How I work, where I am biased, and how to correct me.

My bias, stated plainly

• GreenMedInfo is selective by design. It is a curated index of peer-reviewed studies on foods, herbs, nutrients and other natural interventions, plus research on unintended harms of conventional treatments. Editors choose the studies. It is not a systematic review, and it does not claim to represent all of PubMed. A study that is not in the index may exist, and may point the other way.
• I am an advocate. I write from conviction about informed consent, health freedom and free speech. I am the founder of Restore the First and was a named plaintiff in Finn v. Global Engagement Center. I am not a disinterested observer of the fights I report on.
• I have money at stake. GreenMedInfo earns revenue from memberships, donations and product sales. I founded Regenerate Company, and I make money if you buy its products. Every role and interest is on the disclosures page.

What this means for you: when I cite a study, assume I chose it because it supports the point I am making. Do not treat its absence from my pages as absence from the literature.

Why I work this way

The charge is old and I have heard it often. In a March 31, 2021 article, McGill University’s Office for Science and Society called GreenMedInfo “a colossal exercise in the cherry-picking of scientific studies,” curated by a man with “a bachelor’s degree in philosophy.” Both facts are accurate. I have a philosophy degree from Rutgers, and the index is curated.

My answer, in an August 2025 essay called “The Orchard of Deception,” is that every body of evidence is picked from a larger orchard, and the honest question is whether the picking is declared and whether the orchard itself is balanced. I make three arguments, and I limit myself here to the parts that can be checked.

1. Published evidence is itself selected. In a 2008 analysis of FDA records for 74 antidepressant trials, Turner and colleagues found that 31 percent were never published. The published literature made it appear that 94 percent of trials were positive. The FDA’s own review showed 51 percent. Journal data inflated the apparent effect size by 32 percent overall. Ioannidis (2005) argued on statistical grounds that most published research findings in many fields are likely to be false.
2. Money follows what can be patented. The National Center for Complementary and Integrative Health, the NIH’s lead agency for this field, has had a budget of about $170 million in recent years, against a total NIH program level of roughly $48 to $49 billion, or about 0.35 percent (NIH FY 2025 budget tables). Other NIH institutes also fund some natural-product research, so this understates the total, but it shows the scale.
3. A declared counterweight is better than a hidden tilt. If the mainstream literature under-samples natural interventions, someone has to collect the studies it leaves out. GreenMedInfo says that is its purpose. I do not claim neutrality. I claim to be open about what the selection is.

What I do to keep the bias from misleading you

• Link to the study. GreenMedInfo entries open the underlying PubMed record so you can read the study and judge it.
• Source and date every figure. Each number on this site names its source and the date it was checked.
• Quote exactly, attribute carefully. Quotations are checked against the original. A claim that rests on a person’s own account is marked “by his account” or “she said.”
• Separate what a document shows from what I infer. In my investigative work, a claim about what a record says is footnoted to the record. My interpretation is labeled as interpretation.
• Publish failures. The launch essay for Scalar Flower leads with a result that fell from p = 0.003 to p = 0.83 under a control committed to in advance. I count that as the method working.
• Disclose interests where they apply. A page about a product I sell says so on the page.

How to read my work

• Open the study. Check the design (trial, cohort, case report, lab experiment) and the sample size.
• Check whether it was done in people, animals or cells. Cell and animal findings often do not carry over.
• Look for the opposing evidence. If you cannot find any, search for it yourself.
• Check who funded it, and who wrote it.
• Treat my headlines as arguments, and the footnotes as the evidence.
• Nothing here is medical advice. Talk to a clinician who knows your history.

Corrections policy

What counts as a correction: a factual error, a misquotation, a wrong date or figure, a broken or misread source, or a claim that has been overtaken by a later record. A disagreement over interpretation is not a correction, though I will read it and may answer it.

What I do: I check the claim against a primary source, fix the page, and record the change below with its date and the reason. I do not edit silently. If a correction changes the meaning of a page, a note appears on that page.

How to report one: send the page, the passage and a source to sayerji@greenmedinfo.com. For corrections to GreenMedInfo articles, use the GreenMedInfo corrections page.

Corrections log

1. Oct 4, 2026

   Added this methodology and bias statement to the corrections page.

2. Oct 4, 2026

   Site launched. Stand for Health Freedom is described as “founding board member (2019–2024)” to match the organization’s own About page, which lists other people as co-founders. Earlier biographies of mine use “co-founder.”

Last reviewed October 4, 2026.