Click here to download a Microsoft Excel file containing the data used in the Geography of Social Isolation in U.S. Older Adults tool.

The interactive tool was created in JavaScript using the Leaflet library. Data were processed using R and converted from shapefile to GeoJSON using rgdal library. Multiple administrative data sources were accessed in the development of the tool, including the Census Bureau American Community Survey, National Center for Health Statistics, and the HRSA Area Health Resource file. A complete list of each variable and its data source can be found in the table below.

Several measures of social isolation were estimated using a small area estimation methodology, combining direct survey estimates from the National Social Life, Health, and Aging Project Survey (NSHAP) with model based predictions incorporating variables related to social isolation. The small area estimates were generated by creating a linear statistical model relating the NSHAP direct survey estimates in large counties with multiple NSHAP respondents to various explanatory variables.

The tables below present the data sources and definitions for the variables included in the tool. 

Counties are categorized as urban or rural based on the USDA Economic Research Services 2013 Urban Influence Codes (UIC), with UIC codes 1-2 corresponding to urban counties, and UIC codes 3-12 corresponding to rural counties. County UIC codes are based on the size of the largest city or town and proximity to metro and micropolitan areas as defined by the census.

This tool allows users to create county-level maps illustrating relationships between indices of social isolation (base map) with demographics and health outcomes (overlay).

Here are the step-by-step directions to explore various features of the interactive mapping tool:

To select a base map, choose a measure of social isolation from the pull-down menu. For example, you might select the measure of loneliness scores among the population of those 50 and older. The map will then depict this base map data in the form of varying shades of blue (by quantiles) at the county-level. Buttons labeled 50+ (or 65+) indicate that the percentages mapped include only adults 50 (or 65) years and older.

Once you have selected a base map layer, you may choose to select an overlay. To select an overlay, choose a measure from a category of your choice by using its respective pull-down menu. Categories include individual measures or components of social isolation; COVID-19 variables, demographics, health and health behaviors, health care utilization, and health system capacity. For instance, you might wish to view the overlap between loneliness scores and a demographic measure, such as rates of poverty among adults aged 65 and older. The map will then depict this overlay data in the form of pink spots of varying sizes that correspond to the respective percent of the population living in poverty (by quantiles).

When a second layer of data is added to the map, an option will appear on the left hand sidebar to add a scatterplot to the graph. Hovering over the dots in the scatterplot will highlight the locations of the counties they correspond to on the map.

There are two ways of navigating to view data for a particular county.

1. The first is to click on the county directly on the map and select “View Details” from the popup.
2. The second is to click on the list of county data tables and search for a county by state using the appropriate pull-down menu. Then, county-level data for various categories of measures may be selected using the “select data table” pull-down menu.

To view state-level data, click the "state/county" drop down in the upper-right section of the screen and select "State".

Two filters for the map are available:

1. A filter by data value – This will filter out all counties except those in the highest or lowest quantile of the base map variable, overlay variable, or both. Using the data value filter allows the user to find the upper 25% of values (“high basemap only”) or lower 25% of values (“low basemap only”) among states in the country. Additionally, selecting "No statistical outliers" will filter out all counties with extreme base map values.
2. A filter by state - When viewing the county-level map, filter by state removes county shading from all states except the state selected.

When two variables are displayed on the map, a button reading “Open Graph” will appear at the top of the sidebar. Clicking this button will open a scatterplot showing the two mapped variables and a regression line that shows the relationship between the two variables. The map filters will affect which counties are displayed in the scatterplot, and hovering over a dot on the scatterplot will highlight its corresponding county on the map