Mortality tracker DCEG [code] [issues] [notebook] [gitter]

Tracking Cause of Death by State - The Center for Disease Control (CDC) uses the Socrata platform to manage public data assets, which exposes serialization (csv, json and rdf), and also query methods, as URL compositions using SODA API.

loading data from CDC ...


  • Jonas S. Almeida, Ph.D., Chief Data Scientist.
  • Amy Berrington, Ph.D., Radiation Epidemiology Branch Chief.
  • Neal Freedman, Ph.D., Senior Investigator.
  • Meredith Shiels, Ph.D., Investigator.
  • Praphulla Bhawsar, MS, Data Engineer.
  • Bhaumik Patel, MS, Software Engineer.
  • Erika Nemeth, Student intern.
  • Montserrat Garcia-Closas, MD Ph.D., Integrative Tumor Epidemiology Branch Chief.

  • Frequently Asked Questions:

    1) Why was this tool created?

    This tool was created by researchers at the National Cancer Institute Division of Cancer and Epidemiology and Genetics to monitor weekly US trends in overall and cause-specific mortality since the onset of the COVID-19 pandemic. The goal is to monitor the broader impact of COVID-19 on mortality in the USA using data visualization techniques to reveal patterns and generate potential research questions.

    2) Where do the data come from?

    The tool displays provisional vital statistics data from the National Center for Vital Statistics at the Centers for Disease Control and Prevention (CDC). These data are publicly available here: In order to provide the context of the pandemic and because the vital statistics data we present are a lagging indicator, we also display COVID-19 deaths in each state using more recent data obtained from the COVID-19 Data Repository by the Center for Systems Science and Engineering at Johns Hopkins University. ( These data on COVID-19 deaths are obtained directly from states and localities and are thus more recent than data from CDC.

    3) Are there any limitations of these data?

    Yes. The tool displays provisional data which may be revised at any time by CDC. Limitations include that it can take weeks or even months for deaths to appear within the vital statistics system. It can also take months to resolve between certain common causes of death. For this reason, the most recent weeks have substantially fewer deaths than earlier weeks. Also some states are not shown in the tracker because of substantial delays in reporting. Nevertheless, most deaths are accounted for in the vital statistics system within three weeks of occurrence. Accordingly, we have graphed the last three weeks of available data with open circles rather than continuous lines. A fuller discussion of limitations can be found here : In addition, key factors such as age, sex, and race/ethnicity are not released with the provisional cause-specific mortality data. Once available, we hope to add the ability to stratify by these demographic factors in future releases.

    4) Future plans for the tool

    Most importantly, we will incorporate vital statistics data as soon as they are released by CDC. As described above, we plan to incorporate data about age, sex, and race/ethnicity when it becomes available. We are also actively brainstorming ideas for new features and welcome submission of ideas using the GitHub.

    5) How does this tool differ from other tools available elsewhere?

    As discussed above, these data are all freely available from the CDC where they can also be viewed. The goal of this tool was to provide a quick snapshot of how overall- and cause-specific mortality in 2020 compares to past years in each state. In this way, the tool is designed to track the net impact of COVID-19 on US mortality rates. It is not designed to monitor the epidemic directly (i.e. case numbers and hospitalizations)– there are many other websites that track numbers of COVID-19 cases and deaths using more timely data obtained directly from each state.

    6) How do I share this tool?

    You are welcome to share URL for this web application, When you use the tool you'll see that your choices of cause and state are added to it. Sharing that composite URL will cause the tool to open with the corresponding parameters preselected.

    NCI/DCEG - This is a data science research tool, it should not be construed as vetted reference.