Book
The Real Anthony Fauci: Bill Gates, Big Pharma, and the Global War on Democracy and Public Health
Kennedy argues that Anthony Fauci, NIAID director from 1984 to 2022, used NIAID funding and regulatory influence to serve pharmaceutical companies and the Gates Foundation at the cost of public health. Chapters 1 and 2 cover COVID-19: masks, lockdowns, hydroxychloroquine, ivermectin, remdesivir and the vaccine trials. Chapters 3 to 6 cover AIDS-era AZT, NIAID's trial network and the dissent of Peter Duesberg. Chapters 7 and 8 cover NIAID-linked trials on foster children and in Africa. Chapters 9 and 10 cover the Gates Foundation, WHO and Gavi, chapters 11 and 12 earlier epidemic alarms, gain-of-function funding and biodefense exercises. Endnotes follow each chapter; the Publisher's Note refers to 'thousands of references and citations'.
Who made it
- Author Robert F. Kennedy Jr. is 'the chairman of the board and chief legal counsel for Children's Health Defense', counsel to Morgan and Morgan, and founder of Waterkeeper Alliance (Skyhorse author biography, read 10 November 2026).
- Skyhorse Publishing lists the book in its Children's Health Defense series; publication date 16 November 2021, 480 pages, hardcover and ebook, $32.50 (Skyhorse page, read).
- The Publisher's Note is signed by Tony Lyons of Skyhorse and gives the URL ChildrensHealthDefense.org/fauci-book for updates; the Afterword closes with the author's description of founding Children's Health Defense and an invitation to join it (print text).
- The Dedication thanks physicians and scientists the book relies on for its COVID-19 early-treatment claims, including Pierre Kory (listed as president and chief medical officer) and Paul Marik (listed as founder) of the Front-Line COVID-19 Critical Care Alliance, Peter McCullough, Harvey Risch and Tess Lawrie (print text).
- Tony Lyons told The Guardian in January 2022 that the book had sold 750,000 copies in eight weeks (Guardian interview, 27 January 2022).
Its claims checked against the record
73 claims checked:5 supported · 44 partly supported · 19 contradicted · 5 not in the record
A verdict rests on the documents named in its row.
| The work says | Verdict | What the record shows |
|---|---|---|
| Hydroxychloroquine was effective and safe against COVID-19 when given early, and 31 of 32 early outpatient studies show benefit; the randomized trials that found it ineffective tested it only late, in hospitalized patients. | Partly supported | The book is right that RECOVERY and Solidarity enrolled hospitalized patients. Three randomized trials tested early use: Boulware gave it within 4 days of exposure to 821 people (11.8 percent vs 14.3 percent developed illness, P=0.35); Skipper enrolled outpatients within 4 days of symptoms (symptom-severity change, P=0.117, 43 percent vs 22 percent adverse effects); Mitja enrolled outpatients under 5 days from onset (hospitalization 5.9 percent vs 7.1 percent, no viral-load difference). None found a benefit. The '31 of 32' tally comes from the c19study.com aggregator, which is not a record. |
| RECOVERY and WHO Solidarity gave hydroxychloroquine at 'borderline lethal' doses (2,400 mg on day 1, then 800 mg a day) in trials designed by Gates Foundation staff to make the drug look dangerous, and the drug arm had 10 to 20 percent more deaths. | Partly supported | The doses match: RECOVERY gave 800 mg at 0 and 6 hours, then 400 mg every 12 hours (2,400 mg in the first 24 hours); Solidarity gave 800 mg at 0 and 6 hours, then 400 mg twice daily for 10 days. A Gates Foundation program officer attended the four-person WHO consultation of 8 April 2020, whose record says 'The selected doses are undoubtedly high' and that serious toxicity was 'unlikely'. Mortality was 27.0 percent vs 25.0 percent in RECOVERY (rate ratio 1.09, 95% CI 0.97-1.23) and 104 of 947 vs 84 of 906 in Solidarity (1.19, 0.89-1.59); neither difference is significant. The FDA's own memorandum found the EUA doses too low to reach antiviral concentrations. No document describes the doses as lethal or the design as intended to harm. |
| The Lancet and NEJM Surgisphere papers on hydroxychloroquine were the basis on which FDA withdrew the hydroxychloroquine EUA on 15 June 2020, and both were retracted as frauds. | Contradicted by the record | The retractions are real: authors lacked access to the raw data and Surgisphere would not transfer it to an auditor; three of the Lancet paper's four authors requested retraction. The NEJM paper concerned cardiovascular drugs and mortality, not hydroxychloroquine. FDA's memorandum states of the Lancet paper that it 'was subsequently withdrawn by the authors and will not be included in this Memorandum', and rests on FDA's pharmacology re-analysis, non-replicated viral-shedding data, NIH guidance and the preliminary RECOVERY result. |
| The Brazilian chloroquine trial published in JAMA Network Open used a lethal dose, killed 39 percent (16 of 41) of the high-dose arm, and was run by Gates-funded investigators. | Partly supported | The numbers match: 16 of 41 (39.0 percent) deaths in the 600 mg twice daily arm vs 6 of 40 (15.0 percent) in the low-dose arm, and the safety board recommended stopping the high-dose arm. The paper reports the association was not significant after age adjustment, lists funders that do not include the Gates Foundation (the book itself says Gates 'did not fund the JAMA study directly'), and does not use the words homicide or poisoning. |
| Ivermectin cut COVID-19 deaths by 62 to 86 percent in more than 20 randomized trials and six of seven 2021 meta-analyses, and prophylaxis protected 6 of 7 people; Hill's WHO meta-analysis and Lawrie's review showed large mortality benefits. | Contradicted by the record | The Cochrane review published four months before the book (14 trials, 1,678 participants) rated the mortality evidence very-low certainty (inpatients RR 0.60, 95% CI 0.14-2.51). The Hill meta-analysis was retracted in February 2022 because one of its largest component studies 'was withdrawn due to fraudulent data'; the Bryant/Lawrie paper carries an Expression of Concern stating that excluding the suspect data 'appears to invalidate' its mortality finding. Large placebo-controlled trials published after the book found no benefit: ACTIV-6 (1,591 participants at 400 ug/kg, median recovery 12 vs 13 days; 1,206 at 600 ug/kg, HR 1.02) and TOGETHER (679 vs 679, primary outcome 14.7 percent vs 16.3 percent). |
| Hector Carvallo's randomized placebo-controlled Argentine trial found 100 percent efficacy: no infections among 788 workers on weekly ivermectin versus 58 percent of 407 controls. | Contradicted by the record | The synopsis describes 229 health workers (131 on ivermectin plus carrageenan, 98 controls on standard care), with 0 vs 7 positive swabs; the control group received standard care, not placebo. The 788, 407 and 58 percent figures do not appear in it. BuzzFeed reports a published version with different counts and a hospital that said it had no record of the study; a registry entry is not in the documents cited. |
| The NIH COVID-19 Treatment Guidelines Panel moved ivermectin to 'neither for nor against' on 14 January 2021, hid its members' names, and one member, Susanna Naggie, then received a $155 million grant for ivermectin studies. | Partly supported | The statement says there are 'insufficient data to recommend either for or against the use of ivermectin'. NIH's April 2021 release announced 'an initial investment of $155 million' for ACTIV-6, a platform trial of several repurposed drugs coordinated by Duke, where Naggie was principal investigator; the money funded a multi-drug platform trial in which ivermectin was one arm. The panel roster, the FOIA releases and the court filings on member secrecy were not retrievable. |
| In late August 2021 Fauci spread a false story that poison-control centers in Mississippi and Oklahoma were flooded with ivermectin overdoses; officials in both states quickly denied that anyone was hospitalized. | Partly supported | The health department's alert said 'At least 70% of the recent calls have been related to ingestion of livestock or animal formulations'; on 25 August the state epidemiologist stated ivermectin calls were about 2 percent of all poison-control calls, so the book is right that the 70 percent figure was misread. The department also confirmed at least one hospitalization for ivermectin toxicity, so the book's statement that Mississippi officials denied any is wrong. Sequoyah Northeastern Health System said it had treated no ivermectin patient; a second hospital where the source physician worked said it had seen a handful. per the book the 'calls are spiking in places like Mississippi and Oklahoma' line is the host's. |
| NIAID's remdesivir trial used a fake 'active comparator' placebo containing sulfobutyl, changed its endpoint twice mid-trial to hide the lack of a mortality benefit, and Fauci announced the result before publication. | Partly supported | The paper states the primary outcome 'was initially a comparison of clinical status at day 15' and was changed to time to recovery on 22 March 2020, after 72 patients were enrolled, 'by trial statisticians who were unaware of treatment assignments'. That is one change, made before any outcome data existed. The control was 'a matching placebo' (saline at some sites); no sulfobutyl comparator is described. Results: median recovery 10 vs 15 days; 29-day mortality 11.4 percent vs 15.2 percent (HR 0.73, 95% CI 0.52-1.03, not significant). The 29 April 2020 announcement of preliminary results preceded publication, as the book says. |
| NIAID's 2019 Ebola trial pulled remdesivir for lethal toxicity (54 percent of the remdesivir group died, with multiple organ failure and kidney failure), so Fauci knew remdesivir was 'hideously dangerous' before enrolling COVID-19 patients. | Contradicted by the record | Twenty-eight-day mortality was 53.1 percent (93 of 175) on remdesivir, 49.7 percent (84 of 169) on ZMapp, 35.1 percent on MAb114 and 33.5 percent on REGN-EB3. The monitoring board ended assignment to remdesivir and ZMapp because the other two drugs were 'superior to ZMapp and remdesivir with respect to mortality'. After adjudication, four events in three patients were judged possibly related to study drugs, one in the remdesivir arm (hypotension during the loading dose). |
| Remdesivir has no clinical efficacy against COVID-19, according to every legitimate study; the Chinese Lancet trial was shut for potentially lethal side effects (12 percent vs 5 percent serious injuries) and WHO's 11,266-patient trial found no benefit. | Partly supported | Solidarity (11,330 randomized, 405 hospitals) found remdesivir deaths of 301 of 2,743 vs 303 of 2,708 (RR 0.95, 0.81-1.11) and 'little or no effect' on mortality, ventilation or hospital stay, as the book says. Wang et al. ended early when Wuhan enrolment fell; serious adverse events were 18 percent on remdesivir and 26 percent on placebo, and the 12 percent vs 5 percent figures are discontinuations because of adverse events (18 vs 4 patients). ACTT-1 reported faster recovery (rate ratio 1.29), so 'no efficacy in every study' is wrong for time to recovery. Solidarity appeared as a preprint in mid-October 2020 and in NEJM on 2 December 2020. |
| In Pfizer's six-month trial data far more people died in the vaccine group: 20 all-cause deaths against 14 on placebo (42.8 percent more), with five deaths left out of Table S4, four to five times the cardiac deaths, and 1 COVID death against 2. | Contradicted by the record | The paper reports: 'During the blinded, placebo-controlled period, 15 participants in the BNT162b2 group and 14 in the placebo group died', then 3 vaccine-group and 2 original-placebo deaths after unblinding, with none judged related to the vaccine. The book's 20 adds all five post-unblinding deaths to the vaccine side, although two occurred in original placebo recipients. FDA's review counts 38 deaths from dose 1 to the 13 March 2021 cutoff, 21 in the vaccine group and 17 in placebo, 'None of the deaths were considered related to vaccination.' Table S4 itself could not be opened. |
| FDA allowed Pfizer to end a three-year trial after six months and vaccinate the placebo group so long-term harms could not be detected; the vaccine prevents one COVID death per 22,000 recipients, and relative-risk reporting hid this. | Partly supported | FDA's document says participants were 'expected to participate for a maximum of approximately 26 months' (the book says three years) and that since 14 December 2020 participants were 'progressively unblinded' and placebo recipients offered vaccine; the blinded comparison therefore ended after about six months, as the book says, while follow-up of the original cohort continues. The 1-in-22,000 figure rests on 2 vs 1 COVID deaths, an outcome the trial was not powered to measure; the paper reports efficacy of 91.3 percent against COVID-19 and 96.7 percent against severe COVID-19. Table S4 could not be opened. |
| A 2010 HHS study showed VAERS detects 'fewer than 1 percent of vaccine injuries'; AHRQ's automated Harvard Pilgrim system showed vaccines causing serious injury in 1 of every 40 recipients, and CDC killed the project. | Contradicted by the record | The report cites 'fewer than 1% of vaccine adverse events are reported' as background from earlier literature and does not present it as a finding. Its results: 715,000 patients and 1.4 million doses of 45 vaccines; 35,570 possible reactions (2.6 percent of vaccinations) flagged by automated rules for clinician review, which are candidate events flagged by code; none was confirmed as an injury. The report says the planned evaluation against VAERS and the Vaccine Safety Datalink did not happen because 'due to restructuring at CDC and consequent delays... it became impossible to move forward'. It does not say CDC terminated the project. |
| COVID-19 deaths and hospitalizations are higher among the vaccinated than the unvaccinated in the United States and abroad (UK ONS: death rate higher among the vaccinated by October 2021), which shows 'pathogenic priming' has arrived. | Contradicted by the record | CDC's surveillance in 13 U.S. jurisdictions found age-standardized death-rate ratios (not fully vaccinated vs fully vaccinated) of 16.6 for 4 April-19 June and 11.3 for 20 June-17 July 2021, with 9 percent of 6,748 deaths in fully vaccinated people. In the Israeli data presented to FDA, 60 percent of severe active cases were fully vaccinated in a population that was over 60 percent doubly vaccinated, which the presenter used to argue for third doses. Dalrymple reports it contains numbers and no rates and that its age adjustment split the population at 50. No document opened shows antibody-dependent enhancement in COVID-19 vaccine recipients. |
Show the other 58 claims checked
| The work says | Verdict | What the record shows |
|---|---|---|
| Israel's director of public health told FDA's advisory panel on 17 September 2021 that vaccines cannot stop transmission, sickness or death: 60 percent of people in severe and critical condition and 45 percent of those who died in the fourth wave were doubly vaccinated. | Partly supported | Dr Sharon Alroy-Preis said that 60 percent of severe cases were fully vaccinated. On deaths she said about 1,000 people died in the fourth wave and 'But 40 percent of them are unvaccinated and 54 percent of them received two doses and did not have the chance to receive the third dose yet', so the book's 45 percent is a misquote. In the passages read she frames the rise as 'Is that a Delta issue, or is that a waning immunity issue?', and the meeting concerned a third dose. The transcript was searched by keyword and not read in full; no passage located has her saying vaccines cannot reduce death. |
| Fauci's private emails show he believed ordinary masks had little or no efficacy; his later explanation that he downplayed masks to protect supply was a cover. | Supported by the record | The email excerpt reads 'The typical mask you buy in the drug store is not really effective in keeping out virus, which is small enough to pass through the material.' On 8 March 2020 Fauci told 60 Minutes there was 'no reason to be walking around with a mask' and gave both the protection and the shortage reasons publicly. The book's description of the email is accurate. |
| Fauci raised his herd-immunity estimate from 70 percent to 80-90 percent 'not based on science but in response to polling' and admitted to the New York Times that he had told two 'noble lies'. | Partly supported | The Times quote as read on air: 'when newer surveys said 60 percent or more would take it, I thought, I can nudge this up a bit. So, I went to 80, 85.' Asked whether he based the figure on what people could accept, Fauci answered 'No, I mean, it's a bit of that', and described the numbers as 'pure estimates' extrapolated from measles. He gave 85, not 90, and did not use the word 'lie'. |
| Fauci said in July 2020 that PCR positives at high cycle thresholds are 'just dead nucleotides, period' yet did nothing, and CDC admitted in August 2021 that its PCR tests could not distinguish COVID-19 from other viral illnesses. | Partly supported | The Fauci quote is attested ('if somebody does come in with 37, 38, even 36, you got to say, you know, it's just dead nucleotides, period'); the audio was not heard. The CDC alert is dated 21 July 2021 and says CDC will withdraw its EUA request for the original assay, 'first introduced in February 2020 for detection of SARS-CoV-2 only', and encourages a multiplex method that also detects influenza. It does not say the panel could not tell COVID-19 from other illnesses. |
| CDC's statement that 98 percent of hospitalizations and deaths were among the unvaccinated used January-June 2021 data, when almost no one was vaccinated, and CDC counts anyone as unvaccinated unless more than two weeks past the second dose. | Partly supported | CDC Director Rochelle Walensky's figure was 99.5 percent, taken first from 'preliminary data from a collection of states over the last six months' and then from 'a five-month study from January to May'; CDC did not publish the underlying data. MMWR 70(37) defines fully vaccinated as 14 or more days after the last recommended dose and counts the rest as not fully vaccinated. The 5 August video the book cites was not located. Record not produced. CDC's underlying data for Director Walensky's 99.5 percent unvaccinated-share figure (preliminary state data, then a five-month study) was never published, as PolitiFact documents in its 22 July 2021 check quoting Walensky's 1 and 20 July 2021 statements. |
| The United States had 4 percent of the world's population but 14.5 percent of COVID-19 deaths, 2,107 deaths per million by 30 September 2021 against 139 per million in Japan, and its life expectancy fell 1.9 years, 8.5 times the peer-country average; because COVID-19 killed mainly the elderly, lockdowns must explain the loss. | Partly supported | The counts match: 696,889 of 4,812,680 global deaths (14.48 percent), about 2,093 and 140 deaths per million, and Woolf's 1.87-year fall, 8.5 times the peer average, with 3.88, 3.25 and 1.36 years for Hispanic, Black and White Americans. The lockdown explanation is not in either source: NCHS attributes 73.8 percent of the 2019-2020 decline to COVID-19 (90.0 percent for Hispanic people; 59.3 percent for Black people, with unintentional injuries 11.9 percent and homicide 7.7 percent). |
| Under Fauci, chronic disease 'mushroomed to afflict 54 percent of children, up from 12.8 percent' in 1984, and autism rose from 2-4 per 10,000 to one in 34. | Not in the record | The 54 and 12.8 percent figures cite an advocacy-group article. CDC's latest autism estimate at publication was 18.5 per 1,000 (about 1 in 54) for 8-year-olds in 2016; the estimate for 2018, published 3 December 2021 after the book, was 23.0 per 1,000 (1 in 44). One in 34 is not a CDC national figure. |
| Fauci 'directly and indirectly' controls 57 percent of global biomedical research funding through NIH, the Gates Foundation and the Wellcome Trust. | Contradicted by the record | The cited article says 'NIH and the Gates Foundation together represent 57 percent of all funding for global-health research and development' (about 40 percent NIH, 17 percent the Foundation). It does not mention the Wellcome Trust, concerns global-health R&D rather than all biomedical research, and says nothing about control. NIAID is one NIH institute. |
| Fauci and NIAID officials may pocket up to $150,000 a year in royalties on products they help develop under a 'secretive, unpromulgated HHS policy', and 45 percent of FDA's budget comes from the pharmaceutical industry through user fees. | Partly supported | The cap is statutory: 'Payments made under this section shall not exceed $150,000 per year to any one person, unless the President approves a larger award', paid to inventors or co-inventors whose rights were assigned to the government. FDA at a Glance states 45 percent of the FY2020 budget ($2.7 billion) is paid by industry user fees, a total that includes tobacco fees. The amount Fauci received is not in any document opened. |
| NIAID, under Fauci, funded EcoHealth Alliance coronavirus research that included subawards to the Wuhan Institute of Virology, and the funding carried 'gain-of-function' experiments there. | Supported by the record | OIG Table 1 lists grant R01AI110964, 'Understanding the Risk of Bat Coronavirus Emergence', awarded $3,748,715 from FY2014 with the Wuhan Institute of Virology as a subrecipient; GAO counts three WIV subawards totaling $1,413,720 disbursed. NIH's 20 October 2021 letter (via FactCheck.org) says a 'limited experiment' under the grant made mice sicker with a chimeric bat coronavirus than with WIV1 and that EcoHealth failed to report it right away. In a 2024 interview quoted in the House Select Subcommittee staff report, NIH's Lawrence Tabak accepted the NIH definition of gain-of-function ('research that modifies a biological agent so that it confers new or enhanced activity') and called the experiment 'certainly an example of generic gain-of-function'. Fauci told the Subcommittee he used only the narrower P3CO definition; NIH did not refer the work for P3CO review, and OIG records the non-referral and says it does not evaluate it. The book gives none of these documents. |
| Fauci defied President Obama's 2014 gain-of-function moratorium, dodged NIH's internal safety review, and laundered money to Chinese scientists with military affiliations through Peter Daszak's EcoHealth Alliance. | Partly supported | OIG records that NIAID wrote to EcoHealth on 13 May 2016 that the research 'could be gain-of-function and subject to the funding pause', and on 7 July 2016 determined it was not subject to the pause; in 2018 NIAID re-reviewed under the HHS P3CO framework and did not refer it. OIG states it does not evaluate those determinations. GAO counts two subawards totaling $514,129 to the Beijing Institute of Microbiology and Epidemiology, a subunit of the Academy of Military Medical Sciences, passed through U.S. university awards. The word 'laundered' and the DARPA payments are not in either report. |
| COVID-19 'emanated from a Fauci-funded Little Shop of Horrors in Wuhan'. | Not in the record | ODNI states the intelligence community 'remains divided': four elements and the National Intelligence Council assess natural exposure with low confidence, one element a laboratory-associated incident with moderate confidence, three are undecided. It also says a determination would need information that 'a laboratory in Wuhan was handling SARS-CoV-2 or a close progenitor virus before COVID-19 emerged'. GAO reports NIH's analysis found the bat coronaviruses in EcoHealth's reports genetically distant from SARS-CoV-2. Record not produced. The Wuhan Institute of Virology's laboratory notebooks and electronic files for the EcoHealth-subawarded work were never provided to NIH or EcoHealth; HHS OIG A-05-21-00025 records that WIV did not respond to the requests, and the HHS debarment memorandum (exh. Exhibit 116) lists EcoHealth's failure to take active steps to obtain them as a ground. |
| The U.S. government invested $6 billion in Moderna's COVID-19 vaccine, which Fauci's NIAID 'co-owns' as patent holder, and Fauci's deputies stand to collect royalties of $150,000 a year. | Partly supported | STAT's readable text says the federal government 'has now invested about $6 billion', of which about $5.75 billion went to Moderna to 'develop, test, and manufacture' the vaccine and secure 300 million doses. The readable text names no agency and does not mention NIAID. The patent dispute over NIAID scientists as co-inventors surfaced in 2021 and was not resolved in any document opened. |
| Fauci funneled $125 million to Moderna Therapeutics for an mRNA Zika vaccine, and Zika never caused microcephaly in the United States. | Contradicted by the record | The funder was BARDA (HHS ASPR) with an $8 million award and options up to $125 million; the release does not mention NIAID. Mlakar recovered the complete Zika genome from a microcephalic fetal brain. Honein, using the CDC U.S. Zika Pregnancy Registry, found Zika-associated birth defects in 26 of 442 (6 percent) completed pregnancies with laboratory evidence of Zika, with 18 infants having microcephaly. |
| The AZT phase 2 trial was a double-blind placebo-controlled study of 3,200 people, stopped early by Fauci's team with 19 deaths on placebo and 1 on AZT, and FDA then gave AZT a 'fast-tracked Emergency Use Approval' in March 1987. | Partly supported | Fischl randomized 282 patients with AIDS or advanced ARC, not 3,200, for up to 24 weeks; deaths were 19 on placebo and 1 on AZT, as the book says. The 20 March 1987 announcement describes an ordinary marketing approval limited to AIDS and advanced ARC patients with specified findings; the announcement does not use the term 'Emergency Use Approval'. FDA's table lists a review time of 3.5 months. |
| NIAID's official AZT report listed no adverse effects while AZT patients received six times as many blood transfusions as placebo patients (30 vs 5). | Partly supported | The companion toxicity paper reports repeated red-cell transfusions in 21 percent of AZT and 4 percent of placebo recipients (P<0.001), hemoglobin below 7.5 g/dL in 24 percent vs 4 percent, and neutropenia in 16 percent vs 2 percent; on 145 and 137 patients that is about 30 vs 5, matching the book's counts. The statement that the report listed no adverse effects is contradicted: the toxicity was published in full in the same issue. |
| At 1,500 mg a day AZT was 'absolutely fatal'; AZT, not HIV, caused most AIDS deaths from 1987 to 1996, and life expectancy rose in 1990 only because FDA lowered the dose. | Partly supported | NIAID's own dose trial of 524 patients found 18-month survival of 52 percent on 1,500 mg a day and 63 percent on the reduced dose (p=0.012), with hemoglobin below 8 g/dL in 39 percent vs 29 percent; the book is right that the high dose was more toxic. ACTG 019 found AIDS in 33 placebo patients against 11 on 500 mg and 14 on 1,500 mg. Mortality fell with combination therapy including protease inhibitors (HOPS deaths 29.4 to 8.8 per 100 person-years, 1995 to 1997; MACS AIDS hazard 0.35), which does not fit 'AZT caused the deaths'. |
| A Nature study of British hemophiliacs showed HIV-positive hemophiliacs began dying only in the year AZT was introduced. | Contradicted by the record | Among 2,448 severe hemophiliacs the annual death rate stayed at 8 per 1,000 in HIV-negative patients and rose to 81 per 1,000 in HIV-positive patients by 1991-92. Excess death rates in the HIV-positive group were 19, 34, 53 and 76 per 1,000 in 1985-86, 1987-88, 1989-90 and 1991-92, so excess mortality was present in 1985-86, before AZT was approved in March 1987. HIV-negative hemophiliacs, who received the same blood products, showed no rise. |
| The Concorde study showed 'the more AZT, the more death'. | Partly supported | Concorde randomized 1,749 symptom-free patients to immediate or deferred AZT (median follow-up 3.3 years): 3-year survival was 92 percent vs 94 percent (log-rank P=0.13) and progression to AIDS or death was 18 percent in both arms. Early AZT gave no survival benefit; the trial does not show excess deaths with more AZT. |
| No one has been able to point to a study showing that HIV causes AIDS; the hypothesis has not met Koch's postulates and HIV is a harmless passenger virus. | Contradicted by the record | Seropositive hemophiliacs died at about ten times the rate of seronegative hemophiliacs exposed to the same blood products (81 vs 8 per 1,000, 1991-92). PACTG 076 randomized 477 HIV-positive pregnant women: transmission to infants was 8.3 percent on AZT and 25.5 percent on placebo. The NIAID-funded MACS cohort showed the AIDS hazard falling to 0.35 after potent antiretroviral therapy. Chigwedere estimated more than 330,000 South African deaths in 2000-2005 from not implementing available treatment. The author writes 'I take no position on the relationship between HIV and AIDS' and, in a parenthesis, 'I believe that HIV is a cause of AIDS'. |
| HIV antibody and PCR tests do not detect HIV: antibody tests have no recognized standard and cross-react with tuberculosis, pregnancy and influenza, viral-load tests were 'never validated', and HIV has never been isolated. | Contradicted by the record | Burke tested 135,187 low-prevalence military applicants with repeat ELISA, Western blot and supplementary assays and measured a false-positive rate of 1 in 135,187. Rich describes three people falsely diagnosed by viral-load results during primary infection and advises caution with those assays; it does not say HIV was never isolated. Barre-Sinoussi 1983 reports isolation of a retrovirus and its transmission to cord-blood lymphocytes. |
| Poppers and recreational drugs caused Kaposi sarcoma and the early AIDS cases; in 1990 CDC scientists reported in the Lancet that Kaposi sarcoma was common in young gay men who did not have HIV; Fauci quashed the poppers hypothesis. | Contradicted by the record | Beral analyzed persons with AIDS reported to CDC and concluded Kaposi sarcoma 'may be caused by an as yet unidentified infectious agent'; it did not report the disease in HIV-negative young gay men. In 1994 Chang found herpesvirus DNA (later named KSHV) in over 90 percent of AIDS-associated Kaposi sarcoma tissue. In the San Francisco Men's Health Study, HHV-8 antibody was present in 37.6 percent of men with homosexual activity and none of 195 exclusively heterosexual men. In MACS (3,223 men) heavy poppers use was not significantly associated with Kaposi sarcoma (IRR 0.84, 95% CI 0.42-1.67). The quoted 1994 Gallo meeting could not be checked. |
| AIDS without HIV was known to Fauci and CDC and kept secret until 1992; CDC admitted 82 certified cases in 15 states, Duesberg documented over 4,000, and Fauci invented 'idiopathic CD4+ lymphocytopenia' to explain them away. | Partly supported | CDC itself published the cases: MMWR 1992 (26 case-patients, 'no epidemiologic linkage', cause unknown) and NEJM 1993 (47 patients found among 230,179 AIDS reports; 29 had no HIV risk factor; 23 contacts and blood donors tested normal, so no new transmissible agent was found). 'Kept secret' is contradicted by those publications. The figures 82 and 4,000, and that Fauci coined the label, are not in the documents cited. |
| Gallo's 1984 Science paper found HIV in only 26 of 72 AIDS patients and was the 'only basis' for the HIV-AIDS claim; Gallo took Montagnier's virus, a French suit forced him to disgorge half his proceeds, and two U.S. inquiries examined his ethics. | Partly supported | Gallo's abstract reports virus isolated from 26 of 72 AIDS patients, 18 of 21 pre-AIDS patients, 1 of 22 normal homosexual men and 0 of 115 heterosexual controls, with antibodies in most AIDS patients; 26 of 72 is not the only basis. Montagnier's 1983 paper describes a virus isolated from one patient, which contradicts the book's '72 percent of 44'. ORI found Gallo guilty of misconduct on 29 December 1992 and withdrew the finding on 12 November 1993 after the Popovic appeal; the book omits the withdrawal. The 1987 Reagan-Mitterrand agreement divided royalties between the governments. 'Stole' is not a finding in any document opened. |
| Fauci never funded to completion a single grant exploring whether HIV causes AIDS, and NIAID funded only research assuming Gallo's virus is the cause. | Contradicted by the record | The Multicenter AIDS Cohort Study followed HIV-positive and HIV-negative men over time and produced the time-to-AIDS and mortality data in Detels 1998; it is funded mainly by NIAID. The narrower statement that NIAID cancelled its CHIPS contract in 1996 was not checked. |
| Fauci rigged NIAID's trial of AL 721 to fail and cancelled it when it confirmed AL 721 stopped viral replication, after Gallo and Laurence had found it effective. | Contradicted by the record | The NIAID AIDS Clinical Trials Group study (40 men, AL 721 for 8 weeks) found it well tolerated with 'no sign of immune restoration or antiviral effect'; p24 antigen stayed positive in all three antigen-positive men, and the authors state the data 'provide no support for the use of this compound as an antiretroviral agent'. The quotations attributed to Fauci about rigging the trial come from Nussbaum's book and were not checked. |
| For years Fauci refused to test or recommend aerosol pentamidine and Bactrim against Pneumocystis pneumonia, at a cost of 17,000 lives; at a 28 April 1988 hearing he admitted he would take them himself. | Partly supported | A community-based trial begun in 1987 (408 patients, 12 centers) found aerosolized pentamidine effective (8 PCP episodes on 300 mg vs 22 on 30 mg, p=0.0008). CDC and NIH issued PCP prophylaxis guidelines in 1989, which cuts against the statement that no guidance was ever issued. The hearing transcript, the 17,000 estimate (attributed to unnamed activists) and the funding of the community trial could not be verified. |
| In 1987 the Wall Street Journal won a Pulitzer Prize for investigating an HHS scheme to misrepresent AIDS as a general pandemic to win funding. | Not in the record | No Wall Street Journal winner for AIDS reporting appears in the 1987 list. The Journal's AIDS-related prize in the lists opened is the 1997 National Reporting award 'for its coverage of the struggle against AIDS in all of its aspects'. The cited Kaiser page does not mention the Journal or a Pulitzer. |
| Between 1985 and 2005 at least 532 New York City foster children were enrolled in NIAID-linked HIV trials; 80 died in foster care and 25 while enrolled in a medication trial; trials proceeded without the city's required review and with consent defects. | Supported by the record | Executive Summary: 532 children in 88 studies, 1985-2005; 'Eighty of the 532 children... died while in foster care; 25 of them died while enrolled in a medication trial.' Sixty-four children were in 30 medication trials not reviewed by the city's medical advisory panel and 21 in three trials the panel reviewed and did not recommend; 21 percent of in-care medication-trial enrollments lacked a signed consent form on file, at least 7 consents were signed by people not legally authorized, and 30 percent of child-welfare files were partly lost. The same summary states Vera 'did not find, however, that any child's death was caused directly by clinical trial medication', and that the 64 percent Black and 30 percent Latino enrollment 'paralleled the demographics of children with HIV infection in New York City', which contradicts the book's discrimination inference. |
| The federal Office for Human Research Protections found that drug companies, principal investigators and government officers failed to secure independent advocates and equitable selection for foster children, showing a 'toxic culture' at NIAID. | Partly supported | The letters record OHRP's determinations that the institutional review board 'failed to obtain sufficient information regarding the selection of wards of the state and foster children', the parental-permission process and safeguards (45 CFR 46.111). The findings name hospital review boards; none names NIAID or a drug company, and both letters accept corrective actions and close the matter. Fauci is on the copy line of both. |
| NIAID's Uganda trial HIVNET 012 dropped its placebo group once in the field, leaving an AZT-versus-nevirapine comparison, to mask injuries; the Institute of Medicine panel that reviewed it was a whitewash. | Partly supported | IOM: the trial had two placebo arms; after a Thai trial reported in February 1998, 'HIVNET 012 researchers formally dropped the placebo arms in a letter of amendment', which IOM found 'was reviewed and approved appropriately'; the modified trial 'was not able to demonstrate the safety of the two active regimens relative to untreated controls', as the book argues. IOM also found 'no evidence of misrepresentation of the study results' and recorded lapses (no guardian consent where the mother died during follow-up, some under-reporting of adverse events). The report is dated 2005, not 2004, and lists no grant information, so the book's 'six of nine members were NIAID grantees' could not be checked. |
| The Westat audit found the Uganda team lost the logbook recording adverse events and deaths, thousands of unreported adverse events, no Good Clinical Practice training, and under-reporting of at least fourteen deaths. | Partly supported | Westat: the investigators 'noted that many ('thousands') of unreported AE's and SAE's occurred', 'had never had GCP training, and had never attempted a Phase III trial', and 'Total outcomes were clearly under-reported'. What was lost was one of two Master Logs, recently rewritten after a 'flood'; Westat found previously unreported deaths in the visiting nurses' cards. IOM concluded the flooded logbooks were secondary materials and found 'no evidence of and only a very limited opportunity for either unreported deaths or erroneous reports of deaths'. The 'fourteen deaths' figure is not in the documents cited. |
| Investigators applied lenient seriousness definitions in Uganda (life-threatening events scored 'not serious') without ethics-board approval, and OHRP documented it. | Supported by the record | OHRP found 'no documentation that the investigators obtained IRB approval for the changes in reporting criteria and toxicity severity grading' and stated the alteration 'may have represented a failure to minimize risk to the subjects'; it also found the Ugandan board failed continuing review and record-keeping. IOM later 'finds no evidence that the definitions used... placed human subjects at increased risk'. |
| NIAID's Division of AIDS director Edmund Tramont rewrote Dr. Betsy Smith's safety review of HIVNET 012 and reopened the Uganda site over Dr. Jonathan Fishbein's objection days before President Bush's visit. | Supported by the record | AP, quoting NIH records: Tramont wrote on 23 January 2003 'I need to see the primary data - too much riding on this report'; the report that reached FDA said the study 'has demonstrated the safety of single dose nevirapine'; asked how Smith's report was changed, Tramont said 'I wrote it'. On 8 July 2003 he wrote 'I am convinced that this site is ready to resume' after Fishbein wrote 'I am not convinced that the site is indeed prepared to become active'. NIH acknowledged both actions and called the overruling an 'honest difference of opinion'. The book's added claim that the rewrite led to an FDA approval in December 2002 is contradicted by the dates. |
| Nevirapine, 'which had never won FDA approval as a safe and effective drug', was approved for preventing mother-to-child transmission on the strength of the manipulated Uganda data and shipped worldwide by WHO. | Contradicted by the record | FDA approved Viramune (nevirapine) on 21 June 1996 for treatment of HIV-1. The supplemental mother-to-child indication was withdrawn by Boehringer Ingelheim in March 2002 and, per IOM's Executive Summary, was never approved; the book states both that it was withdrawn and that FDA approved it in December 2002. The WHO/UNAIDS statement says the irregularities 'did not warrant any change in the recommendations'. |
| Joyce Hafford died of nevirapine liver failure in a DAIDS-sponsored Memphis trial, her family was told she died of AIDS, and DAIDS staff knew the drug caused it. | Partly supported | AP reports NIH files list 'drug-induced hepatitis' as the cause of death on 1 August 2003, that NIH's review found the Memphis hospital delayed reviewing labs, that the family was told she died of AIDS complications until AP gave them NIH documents, and that NIH acknowledged the drugs, 'most likely nevirapine', caused the death. |
| Dr. Jonathan Fishbein was fired by Fauci's circle for raising problems with nevirapine and interleukin-2 trials, and the Merit Systems Protection Board found 'wrongful retribution'. | Partly supported | NIH fired Fishbein for 'poor performance' weeks after he was recommended for a bonus; an internal NIH report found the effort gave the 'appearance of reprisal'; he was reinstated effective 12 December 2005. Government Executive says the MSPB ruled on whistleblower coverage for Title 42 hires and does not say it decided retaliation, so the book's 'wrongful retribution' finding is not in the records. On interleukin-2, AP reports Fauci and Lane each received $45,072.82 in royalties from 1997 and that Fauci said 'I'm going to give every penny of it to charity'. |
| Gates is the unofficial top sponsor of WHO: for 2018-2019 the U.S. gave $604.2 million, the Gates Foundation $431.3 million and Gavi $316.5 million, and Gates's total exceeds $1 billion. | Partly supported | The ranking of U.S., Gates and Gavi as the three largest voluntary contributors holds in outline (2018 and 2019 donor figures compiled by KEI from the WHO annexes: U.S. about $615 million, Gates about $455 million, Gavi about $389 million), but the book's three dollar figures do not match WHO's annexes and omit U.S. assessed contributions. For 2020-2021 WHO's own table ranks Germany first ($721.3 million), the Gates Foundation second ($375.5 million), the U.S. fourth ($340.2 million) and Gavi seventh. Kennedy and Thakrar (BMJ Global Health, 2025) put the Foundation's share at 9.5 percent of WHO revenue in 2010-2023 and report that all 640 of its grants are earmarked. I found no WHO document putting Gates's total above $1 billion for the biennium. |
| Over half of WHO's total budget now goes to vaccines, and WHO spent half of its total budget on polio at Gates's direction. | Contradicted by the record | The approved 2020-2021 budget was $5,840.4 million: base programmes $3,768.7 million, polio eradication $863.0 million (14.8 percent), emergency operations $1,000.0 million, special programmes $208.7 million. Polio is the largest vaccine line and is under 15 percent. No WHO document opened puts vaccines above half the budget. |
| In 2011 Gates ordered all 193 WHO member states to 'make vaccines a central focus of your health systems' and said 'Our priorities are your priorities'. | Partly supported | The speech reads 'Our priorities are your priorities - to make motherhood safer; to ensure that newborns survive their first 30 days; to provide children with a nutritious diet', followed by HIV, TB and malaria. The vaccine sentence is 'And all 193 Member States, you must make vaccines a high priority in your health systems', which closes a list of asks to donors and industry. |
| Gavi receives $1.16 billion a year from the U.S. government, five times what Gates gives WHO, and Gates launched Gavi in 1999 with $750 million. | Partly supported | Gavi's release describes $1.16 billion for fiscal years 2020 to 2023, about $290 million a year. Gavi's donor profile (exh. Exhibit 059) states 'Its initial US$ 750 million commitment in 2000 was catalytic in establishing Gavi'. |
| Gates handpicked Tedros Adhanom Ghebreyesus as WHO Director-General in 2017, his key qualification was loyalty to Gates, he sat on the Gavi and Global Fund boards, and he ran a terror group. | Partly supported | Member states elected him at the World Health Assembly on 23 May 2017; Gavi says he sat on its board in 2008-2009 and the Global Fund reports he chaired its board from 2009. No document opened shows Gates selected him or supports the loyalty and terror-group assertions. |
| Gates and his foundation hold equity in Gilead, Biogen, AstraZeneca, Moderna, Novavax and Inovio, and stocks and bonds in Merck, GSK, Lilly, Pfizer, Novartis and Sanofi, so he profits from the vaccines he promotes; the foundation owns a $6.5 million stake in Gilead. | Partly supported | The foundation's 2016 Form 990-PF lists common stock in Merck (27,200 shares, $1,601,264), Novartis (79,812 shares, $5,818,930), Sanofi (16,181 shares, $1,312,445) and Amgen, and corporate bonds of Gilead ($2,964,306), GlaxoSmithKline, Merck, AbbVie and Amgen. It lists no Lilly, Pfizer, Biogen, AstraZeneca, Moderna, Novavax or Inovio, and no $6.5 million Gilead holding; the Trust's 13F filings from 2018 to 2021 list none of these companies. Direct equity investments in vaccine developers are documented: BioNTech $55 million (2019) and CureVac about $52 million (2015). |
| A 2017 Danish-authored study showed DTP vaccine raised mortality in girls tenfold; a WHO-commissioned review confirmed it and WHO and Gates ignored it. | Partly supported | Mogensen analyzed a natural experiment in Bissau in the early 1980s (allocation by birthday, not randomized): hazard ratio 5.00 (95% CI 1.53-16.3) for DTP with or without OPV, and 10.0 (2.61-38.6) in the DTP-only group; the abstract concludes 'DTP was associated with increased mortality; OPV may modify the effect'. Higgins (34 birth cohorts) reports a possible increase with DTP, relative risk 1.38 (0.92-2.08) from studies at high risk of bias, stronger in girls, and lower mortality with BCG and measles vaccine; the authors recommend on-schedule vaccination and call for randomized trials. |
| In the Gates-funded RTS,S malaria trial, 151 infants died, the vaccine doubled girls' mortality and raised meningitis ten-fold, and reactogenic rabies and meningitis vaccines were used as controls to mask harm. | Partly supported | The comparators were rabies (VeroRab) and meningococcal C (Menjugate) vaccines. Lancet 2015 reports meningitis as a serious adverse event in 22 children (11 and 10 in the two RTS,S arms, 1 in control) and overall serious adverse events 'balanced between groups'; the girls' mortality ratio of 1.91 (1.30-2.79) against 0.84 in boys is a post hoc finding. The 151 and 1,048 counts are not in the documents cited. WHO recommended the vaccine on 6 October 2021 after a pilot reaching over 900,000 children. |
| Gates operatives forcibly vaccinated African children with MenAfriVac in 2010 and about 50 of 500 vaccinated children were paralyzed. | Not in the record | The only cited source repeats a Chadian newspaper report about Gouro, Chad, dated December 2012. No clinical record of paralysis in 50 of 500 children was found. |
| India's polio campaign, funded by Gates, caused a vaccine-strain epidemic that paralyzed 491,000 children in 2000-2017; by 2018 WHO conceded 70 percent of global polio cases came from Gates's vaccines, and polio had disappeared from Congo, the Philippines and Afghanistan until Gates reintroduced it. | Partly supported | Dhiman et al. report 'an additional 491,000 paralyzed children above our expected numbers' of non-polio acute flaccid paralysis against an assumed baseline and state that an association 'does not prove a causal relationship'. The 2018 counts were 33 wild and 104 circulating vaccine-derived cases, about 76 percent vaccine-derived, so the 70 percent figure is the right size; vaccine-derived cases arise from oral vaccine in under-immunized populations. Afghanistan never interrupted wild poliovirus and reported 21 wild cases in 2018. WHO certified the South-East Asia Region polio-free in March 2014, not 2016. |
| Gates-funded HPV vaccine 'experiments' on 23,000 girls in India killed seven, severely injured 1,200, used forged consent, and drew excoriation from the Indian Parliament. | Partly supported | .13 states PATH, 'under the pretext of observation/demonstration project has violated all laws and regulations... its sole aim has been to promote the commercial interests of HPV vaccine manufacturers'..12 records 9,543 consent forms in Andhra Pradesh, 1,948 with thumbprints and 2,763 signed by hostel wardens. Seven deaths are recorded (five Gardasil, two Cervarix); the ICMR inquiry found alternative causes (fever, malaria, snakebite, pesticide poisoning, drowning) and 'unlikely to be related to the vaccine', and the Committee wrote a vaccine role 'cannot be ruled out'. The report says the project was 'reportedly funded by' the Gates Foundation. The 1,200 severe events are not in the documents cited. |
| In his 2010 TED talk Gates said new vaccines could lower world population by 10 to 15 percent, and that month he pledged $10 billion to WHO. | Partly supported | The sentence is accurate: 'if we do a really great job on new vaccines, health care, reproductive health services, we could lower that by, perhaps, 10% or 15%.' It comes in a talk on CO2 emissions, where population is one factor of an emissions equation and refers to a projection of about nine billion people. The $10 billion was announced at Davos on 29 January 2010 as a ten-year pledge for vaccine research and delivery. |
| Kenya's 2014 tetanus campaign was a sterilization program: WHO gave a five-dose schedule only to girls, laboratories found hCG in the vaccine, and WHO admitted developing sterility vaccines. | Not in the record | Kenyan Catholic bishops reported beta-hCG in some vials; a joint bishops-ministry committee later tested 59 vials and, per press accounts read in part, three vials from the bishops' own set were positive and the rest negative. No primary laboratory report was found. WHO describes three tetanus doses to women of childbearing age in high-risk districts; no record links the hCG contraceptive-vaccine research to the Kenyan campaign vaccine. |
| Gates's Depo-Provera program doubles HIV risk, Gates suppressed the finding, and WHO's reviews were bought. | Partly supported | Heffron (funded by NIH and the Gates Foundation) found HR 1.98 (1.06-3.68) for HIV acquisition with hormonal contraception generally in 3,790 serodiscordant couples. The randomized ECHO trial (7,715 women, published 2019, before the book) found HR 1.04 (0.82-1.33) for DMPA-IM against copper IUD with no significant difference, and the book does not mention it. No evidence of suppression or bribery was found in these sources. |
| The 1976 swine flu 'pandemic' produced one casualty, not the million predicted; officials told Congress the Fort Dix strain was the 1918 virus while knowing it was an ordinary pig virus; about 49 million were vaccinated, with Guillain-Barre, deaths and injuries. | Partly supported | The Fineberg quotation is verbatim. Krause, who ran NIAID at the time, says about 200 Fort Dix men were infected and 'with the exception of 1 or 2 deaths' the illness was mild; he writes the virus was thought to be 'a direct descendant' of the 1918 virus, and nothing opened shows officials privately knew otherwise. Krause gives 45 million vaccinated; Schonberger counted 1,098 Guillain-Barre onsets from October 1976 to January 1977, 532 in recent vaccinees, with just under one excess case per 100,000 vaccinations in adults. The 32 deaths, 4,000 injuries and 'seven times' flu figures are not in the documents cited. |
| In 2009 WHO declared a pandemic with fewer than 145 deaths worldwide and deleted 'enormous numbers of deaths' from its definition to trigger $18 billion in vaccine contracts, with advisers who had undisclosed industry ties. | Partly supported | WHO's update of 11 June 2009 reports 28,774 cases and 144 deaths in 74 countries. The Review Committee found one online WHO document described pandemics as causing 'enormous numbers of deaths and illness' while the official definition 'was based only on the degree of spread', that WHO altered online documents 'without notice or explanation', and that WHO showed 'lack of timely disclosure of relationships potentially constituting a conflict of interest among experts'. The same report says it 'found no evidence of attempted or actual influence by commercial interests'. The $18 billion figure is not in it. |
| The Gates/Fauci dengue vaccine Dengvaxia killed at least 600 Philippine children through 'pathogenic priming', and Fauci minimized the deaths. | Partly supported | WHO states that a November 2017 retrospective analysis 'showed an excess risk of severe dengue in seronegative vaccine recipients' of CYD-TDV, as Halstead had warned. CYD-TDV is Sanofi Pasteur's vaccine; the NIAID candidates (TV003/TV005) are separate products, and the book merges them. The figure of 600 deaths is an allegation: the Department of Health panel reviewing 14 deaths linked three as possibly vaccine-related, and no document opened establishes causation for 600. |
| The 2019 Crimson Contagion exercise 'precisely predicted' COVID-19: 110 million illnesses, 7.7 million hospitalizations and 568,000 deaths, with Fauci, Redfield and Azar taking part. | Contradicted by the record | The report describes a novel avian influenza H7N9 scenario exercised 13-16 August 2019 with '110 million forecasted illnesses, 7.7 million forecasted hospitalizations, and 586,000 deaths', which are scenario inputs for an influenza exercise; the exercise did not model SARS-CoV-2. The names Fauci, Redfield and Azar do not appear in the text searched. |
| Dark Winter (June 2001) was a Pentagon war game led and named by Robert Kadlec that rehearsed coercive quarantine, censorship, masks and lockdown, and the 2001 anthrax letters were traced to one of three labs. | Partly supported | The Dark Winter script is headed as the property of Johns Hopkins Center for Civilian Biodefense, CSIS, ANSER and the Memorial Institute for the Prevention of Terrorism; 'Kadlec' does not appear in it, the Pentagon is not named as sponsor, and the legal briefing covers forcible inoculation and quarantine, with masks appearing only as a supply shortage. The Amerithrax summary attributes the spores to flask RMR-1029 kept by Bruce Ivins at USAMRIID, says no 'aerosolizing enhancers (e.g., fumed silica, bentonite...)' were found, and names no list of three labs. The National Research Council found 'it is not possible to reach a definitive conclusion about the origins of the B. anthracis in the mailings based on the available scientific evidence alone'. |
| The Rockefeller Foundation's 1910 Flexner Report recentered medicine on germ theory and closed more than half of U.S. medical schools. | Contradicted by the record | The report was published in 1910 by the Carnegie Foundation for the Advancement of Teaching, as 'a comprehensive report to the foundation, by Abraham Flexner'. The Rockefeller Foundation was incorporated in 1913. |
| Gates has given at least $250 million to media organizations, including NPR, the BBC and The Guardian, to neutralize press scrutiny. | Partly supported | CJR reviewed nearly 20,000 grants through June 2020 and found 'more than $250 million going toward journalism', including $17.5 million to NPR since 2000. The purpose 'neutralize' is the book's inference and is not in CJR. |
What the record shows that it leaves out
- NIH and HHS records on the EcoHealth grant show the oversight the book alleges was thin: the Year 5 progress report was filed two years late and NIH did not follow up; NIH terminated the WIV subaward in August 2022; HHS debarred EcoHealth and Peter Daszak in January 2025. (HHS OIG A-05-21-00025, https://www.oversight.gov/sites/default/files/upload-reports/52100025.pdf; GAO-23-106119 app. II; exh. Exhibit 116)
- The Tabak letter of 20 October 2021, a month before the book, told Congress that a 'limited experiment' under the EcoHealth grant made mice sicker and was not reported immediately as the grant required. The book does not cite it. (FactCheck.org, October 2021, exh. Exhibit 117)
- FDA's own revocation memorandum of 15 June 2020 concluded the EUA hydroxychloroquine doses were too low to reach antiviral concentrations and that a substantial increase 'would not be acceptable due to toxicity concerns'. (https://www.fda.gov/media/138945/download)
- Randomized trials of early outpatient treatment and post-exposure prophylaxis with hydroxychloroquine (Boulware 2020, Skipper 2020, Mitja 2020) found no benefit; the book discusses only hospital trials as the evidence against the drug. (doi 10.1056/NEJMoa2016638; doi 10.7326/M20-4207; doi 10.1093/cid/ciaa1009)
- Cochrane's 2022 ivermectin update excluded 7 of 14 trials for lack of prospective registration or non-randomization; the Elgazzar preprint was withdrawn on 14 July 2021; the Hill meta-analysis was retracted. The book's ivermectin figures rely on the same literature. (https://pubmed.ncbi.nlm.nih.gov/35726131/; https://www.researchsquare.com/article/rs-100956/v2; https://pmc.ncbi.nlm.nih.gov/articles/PMC8826014/)
- NIH's COVID-19 Treatment Guidelines Panel recommended against ivermectin outside trials in April 2022 after ACTIV-6 and TOGETHER; the book predates these. (JAMA 2022;328:1595; N Engl J Med 2022;386:1721)
- ACTT-1 reported that 51 patients (4.8 percent) were unblinded and 26 placebo patients crossed over after the interim analysis, with sensitivity analyses giving similar results. The book presents the unblinding as Fauci's unilateral act. (Beigel et al., PMC7262788, Results)
- The Pfizer six-month paper reports 30 of 31 severe COVID-19 cases in placebo recipients and no myocarditis through six months; the book mentions neither. (Thomas et al., PMC8461570)
- MMWR 70(21) (28 May 2021) reported 10,262 vaccine-breakthrough infections among about 101 million fully vaccinated people as of 30 April 2021: 995 hospitalized and 160 deaths, median age of decedents 82. (https://www.cdc.gov/mmwr/volumes/70/wr/mm7021e3.htm)
- FDA's 2025 approval letter for Comirnaty records Pfizer's commitment to a randomized, double-blind, placebo-controlled study in healthy adults aged 50 to 64, and a 2026 FDA letter releases Pfizer from a separate placebo-controlled postmarketing commitment (post-COVID-19 vaccination syndrome or Long COVID symptoms). Both documents post-date the book and bear on its statements about placebo-controlled testing. (exh. Exhibit 020 (https://www.fda.gov/media/188439/download); exh. Exhibit 021 (https://www.fda.gov/media/194509/download))
- NCHS attributed 73.8 percent of the 2019-2020 U.S. life-expectancy decline to COVID-19 (90.0 percent for Hispanic people); the Introduction's life-expectancy section omits it. (https://www.cdc.gov/nchs/data/vsrr/vsrr015-508.pdf)
- PACTG 076 (1994): AZT in 477 HIV-positive pregnant women cut mother-to-child transmission from 25.5 percent to 8.3 percent; ACTG 019 (1990): AZT 500 mg a day cut AIDS cases in symptom-free patients (33 on placebo, 11 on 500 mg) and was better tolerated than 1,500 mg. The book discusses AZT toxicity and does not cite either trial. (N Engl J Med 1994;331:1173; N Engl J Med 1990;322:941)
- ORI withdrew its misconduct finding against Robert Gallo on 12 November 1993 after the Popovic appeal; the book reports the 1992 finding without the withdrawal. (https://ori.hhs.gov/images/ddblock/vol2_no1.pdf)
- The Vera Institute found no child's death was caused directly by trial medication, no evidence of incentive payments, and no evidence that children were removed from families for refusing consent; it also found no evidence foster children were specially targeted for trials. (Vera Executive Summary, findings 1, 2, 4, 6, 9, 11, 13)
- IOM found HIVNET 012 was subject to FDA regulations and IRB oversight and found no evidence of misrepresentation of the study results; it also documented real lapses (no guardian consent where the consenting mother died during follow-up, under-reporting of non-serious adverse events). (IOM 2005, Executive Summary, https://www.nationalacademies.org/read/11264/chapter/2)
- WHO's 2020-2021 contributor table ranks Germany first, the Gates Foundation second and the United States fourth; the book gives 2018-2019 shares only. (https://www.who.int/about/accountability/results/who-results-report-2020-mtr/budget-summary-details)
- The Parliamentary Standing Committee's 72nd report also faulted the Drug Controller General's office, ICMR and state governments for oversight failures, not only PATH. (72nd Report, recommendations)
- The ECHO randomized trial (2019) found no significant HIV difference between DMPA-IM and a copper IUD; WHO recommended RTS,S on 6 October 2021 after a pilot reaching over 900,000 children. (Lancet 2019 (PMID 31204114); https://www.who.int/news/item/06-10-2021-who-recommends-groundbreaking-malaria-vaccine-for-children-at-risk)
Documented efforts and withheld records
Each item is a documented act, with its source: an act against the work, its author or its sources, or the withholding of records on its subject. The list states what was done and does not state why.
- 10-11 February 2021. Instagram permanently removed Robert F. Kennedy Jr.'s account. A Facebook spokesperson said: 'We removed this account for repeatedly sharing debunked claims about the coronavirus or vaccines.' Facebook left his Facebook page up. NPR (via WGBH), 11 February 2021, quoting the Facebook statement
- 2021; opinion 26 June 2024. White House officials pressed Facebook in 2021 to remove the accounts of the 'disinformation dozen', twelve people that include Kennedy. The Supreme Court recorded the request and held that the plaintiff before it had not shown a link between it and the moderation she experienced; it did not decide the First Amendment question. Murthy v. Missouri, 603 U.S. ___ (2024), slip op. (exh. Exhibit 067)
- 26 August 2024 (describing 2021). Meta's chief executive wrote to the House Judiciary Committee that in 2021 senior Biden administration officials, including the White House, 'repeatedly pressured our team for months to censor certain COVID-19 content, including humor and satire'. Rep. Nancy Mace letter of 11 September 2024 quoting the Zuckerberg letter (exh. Exhibit 867)
- 17-18 August 2022. Meta removed the Facebook and Instagram accounts of Children's Health Defense, the organization Kennedy chairs and that carries the book in its series. A Meta spokesman said: 'We removed these accounts for repeatedly violating our policies.' Associated Press, 18 August 2022, quoting Meta (as carried by NBC News)
- Filed August 2020; Ninth Circuit opinion 9 August 2024. Children's Health Defense sued Meta, Mark Zuckerberg, the Poynter Institute and Science Feedback over Facebook's warning labels, fact-checks and donation-button restrictions (filed August 2020). The Ninth Circuit affirmed dismissal 2-1, holding that the harm came from Meta's own policy and that CHD alleged no government coercion or agreement; Judge Collins dissented in part. Press reports (Mealey's, MediaPost) say the Supreme Court later denied certiorari. Children's Health Defense v. Meta Platforms, Inc., No. 21-16210 (9th Cir.) (read in full text)
- 14 February 2024. Kennedy and Children's Health Defense, as plaintiffs in Kennedy v. Biden, moved for a preliminary injunction against federal officials' contacts with platforms. The district judge held the motion pending Murthy, found Kennedy likely to succeed against the White House, Surgeon General, CDC, FBI and CISA defendants, denied it as to State Department, NIAID, FDA, Treasury, EAC and Commerce defendants, and stayed the order until the Supreme Court acted in Missouri v. Biden. Memorandum Ruling, Kennedy v. Biden, No. 3:23-cv-00381 (W.D. La.), Doc. 38
- Late 2021 to early 2022 (exact dates not given); Lyons essay 29 March 2022. The New York Times declined Skyhorse's advertisement for the book. Skyhorse's publisher Tony Lyons quotes the Times as first saying 'We cannot accept ads that contain ad hominem attacks', later 'we can't accept the ad due to misinformation', and then 'The Times retains the right to decline an advertisement offered to us'. Tony Lyons (Skyhorse), Substack repost of his Children's Health Defense essay; the Times statements are as quoted by Lyons
- Late 2021 to early 2022; Lyons essay 29 March 2022. Twitter refused to let Skyhorse advertise the book, and YouTube took down a televised book signing and discussion of it. The source is Lyons's own account; no platform statement was found. Tony Lyons (Skyhorse), Substack repost of his Children's Health Defense essay
- Around 2 December 2021. Amazon rejected a Skyhorse promotion for the book, with the message that the ad 'contains content that is not allowed for advertising' and asking Skyhorse to remove 'any content related to political campaigns, elections, or political issues of public debate'. Skyhorse is the source of the account. Washington Times, 2 December 2021, as excerpted by ADN America (the Times page returned 403)
- Request 3 September 2020; article 1 October 2021. The Intercept requested the EcoHealth grant records from NIH on 3 September 2020. NIH's records show the year-four progress report (due April 2018) was submitted on 16 September 2020, less than two weeks later. The first release omitted the year-five report, and The Intercept sued under FOIA for the missing reports. The Intercept, Hvistendahl and Lerner, 1 October 2021
- February 2021 emails; released and testified May 2024. NIAID senior adviser David Morens wrote in February 2021 that he had learned 'from our FOIA lady here how to make emails disappear after I'm FOIA'd but before the search starts', and that he had deleted earlier emails after forwarding them to Gmail. The House Select Subcommittee released the emails in May 2024; Morens testified on 22 May 2024 that he had not realized his government emails were federal records, and said Moore gave him no advice on avoiding FOIA. House Select Subcommittee on the Coronavirus Pandemic hearing 1 May 2024 and 22 May 2024; National Security Archive (Unredacted), 31 May 2024
- 30 September 2024 (subpoena); October 2024 (deposition). The House Select Subcommittee subpoenaed Margaret Moore, the former head of NIAID's FOIA office, for deposition on 4 October 2024. Her counsel said she would invoke the Fifth Amendment, and Fox News reported she did. House Oversight Committee release, 30 September 2024; Fox News
- Production dated 21 July 2025; reported 30 September 2025. NIH released 2016 internal emails of its ad hoc committee that reviewed EcoHealth's Wuhan experiments to U.S. Right to Know only as a partial production, filed in USRTK v. NIH, No. 24-cv-02785 (D.D.C.), a FOIA suit still open in 2025. U.S. Right to Know, 30 September 2025
- 15 May 2024; debarment notices 17 January 2025. HHS suspended EcoHealth Alliance on 15 May 2024 and debarred it for five years, and moved to debar Peter Daszak, on grounds that include the late year-five report and a lockout claim that NIH's forensic review contradicted. EcoHealth ended Daszak's employment effective 6 January 2025. HHS Notice of Debarment, redacted copy released by House Oversight (exh. Exhibit 116)
- 6 January 2022. FDA proposed to release the Pfizer-BioNTech vaccine licensing records at 500 pages a month, which the plaintiffs said meant about 55 years for more than 329,000 pages. Judge Pittman rejected it and ordered 12,000 pages by the end of January 2022 and 55,000 pages every 30 days, permitting redaction only where a 'privilege, exemption, or exclusion' applies. He wrote the request 'is of paramount public importance'. Public Health and Medical Professionals for Transparency v. FDA, No. 4:21-cv-1058-P (N.D. Tex.); order as reported by BioSpace
- Suit reported 31 December 2021; first release 3 October 2022. CDC declined a FOIA request from the Informed Consent Action Network for v-safe data, while giving deidentified v-safe data to its contractor Oracle. ICAN sued; under a court order CDC released the check-box fields of over 10 million users and withheld the free-text responses. Fox News via Siri & Glimstad, 31 December 2021; ICAN/Siri & Glimstad press release, PR Newswire, 3 October 2022 (plaintiff's own account; no CDC statement in it)
- 5 January 2024. CDC withheld about 7.8 million v-safe free-text responses (October 2020 to December 2022), citing personal-information review that a CDC declaration put at 59 analyst-years for its 13-analyst FOIA office. Judge Kacsmaryk found the burden claim unpersuasive, granted summary judgment to the requester and ordered production with exempt personal information redacted. Freedom Coalition of Doctors for Choice v. CDC, No. 23-102 (N.D. Tex.), DOJ OIP summary
Published responses
- Theodore Dalrymple, Claremont Review of Books (review, Summer 2022): Reports checking five of the medical papers the book cites and finding each misrepresented. Details he gives: an Argentinian trial the book calls randomized, which in his account states in its own text that it was unrandomized; the Ferguson projection, which he says the book wrongly calls off by more than an order of magnitude; a UK ONS report that he says contains numbers and no rates; an ivermectin trial of 12 patients per arm; and that Kennedy devotes pages to denying that HIV causes AIDS and then says he believes it does (this matches Ch. 5).
- Anthony Fauci (interview with Yahoo News, reported by TheWrap, 22 December 2021): Called Kennedy 'a very disturbed individual' and said it is a shame that he is attacking Fauci's career; the report also gives Fauci saying he does not think Kennedy is inherently malicious. The article mentions no legal action.
- Anthony Fauci (as reported by The Hill, 2 July 2024): Listed on the book's Wikipedia page as saying 'I don't know what's going on in his head, but it's not good'.
- Tony Lyons, Skyhorse Publishing (Publisher's Note, print edition; and The Guardian, 27 January 2022): The Publisher's Note invites readers to 'Tell us where you believe something is incorrect, make your best arguments, and offer the best available support', says every suggestion 'will be heard and considered', and promises 'updates, additional notes, and new references' through a QR code. In The Guardian interview Lyons said the book had sold 750,000 copies in eight weeks and has a number one rating on Amazon.
- Julie Matthaei, Review of Radical Political Economics (SAGE), book review: Listed in the candidate record from a search summary as noting about 2,200 endnotes drawn from primary sources, journal articles and interviews, and reviewing the book favorably.
Where it connects
- What Was Known and When: NIAID's EcoHealth grant: the late year-four and year-five reports, the Wuhan notebooks never obtained, the Morens and NIAID FOIA-office emails, and the 2025 debarment.
- The Template Goes Global: Gates Foundation and Gavi funding of WHO, and the book's Chapters 9 and 10 on the Foundation's holdings, which the audit checks against the 990-PF and WHO tables.
- The Studies Not Run: The hydroxychloroquine and ivermectin trial record, Pfizer's unblinded six-month data and the FDA and CDC data (licensing files, v-safe free text) released only under court order.
- Managing the Narrative: Platform moderation of Kennedy and Children's Health Defense after White House and CDC contacts, as documented in Murthy v. Missouri, Kennedy v. Biden and Zuckerberg's 2024 letter.