Covid-19 Analytics

4/25/2020: Added Nursing home statistics from https://data.medicare.gov/data/nursing-home-compare at a County level (# of Nursing Home residents per 100k of population, # of Reported Nursing Home Incidents per 100k of population.) Also incorporated income metrics from the American Community Survey at a county level (Median Household Income and % of households <$35k).

Release 4/20/2020: Incorporated County level death projections based upon the IMHE model, with unreported death projections allocated to County based on a mix of the # of reported confirmed cases in the last 14 days (80%) and population (20%). This allows for analysis of a normalized target, deaths per 100k of population, across various descriptive statistics like population density, temperature, activity levels, historical influenza and pneumonia death rates, age demographics, etc. For each of the descriptors included within the County level report, the grouped ranges were determined after reviewing county results broken into deciles across population density ranges. Detailed data as of 4/18/2020 can be downloaded here.

Initial Release 3/21/2020: This site will bring together data from a number of sources related to Covid-19. The hope is to provide some context and insights into the progression of the global pandemic. Data is assembled from a number of sourcesLet's start by looking at how the active/open case inventory has trended around the world. An active case is one that's confirmed and the patient has not yet recovered or died (according to work done by Johns Hopkins). The Hubei region of China was the epicenter of the outbreak and their open/unresolved case inventory peaked 24 days after logging the 500th open case. Active cases consistently decreased through day 54, where they are currently observed to be at levels similar to the beginning of the crisis. The remainder of China peaked at 18 days and is predominantly recovered as of this writing. Other countries are beginning to reveal their path and unique characteristics. The chart below is interactive, clickable and can be used to display a few metrics over time (Active/ Unresolved Cases, Confirmed Cases, Recoveries, Deaths and Death rates).

Testing update 3/22/2020: Comparing testing for Covid-19 across countries is complicated due to a lack of clear standards for tracking and reporting tests and their results. Our World in Data has done a nice job pulling together estimated tests performed to date for most International regions and the covidtrackingproject has a daily process to extract testing data for each state in the US. This data is as good as we have, but should be used with caution. A new page has been added to the project above detailing a point in time comparison of tests vs confirmed cases by region as of 3/20/2020.

update 3/23/2020: added page to display charts on log basis and a diagnostics page for major areas.

update 3/24/2020: added pages with detail on individual US States

Update 3/30/2020: Added dashboard for United States. Click on a state, or hold ctrl to select multiple states. Click upper right corner to open full screen.

4/2/2020: Added demographic data to allow for benchmark analysis. Benchmarks added:

1) Covid Deaths / Prior Year Total Deaths (1k): This metric is available at a county level and compares Covid-19 Deaths to date against Census Bureau Estimated Total Deaths in 2019. Census Bureau details were obtained from the components of population change in 2019.

2) Covid Deaths / Over65 Population (100k): Also available at a County level and expresses the Covid-19 Deaths recorded to date against the Census Bureau estimated population older than age 65. The Older than 65 population estimates were obtained from age group break-outs thru July 1, 2018. This 2018 estimate was brought forward to 2019 based upon the ratio of 2019 Overall Population Estimate / 2018 Overall Population Estimate.

3) Covid Deaths / Avg 2015-18 Influenza & Pneumonia Deaths: Reported Covid-19 Deaths to date in relation to the average number of deaths between 2015-18 due to influenza or pneumonia. Influenza and pneumonia deaths were obtained from the CDC Wonder - Underlying cause of death request system at a county level.

Census Tiger files were also consolidated to analyze the effect of population density at a county level. The section that follows is included to provide a frame of reference for these metrics using actual 2017 deaths that were attributed to Influenza and pneumonia. It is important to note that Covid-19 deaths are early in their emergence patterns, and are expected to exceed the death rates experienced with influenza as the outbreak continues to progress.

4/6/2020 (update 4/8/2020): Working to add new fields to help convey the impact of population density and temperature on Covid-19 mortality in the United States. March 2019 temperature statistics were obtained from NOAA's ftp site: ftp://ftp.ncdc.noaa.gov/pub/data/cirs/climdiv/ . There is an apparent relationship between Covid-19 mortality, population density, and average temperature. With the highest mortality occurring in high density Counties with a March 2019 average temperature in the range of 32-42F. Analysis will be updated when March 2020 Average Temperature statistics become available at a County level. An excel file with county level detail supporting the graphic below can be downloaded here. March 2020 data became available on 4/8 and has been included. A simlar relationship is persists.