We take no money from the companies we review: no ads, no affiliate links, no sponsorships. Our editorial independence
Trust

Corrections

When we get something wrong, we fix it and log it here, with the date, the page, and what changed. A clear record beats a quiet edit.

Our correction policy

  • Substantive errors (wrong data, misread studies, outdated pricing presented as current) are corrected in the article and logged below.
  • The article's "Updated" date changes whenever content materially changes.
  • Typos and formatting fixes are not logged.
  • Spotted an error? Tell us. We genuinely want to know.

Correction log

DatePageWhat changed
2026-06-13 /guides/first-month-on-glp1-week-by-week/ Was: Gave Zepbound's placebo-arm side-effect rates as specific figures (about 9% and 7%) and said injection-site rotation is recommended by “the labels.”Now: Softened the placebo figures to “single-digit rates” because we could not pin the exact percentages to the prescribing information, and corrected the wording to “the manufacturer instructions.” Changed during a full accuracy audit that cross-checked every clinical figure against its primary source.
2026-06-13 /comparisons/wegovy-vs-zepbound/ Was: Cited the SURMOUNT-5 head-to-head trial loosely (“~750 participants, industry-funded”) without tying the result to the trial's primary endpoint or full citation.Now: Specified the result, tirzepatide (Zepbound) 20.2% vs. semaglutide (Wegovy) 13.7% average weight loss at 72 weeks, with the full citation (NEJM 2025, NEJMoa2416394; 751 participants), and added the SELECT cardiovascular-outcomes context for Wegovy. Fact-checked and tightened during the accuracy audit, against the primary NEJM sources.
2026-06-12 / Was: The homepage and a few other pages carried review-volume claims inherited from the design mockup: “2,000+ consumer reviews reviewed” and “thousands of verified consumer reviews.”Now: Replaced them with a figure that traces to our actual collected dataset (“900+ patient comments analyzed”) and made the homepage count dynamic. Every “n=” on the site must trace to our published research dataset; the inflated mockup numbers did not, so they were removed before launch.