Figures we published wrong
6 statistics that were live on this site and turned out to be wrong, misattributed or missing a caveat that changed what they meant. Each one, where it appeared, and what the page says now.
All were found by our own checks rather than reported to us. The verification log is append-only, so a correction is added to the record rather than replacing what was there — this list can only grow. Errors caught in drafts before they were published are not listed here; these were all live. If you find another, tell us and it will appear here.
What we said
Health anxiety affects 5.7% lifetime, 3.4% experiencing it 'in any given 12-month period' (Sunderland et al., 2013)
On /research/health-anxiety-statistics/
What it says now
Page now distinguishes the three figures the study actually reports: 5.7% lifetime, 4.2% in the past 12 months, and 3.4% current at the time of interview. The 3.4% was previously mislabelled as the 12-month figure.
src/pages/research/health-anxiety-statistics.astro
What we said
Health anxiety carries a 4x suicide risk that is the single largest driver of excess mortality
On /research/health-anxiety-statistics/
What it says now
Live page now states HR 4.14 with the caveat that it did not survive adjustment for comorbid depressive and anxiety disorders. The 'single largest driver' superlative is removed.
src/pages/research/health-anxiety-statistics.astro
What we said
NICE CG113 supports a 60% figure for additional psychiatric diagnoses in anxiety disorders
On /research/health-anxiety-statistics/
What it says now
Live page now states 57% citing Brown et al. 2001, labelled as a specialist-clinic sample. CG113 removed from the references.
src/pages/research/health-anxiety-statistics.astro
What we said
NICE CG113 supports a 90% OCD comorbidity figure
On /research/health-anxiety-statistics/
What it says now
Live page now cites Ruscio et al. 2010 (NCS-R) with the paper's own breakdown. The 90% figure itself was correct.
src/pages/research/health-anxiety-statistics.astro
What we said
The $256bn annual US healthcare cost figure is driven by health anxiety
On /research/health-anxiety-statistics/
What it says now
Live page now states plainly that the figure is for somatization broadly, and that the two overlap but are not the same.
src/pages/research/health-anxiety-statistics.astro
What we said
Women are 1.7 times more likely to develop health anxiety than men (McLean et al., 2011)
On /research/health-anxiety-statistics/
What it says now
Page now states the 1.7 ratio is for anxiety disorders as a group and notes the study reports no separate figure for health anxiety. It previously attributed the ratio to health anxiety specifically.
src/pages/research/health-anxiety-statistics.astro
More from the research library
Treatments compared
Six approaches ranked by evidence
Questions answered
14 answers, each with its evidence
Research database
All 10,657 papers and the query behind them
Research hub
Start here
Statistics
Prevalence, mortality, cost
SHAI-14
The measure and what scores mean
Terminology
Health anxiety vs hypochondria