Table of Contents
Last updated: February 14, 2025. This page is updated as changes are detected in CDC data.
In the weeks since President Trump took office, the Centers for Disease Control and Prevention have removed at least 135 data sets and other files from the public data platform, Data.CDC.GOVA Stat analysis shows. The deletions seem to be part of an attempt to scrub the language that is considered reprehensible for the Trump government.
Because Trump had criticized heavily in public health during his campaign, Stat expected that there could be changes in federal health websites. While rumors started swallowing at the end of January that federal websites would be completely removed if they had not yet fully met Trump’s executive orders, we have downloaded and archived all available files from Data.CDC.Gov. We make that archives available; See the comments below. (For an archive of the web pages of the CDC, instead of data, Acasignups.net has a useful archive.)
Stat’s analysis of the data that has been removed since the inauguration shows that the word “gender” has attracted special attention. This is not particularly surprising, because many data sets of public health record data on sex and so far many data sets tend to use the terms “gender” and “sex” somewhat mixed up. Trump’s executive order on “gender ideology” Prohibits specifically all federal agencies to use the word ‘gender’ instead of ‘sex’. I have identified at least 67 items that seem to have been specifically removed because they contain the word ‘gender’, regardless of the context in which it was used.
Other items seem to have finished from other marked words and concepts. Remarkable examples are data sets with regard to the Youth risk behavior system (Yrbss) and the Behavioral risk factor surveillance system (BRFSS), who record both data on now sensitive topics such as gender identity and sexual orientation.
Strangely enough, the web pages for both data sets have been re -published on the most important CDC website after a judicial order. However, these data sets have not yet been re -published at Data.CDC.GOV.
Monitoring of the purification
Stat started regularly, real -time monitoring of the data platform of the CDC on January 31. Can be downloaded. Between then and January 31, a total of 133 items seem to have been removed while 34 new items were added. Of those 133 items, at least 72 were planned to have been removed because they were an outdated type of data visualization No longer compatible with the software That feeds the data platform of the CDC.
This means that 61 of the items that disappeared would not be deleted, and they all contain types of data that are comparable to what has been removed as part of the current purification. In addition, while it makes a backup of all CDC data files (More below) I was able to note that 51 of these items were still publicly available from January 29, which means that they must be removed between then and January 31. The other 10 are activation types that would not have been recorded in the backup of Stat. These 61 items are included in the number of status of missing and censored data.
Since our regular monitoring began, another 74 items have disappeared. This brings the total to 135 deletions.
Various data sets that have been removed were eventually uploaded with adjustments. A good example of a data set that has been re -evaluated with changes Heart disease Dictionary data among American adults (35+) per state/territory and province-2018-2020.
There is one column in this data set called StratificationCategory1of which the content had been the term gender. On February 10, when it was added to the site of the CDC 59,095 bodies of the term gender Was replaced by the term sex. Here is a shortened version of this data set, which illustrates these changes:
Another example is a data set with the name Alzheimer’s Disease and Health Aging -data. This is a data set with 284.142 rows. Originally three columns RowId” StratificationCategory2And StratificationCategoryID2 contain authorities of the term gender. When the dataset was published on February 14 on February 14, 68,838 rows had changed to replace each body of gender of sex. Another short example:
This is extremely typical of the types of changes I have seen – whole datasets that have been brought down and re -published to change one word.
We have made a back -up of data.cdc.gov
Around the same time, Stat began to follow the changes in the website and the data platform of the CDC, Stat started an attempt to download and archive all available files from Data.CDC.GOV.
Below you will find a table that you can use to download original copies from each dataset removed from the site of the CDC. Some of the 135 removed items were data visualizations or infographics linked to other data sets made with the data platform of the CDC. These are not files that can be downloaded. That is why the table below contains fewer items. However, I believe that all underlying data for those items are recorded in the backup of Stat.
Some comments about the use of this data
The Back -up files of Stat were recorded between January 29 and January 31 and they reflect data as it was published on the CDC site at that time. As noted above, some files have been uploaded to the site of the CDC. Files in the backup of Stat can be outdated compared to what is available at the CDC. Unless you have a specific concern about censorship, if the data that you want is available on Data.CDC.GOV, you should get it from there.
There are a few ways to communicate with the data platform of the CDC, mainly via its website or programmatic via an API. In the course of our monitoring there have been small discrepancies in the number of available items between the two options. This probably simply means that these systems do not run perfectly synchronously. We have decided to treat the API as the most important source of truth for the purpose of this analysis. This can lead to light discrepancies between the accounting of Stat and other archive efforts.
Roadmap
This is a work in progress and will be updated as the situation develops. I also plan to expand the functionality and usefulness of this page. A few things that I will have available soon are:
- Publish full differences of all data sets published by the CDC
- Publish the entire collection of Back -up files
- Add better search functionality to help discover files
- Add more metadata to file lists (ie descriptions, publishing and last updated times, etc.)
If you notice that something is missing, or if you have a request or need help, please contact us.
If you are interested in obtaining a complete archive of CDC data for yourself, I recommend using This collection, It was uploaded to the internet archive. This collection seems to be equal to the archive of Stat. The most handy way to download the entire collection is to use a Bittorrent client. You need around 112 GB of free space. The torrent file is located here.