the Demographic Detail page, the Dashboard provides additional information about a city’s demographic breakdown to put the metric values into context. For example, for the race/ethnicity breakdowns for the metric High School Completion, the Dashboard also provides the percentage of the city’s adult population that is each race/ethnicity (see example below). Population breakdowns are different across metrics because the numbers correspond to the group of people relevant to the metric, not necessarily the city's residents overall. For example, the population breakdown for High School Completion describes the racial/ethnic breakdown of adults in a city, whereas the population breakdown for children in poverty describes the racial/ethnic breakdown of children in a city.

Note that sex-specific estimates do not have population breakdowns because these numbers consistently hover around 50% per group across cities.

Why do some metrics have 500 Cities Project as the source in some years and PLACES Project as the source in other years? What is the difference between these data sources?

The PLACES Project is an extension of the 500 Cities Project, a collaboration of the Centers for Disease Control and Prevention (CDC), the Robert Wood Johnson Foundation (RWJF) and the CDC Foundation. The original 500 Cities Project provided data on health measures for 500 of the largest cities across the United States. PLACES Project provides the same data for more geographies, including all U.S. counties, cities, census tracts, and zip code tabulated areas (ZCTAs). For these health measures in years 2013 – 2017, the Dashboard presents 500 Cities Project data for our original 500 cities. For 2018 and forward (2017, for high blood pressure), the Dashboard presents PLACES data for all cities. The Dashboard cautions against using 500 Cities and PLACES census tract or city level data to track changes over time. More information on how to use and interpret our multi-year data is available here.

What does ‘estimate should be interpreted with caution’ mean on the Demographics Overview tab?

The data displayed on the Demographics Overview tab reflect estimates provided by the US Census American Community Survey (ACS). ACS produces these estimates by interviewing some of the people who live in a given city or neighborhood. ACS cannot interview everyone in each place, so the numbers they provide are estimates, or well-informed guesses. For some racial/ethnic subgroups, these estimates are based on responses from only a few people that identify as members of that racial/ethnic group. When this happens, the final estimate is based on information from fewer people and is less reliable than an estimate based on information from more people. The Dashboard adds an indicator to ACS estimates that reach specific uncertainty thresholds in order to inform users that these estimates may be imprecise or that their precision cannot be determined. You can learn more about how this is calculated in the City Health Dashboard Technical Document.

Are the national comparison estimates for the whole U.S.?

No, the national comparison estimates for each metric are calculated using values only from the Dashboard’s cities. These estimates are calculated after censoring criteria are applied.

Where can I find more technical information?

Please refer to the City Health Dashboard Technical Document. Taking Action

How can I use the Take Action section?

The Take Action section provides 5 categories of useful tools created by organizations and websites we trust to help you drive change in your community:

1. Find Policies and Programs includes evidence-based policies and programs and any promising practices to support cities in implementing new ideas

2. Find Partners identifies websites and organizations that can foster collaboration to help you achieve your goals

3. Explore Strategies provides best practices for each stage of a project, curated by audience type

4. Find Funding identifies the diverse funding streams available to users and provides tips on how to obtain this financial support

5. Measure Impact provides tools to help users leverage data to support their work and evaluate programs

Please email us if you have any tools or documents that have been useful in your work that we can include as resources on the Take Action page.

These new sections are meant to provide practical, logistical support around implementation in a way that we currently do not offer on the site.

How can I use the Stories and News section?

The Stories and News section showcases the Dashboard’s blog, any Dashboard-related news, and stories of successful communities integrating health into decision- making. This section may be used to generate ideas on how to improve health outcomes in a community. Please email us if your city is interested in being featured on the site.

How do I submit my city’s story to the Dashboard?

We love hearing how cities are using the Dashboard. Please contact us directly with your city’s story at [email protected].

Contacting the Dashboard

Who should I contact if I have questions, suggestions, or see an error on the site?

We are eager to hear from you. Please contact us directly at [email protected].

How do I stay connected with the Dashboard?

Please subscribe to our mailing list and follow us on social media.

• Subscribe to our mailing list

• Follow us on Twitter at @CityHealthData

• Follow us on LinkedIn

Can I submit a data request?

We plan to offer additional data services for cities and organizations that want to examine their data more deeply, including adding new cities, custom metrics, disaggregations to new geographies, as well as expert assistance for using Dashboard data in your city. Interested groups can engage our data analysts for a number of services. If you would like to learn more or have an idea for a project, please email us for details.

Is the Dashboard available in other languages?

We hope to be able to release a Spanish version of the Dashboard soon.

Metric Specifics

Why is the life expectancy metric no longer available for my city? Why is my city’s life expectancy estimate different than before?

Life expectancy city-level estimates are calculated by aggregating the tract-level data publicly available from USALEEP. The Dashboard recently updated its aggregation method to better account for underlying population distributions, as well as censor city level estimates that were missing significant contributing tract estimates. This resulted in a change to metric estimates and data availability on the website. Please refer to the City Health Dashboard Technical Document for more information.

Why do estimates of third-grade reading scores and absenteeism appear different from estimates I have seen from other sources?

The Dashboard’s estimates for third-grade reading scores and absenteeism may differ from those from other sources because: 1) the Dashboard presents city-level estimates, not school district-level estimates; and 2) the Dashboard includes all public schools (regular, vocational, alternative, and special education) in its calculation. Some other estimates include only traditional schools. More information is available in the City Health Dashboard Technical Document. I notice there are two unemployment metrics. What’s the difference? The Dashboard has two unemployment metrics. Both are important for improving health, well-being and equity in U.S. cities, but they have different data sources, which have important methodological differences that can contribute to differences in estimates The table below provides details.

I noticed that COVID local risk index metric has been removed. What happened?

The national Public Health Emergency for COVID-19 expired on May 11, 2023, as per the U.S. Department of Health and Human Services (HHS). Though the Dashboard acknowledges the current and lasting impact that COVID-19 has on communities across the country, the team doesn’t believe this risk index to still be as relevant. The data is still available for research purposes, please just contact us and we’ll send you what you need.

I noticed that Limited Supermarket Proximity metric has been removed. What happened?

The Dashboard evaluates new metrics and those already on the site continually, with high standards for statistical rigor and external validity. When the team discovers questions about methodology or user-relevance, we remove the metric from the site and continue research and development for potential replacements. In this case, the Dashboard team questioned whether the ½ mile distance to a supermarket was an appropriate cut-off for all congressional districts, urban and rural. We also discovered methodology questions around missing values, so the metric has been removed, and we aim to have a replacement measure of food access/insecurity in the future.

I noticed that the Air Pollution – Particulate Matter metric has been changed from a yearly metric to a monthly metric. What happened?

The Dashboard continues to evaluate the measures and associated data sources on our websites.  We are particularly interested in finding and adding timely and granular data sources when available. The Dashboard opted to present 1 month average estimates starting in 2022 to provide more granular data to assist users interested in considering seasonal or environmental conditions. Please contact our team for further information.

I noticed that age breakdowns for the Uninsured metric are unavailable for 2013 and 2014. What happened?

The Dashboard thoughtfully validates all data before transferring it to the website. This validation includes reviewing new data within the context of older data already on the website. During this process, the Dashboard may find new methodological questions or concerns. This process has occurred with the Uninsured measure when methodological differences were found for the age disaggregation for years 2013 and 2014, therefore the decision was made to remove these breakdowns for these years. The other demographic breakdowns for 2013 and 2014, as well as the age breakdowns for the Uninsured measure for years after 2014 were methodologically uniform and will be maintainedon the website. Please contact our team for further information.

I noticed that there is an unexpected difference in the Preventive Services metrics for 2020 my city. What happened?

Please describe the methodological changes for Preventive Services between 2019 and 2020.

The Dashboard uses Preventive Services data from a data source called PLACES, which, in turn, relies on BRFSS as its data provider. In 2020, BRFSS introduced supplementary questions concerning preventive care in their survey. These additional questions provided individuals with more opportunities to be considered up to date with their core preventative services, resulting in a noticeable increase in adherence across the board between 2019 and 2020.

Last updated: July 26, 2023