Q. I’m in the middle of working with a client on a white paper that has citations to articles found on government agency websites (e.g., Centers for Disease Control and Prevention, National Institutes of Health, Office of Disease Prevention). The writing is completed, we’re in the layout/production stage, and while checking links and confirming URLs in the endnotes, we’re finding that articles and pages that had been referenced in our endnotes have now been removed from the government websites in accordance with the administration’s recent orders. How do we reference reports and articles that are significant but have been removed?
A. Whenever you find a dead link (or a link that works but points to a different version of the cited content), you have a few options, which are the same regardless of why the link no longer works as intended:
- If the cited page can’t be found anywhere, and the author didn’t save a copy, then you can either (a) add “(page no longer available)” or similar wording after the URL (as shown at CMOS 14.104), or (b) ask the author to revise the text and citation to fix the problem (perhaps by referring to and citing a different document). Option a should be used only as a last resort.
- If the author did save a copy, you can add that fact to the notice suggested for option a above: “(page no longer available; copy of original in author’s possession).” Option 1b may still be preferable unless the cited document is crucial to the author’s paper.
- If you can track down an archived version of the page—for example, at the Internet Archive’s Wayback Machine—you can cite that (after making sure that any text that relies on that page, including any direct quotations, remains accurate).
Option 3 is usually best, assuming the content has been archived in a way that allows others to consult it. For example, consider the following URL, which as of March 4, 2025, results in a “Page Not Found” notice:
https://www.cdc.gov/healthcommunication/Health_Equity.html
If you paste that URL into the Wayback Machine, you’ll find dozens of archived versions (or snapshots) dating back to August 25, 2021. Here’s how we’d cite one of those versions, following the recommendations at CMOS 14.104 (and using the “last reviewed” date reported at the bottom of that page in lieu of a publication date; see CMOS 13.16):
1. “Health Equity Guiding Principles for Inclusive Communication,” Gateway to Health Communication, CDC, last reviewed August 11, 2023, archived July 24, 2024, at https://web.archive.org/web/20240724170713/https://www.cdc.gov/healthcommunication/Health_Equity.html.
If you can’t find the page at the Wayback Machine or anywhere else, you’re back to options 1 and 2 at the beginning of this answer.
Links break for many reasons, but there are some basic precautions that authors can take to prevent the scenario described in the question above from happening in the first place:
- First, always save a copy of any web page (including any PDFs) that you consult as you do your research (e.g., as a screenshot or as an HTML or PDF file). Zotero and other citation managers can help with this task. See also CMOS 13.13 and 13.17.
- Second, don’t assume that a page will have been archived by someone else (as was the case for the CDC.gov URL in the example above). Instead, create your own archived version if you can—for example, by pasting the URL into the Save Page Now feature at the Wayback Machine. This won’t always be an option (some content will be blocked from being archived), but at least you’ll have your own personal copy to point to (see previous bullet) if anyone challenges your research.
Copyeditors can help by alerting authors to any page that may need to be archived or saved in one of the ways described above—and by editing citations accordingly. That should keep everyone, including readers, on the same page.