How Trial Friend writes, sources, and fact-checks every word
Trial Friend serves rare disease patients and caregivers researching life-changing decisions. Our accuracy bar is therefore zero room for error. This is how we get there, and how to reach us if we miss.
Sources we trust
Every factual claim on Trial Friend traces back to one of: peer-reviewed medical literature (PubMed, NEJM, JAMA, Lancet, specialty journals), FDA documents (drug labels, approval announcements, advisory committee briefs), ClinicalTrials.gov, NIH and CDC publications, the Genetic Testing Registry, manufacturer Investor Relations press releases for pipeline updates, SEC filings (10-K, 10-Q, 8-K) for company-level facts, and disease-specific patient advocacy organizations recognized by NORD or the EveryLife Foundation.
Sources we do not treat as primary: Wikipedia, drug industry blogs, social media, and AI-generated summaries from other systems. These are useful starting points for finding primary sources but are never cited as the source of a claim.
Inline attribution, not bare numbers
When a Trial Friend post or page cites a statistic, dollar amount, study finding, or named claim, the source is woven into the sentence. The reader should know where a number came from without scrolling to the bottom of the post. The sources list at the end of every blog post is for verification, not as a substitute for inline attribution.
The wall between business and editorial
Trial Friend may display clearly labeled advertisements, sponsored content, or branded patient-education tools from pharmaceutical companies, patient advocacy foundations, research organizations, and other healthcare partners. When this happens, sponsored content is marked as such and is visually distinct from editorial content. Sponsorship is permitted on places like a company's own page, a drug's own page, or other clearly-defined branded spots.
What sponsorship cannot do, ever: influence what we say in editorial content. No advertiser, sponsor, or partner has input into our articles, drug decoder pages, treatment timelines, comparison charts, side effect data, clinical trial coverage, or any other editorial content. Drug coverage in our editorial pages stays analytical regardless of whether the manufacturer has a separate sponsorship relationship with Trial Friend. The presence of an advertisement is not an endorsement. See our Terms of Use for the full advertising policy.
What we do not do
- We do not auto-publish AI-generated content. Every post is reviewed for accuracy, tone, and patient-friendliness before going live.
- We do not let sponsors influence editorial content. Drug coverage, disease pages, comparison charts, and articles are editorially independent regardless of any business relationship with the manufacturer.
- We do not provide medical advice. Information is informational and educational. Treatment decisions belong with you and your physician.
- We do not sell, share, or monetize personal information. Advertisers and sponsors receive aggregate, anonymized statistics only. See our Privacy Policy and Do Not Sell page for specifics.
- We do not refer patients to specific clinics, doctors, or treatment programs in exchange for compensation. Trial listings come directly from ClinicalTrials.gov.
How we handle uncertainty
Medicine is full of areas where the evidence is mixed, the guidelines disagree, or the patient community and mainstream literature don't fully align. When we cover those topics (the IDSA / ILADS divide on Lyme disease, MTHFR clinical actionability, off-label uses of approved drugs, contested diagnostic criteria for emerging syndromes), we name the disagreement rather than pretending it doesn't exist. We tell you what's solidly evidenced, what's debated, and where the literature stands. We don't pick sides for you.
How we use AI
Trial Friend is written and edited by Jason Hunter, and the thinking behind the site is his: what to cover, what patients need to know first, which sources count, and how each page and tool should work. AI tools help stitch that work together and validate it: finding and reading primary sources, drafting and editing text to his direction, checking facts, numbers and links against those sources, and monitoring ClinicalTrials.gov, FDA notices and company filings for changes every day.
AI does not decide what is published. Every article is checked against the sources it cites and approved by Jason before it goes live, and nothing is published automatically. AI output is never treated as a source: every fact on the site has to trace back to one of the primary documents listed above.
Freshness and updates
Drug pipelines move fast. FDA decisions land. Trials read out, get put on hold, or get terminated. Trial Friend runs an automated daily monitor that scans authoritative sources (ClinicalTrials.gov, FDA, SEC 8-K filings) for material changes affecting tracked conditions and drugs. When a change matters, we update the affected pages and (for major events) publish a focused blog post explaining what happened and what it means for patients. Every page has a last-reviewed indicator near the bottom.
Corrections policy
We will get things wrong. When we do, the correction process is fast and transparent. If you spot a factual error, an outdated claim, a missing citation, or a misrepresentation, please email jason@trialfriend.com with the page URL and the specific issue. We aim to acknowledge corrections within 24 hours and publish updates within 72 hours of verification. Material corrections (changed numbers, retracted claims) will be flagged inline on the page.
Contact
For corrections, source questions, partnership inquiries, or feedback on a specific page: jason@trialfriend.com.
For privacy questions: see our Privacy Policy. For general background on Trial Friend: about.