Changes in version 1.0                         

  - Data refresh
  - Fixes for fetching
  - Apple mobility data is now frozen (Apple stopped releasing mobility
    data on April 14th, 2022)
  - Breaking change: The apple_mobility data object's layout has changed
    and it is now more compact. See docs for new variable names.

                        Changes in version 0.9.2                        

  - Data refresh

                         Changes in version 0.8                         

  - Data refresh

                        Changes in version 0.7.8                        

  - Breaking change: nchs_wss, the table of provisional death counts in
    the U.S. across a variety of units of observation, has had its race
    and ethnicity categories updated by the NCHS. The group column has
    been renamed race_ethnicity. A new column obs_unit indicates whether
    the results in a given row are tabulated by Total, by Year, or by
    Month.

                         Changes in version 0.7                         

  - Breaking change: covnat, the ECDC daily dataset, has been renamed to
    covnat_daily. Daily data reporting by the ECDC was discontinued on
    December 14th 2020 and so this table will no longer be updated.
  - A new dataset, covnat_weekly has been added. This is the ECDC's
    replacement dataset. It gives weekly national data from the
    beginning of the pandemic to the present.

                         Changes in version 0.6                         

Breaking change: Removed individual-level NCHS data due to its large
size. The plan is to provide more functions for the user to grab this
and other data directly.

                        Changes in version 0.5.2                        

Breaking change: The google_mobility dataset is getting pretty big,
causing longer load times and also installation failures in some
settings, e.g. on RStudio Cloud instances. I have relocated it from this
package to the new covmobility package. covmobility also contains the
apple_mobility dataset. For now, apple_mobility is still in covdata as
before, though this may change.

                         Changes in version 0.5                         

NCHS/CDC weekly mortality counts for 2014-2020 (useful for assessing
excess deaths) are now included as nchs_wdc.

                         Changes in version 0.4                         

Breaking change: The index variable in the apple_mobility dataset is now
named score. Under the hood, various tidyverse or tidyverse-adjacent
tools such as tsibble and fable use the concept of an index variable in
many functions. Often it is named as index internally. Having a column
in the data actually named index can lead to strange and confusing
errors.

                         Changes in version 0.3                         

Added the CDC's patient-level public-use dataset, as nchs_pud

                        Changes in version 0.2.9                        

Added two tables adapted from the COVID tracking project:

  - covus_race provides total reported cases and deaths by state and
    race
  - covus_ethnicity provides total reported cases and deaths by state
    and Hispanic status

See the COVID Tracking Project Page for more details on the source,
scope, and limits of these counts.

                        Changes in version 0.2.8                        

Added two tables from the National Center for Health Statistics, one on
death counts by sex, age and state; the other on state-level estimates
of deaths by racial and ethnic group.

                        Changes in version 0.2.7                        

The pop_2018 column in covnat has been updated to 2019 population data
and renamed pop. This is a breaking change for any code that used
pop_2018 to calculate, e.g., per capita rates.