We publish our audit record because accuracy claims should be checkable. Every entry on this site runs through an automated fact-audit pipeline (accuracy audit → source-grounded repair → visual QA); this page is generated from that pipeline's own report, not written by hand.
Last full accuracy audit: 2026-09-02 · 10 entries checked · 5 flagged · 7 issues confirmed · 5 corrected · 2 still open
Coronary artery disease — The claim that air pollution is responsible for roughly 25% of deaths from CAD is an overstatement; the attributable fraction for CAD specifically is lower, around 15-20%.
Crohn's disease — The claim that tobacco smokers are three times as likely to develop Crohn's disease as non-smokers is an oversimplification; the established relative risk is approximately 1.5 to 2 times, not three times.
Crohn's disease — The claim that half will require surgery within ten years is outdated; modern data suggests lower rates (e.g., 20-30% within 10 years) and the entry does not note the decline.
Dermatitis — The claim 'Contact dermatitis is twice as common in females as in males' contradicts the entry's own STATS section stating 'no universally established ratio', and is not a universally established fact.
Dwarfism — The incidence of achondroplasia is typically cited as 1 in 15,000 to 1 in 40,000 live births (approximately 2.5 to 6.7 per 100,000), not 4 to 15 per 100,000.
Dyscalculia — The claim 'comorbidity_with_ADHD=11% of children with dyscalculia also have ADHD' is inaccurate; research typically finds comorbidity rates between 20% and 60%.
Dyscalculia — The claim 'In a study, only 65% of children diagnosed with dyscalculia over four years were diagnosed for at least two years' is vague and unsourced, with no widely-known canonical study supporting this specific statistic as a general fact about dyscalculia.
All content is fetchable as JSON at /data/content/ · llms.txt · sitemap
Spotted an error? Suggest an edit — reader reports go straight into the audit queue.