Accountability · Editorial Record
The VitalScope Corrections Log
Every error we've issued since launch — original claim, corrected claim, who caught it, when it was resolved — published in real time, in plain English, with no paywall and no spin.
By the numbers
What the log actually shows.
Benchmarks verified against the Press Gazette 2023 health-publisher study. The first health publication to maintain a public corrections log of this scale.
Editor's note
Errors are inevitable. Hiding them is a choice.
Health journalism moves fast. New trials publish every week, drug labels shift, and a single misplaced decimal in a risk statistic can mislead a reader into skipping a vaccine or starting a supplement they don't need. At a publication that ships 28 to 35 articles a week, the honest answer is that we will sometimes get things wrong. What matters is what happens next.
VitalScope has published every correction in a public log since our first issue in 2017. Not a summary. Not a quarterly footnote. The full record — the original claim, the corrected claim, who flagged the error (a reader, an editor, a peer reviewer), and the timestamp of the resolution. We treat the log the way a hospital treats an incident report: as a working instrument, not a confession.
Every entry has been reviewed by at least one board-certified clinician from our 42-member Editorial Board before it appears here. Nothing is edited for comfort. If we overstated a benefit, we say so and link to the corrected version. If a citation was wrong, we link to the right one. If a reader caught it, we credit them by initials.
The pages below are searchable, filterable by topic, and downloadable as a structured feed for researchers and newsrooms. If you spot something we missed, write to the Editorial Standards Desk. We answer every report within one business day.
Recent entries
The log, as it stands this week.
-
07 Mar 2026
"Daily multivitamin cuts cardiovascular events by 12%"
Original claim A March 2026 brief summarized the COSMOS-Clinic substudy as showing a 12% reduction in major adverse cardiovascular events. Corrected claim The 12% figure refers to a secondary composite endpoint only; the primary endpoint (MACE) showed no statistically significant effect in the substudy population. Flagged by an external cardiologist on March 6; corrected version republished March 7, 06:14 ET.
Caught by: Dr. R. M., board-certified cardiologist Read corrected article → -
03 Mar 2026
GLP-1 agonist study sample size overstated
Original claim Brief cited a JAMA meta-analysis as including 18,420 participants across 14 trials. Corrected claim The pooled cohort was 14,082 participants across 9 trials; the larger figure referenced a separate unpublished registry. Caught internally during routine PubMed cross-check.
Caught by: Internal fact-check desk, 31 minutes after publication Read corrected article → -
28 Feb 2026
SSRI discontinuation risk framing
Original claim Article described SSRI withdrawal symptoms as "rare, affecting fewer than 1 in 50 patients." Corrected claim Recent systematic reviews estimate incidence between 15% and 40% depending on half-life and taper speed; we have revised language and added the cited review. Flagged by a reader and corroborated by an Editorial Board psychiatrist.
Caught by: Reader S.K., corroborated by Editorial Board Read corrected article → -
22 Feb 2026
Atrial fibrillation prevalence in adults under 40
Original claim Brief reported AF prevalence in adults under 40 at "approximately 0.5%." Corrected claim Age-standardized prevalence is closer to 0.1%–0.2% in asymptomatic populations; the higher figure cited a symptomatic cohort and was not generalizable. Updated with CDC NHANES data.
Caught by: Dr. A. P., electrophysiology fellow Read corrected article →
The pipeline
How a flagged error becomes a public entry.
The 38-minute average turnaround is not a marketing line. It is an operational pipeline with four named stages, each owned by a specific role.
-
01
Report received
A reader, an Editorial Board clinician, or our internal cross-check flags a claim. Every report is logged into the Standards Desk queue within five minutes, regardless of source.
Owner Editorial Standards Desk -
02
Source verification
The flagged claim is checked against the original peer-reviewed paper, the FDA/EMA label, and a Cochrane Review where available. Citations are opened inline on the article itself.
Owner Fact-checking team (4 full-time) -
03
Clinician re-review
A board-certified MD or PhD from the 42-member Editorial Board signs off on the corrected wording. No correction ships without a clinical reviewer attached by name and credentials.
Owner Editorial Board (rotating) -
04
Public log entry
The original claim, corrected claim, the catcher, and the resolution timestamp are published here. The corrected article is republished with an inline banner linking back to this entry.
Owner Standards Desk + Engineering
Weekly Brief
The same standard, delivered every Sunday.
612,000 readers get our Weekly Brief — physician-reviewed, citation-linked, with corrections surfaced inline when they happen. Five minutes to read. Built for the kind of person who reads a Corrections Log on purpose.