About
About SuicideData
SuicideData gathers statistics and research from reliable sources and presents them in plain language, following safe-messaging guidance — to improve understanding, reduce stigma, and prevent suicide.
What's here
Statistics
US suicide rates by state, sex, age, race and method, and rates across countries. Every figure carries its source, its year, and what it cannot tell you.
Research library
Peer-reviewed work on prevention, treatment, assessment, postvention and theory, each entry summarized in plain language and carrying its full citation.
News
Current coverage, screened against responsible-reporting guidance. Stories about individual deaths are kept off the list and reachable only if you ask for them.
How suicide is written about here
- No method detail. Population-level figures on means appear because they are the evidence for means restriction; nothing about how any individual died appears anywhere.
- Rates rather than raw counts where a rate is the honest comparison, and the limits of a figure alongside the figure rather than in a footnote.
- Coverage of individual deaths is not published in the news feed. It is held back, and a reader who needs it has to open it deliberately.
- Crisis lines on every page, at the top, whether or not the page is about crisis.
How this was built
This is an AI project. Language models do the reading: one pipeline screens the literature and writes the plain-language summaries, another screens the news against the reporting rules above. The rules come from published guidance, everything rejected is logged so the screen can be audited, and a person reads the output. That last part is most of the work. The screen has quietly binned real prevention coverage, treated the phrasing journalists are asked to use as a warning sign, and placed a Massachusetts radio station in California. Each was caught by reading what it produced.
Where the figures come from
Mortality statistics come from CDC and NCHS, international rates from the WHO Global Health Observatory, and the figures on thoughts, plans and attempts from the National Survey on Drug Use and Health via NIMH. Research is drawn from PubMed, OpenAlex, Europe PMC and Crossref. News is surfaced from aggregators and screened before anything is published.
What the figures cannot tell you
Official counts are believed to understate suicide mortality by roughly 10–30%, and the undercount is worse for some populations — misclassification on death certificates has been estimated at around 33% for American Indian and Alaska Native people. Some figures are also fresher than others: state rates run to 2024, while the splits by sex, age and race stop in 2018 because that is where the published table stops. Each is dated where it appears.
Not medical advice
This is an educational resource. Nothing here is medical advice, a diagnosis, or treatment, and nothing here replaces telling someone. If you need to talk to a person now, 988 is free and answers by call or text, any hour.