The source for this is an economics paper using old-school macro techniques to measure the correlation between life expectancy and the unemployment rate.
Note that the policy conclusion of the paper includes "It is worth clarifying that with this claim, we do not want to suggest that policymakers should refrain from ordering lockdowns, as necessary lifesaving measures, but rather that, if they decide to do so, they should provide alongside enhanced health and economic support for the most vulnerable portions of the population."
Moreover, note that the sentence "Unemployment shock is expected to cause 890,000 additional deaths over the next 15 years." like the whole text you cite seems to suggest a counterfactual of people just continuing their normal work and normal life. While it is surely debatable how effective lockdowns were, their justification was not only saving the lives of old people, but also avoiding an uncontrolled breakdown of the economy; and without a lockdown, many people would probably also have lost their job or stayed home. In some cases, a lockdown ordered by the government may allow more targeted help to those affected.
Some people in my orbit suggested reading Robert F. Kennedy Jr’s book The Real Anthony Fauci.
Here’s my story of wading through a few pages of the book and trying to understand the basis of the claims. My takeaway: there's a lot of sloppiness here, and several times the book’s claim isn’t supported by the citation.
Often, though, the citation is fine. The arguments are carried by shoddy science and massive leaps in causal attribution. For a lot of the anti-vaccine and anti-lockdown takes, there’s at least some study by some scientists showing really bad effects. And the scientists are real—it’s just a small sample or poor design.
Sorting things out honestly is hard, especially with bad faith actors using sloppy language to describe highly selected but serious-sounding findings. My notes on the text follow.
Lockdowns
First, the Introduction has a section on the harms from lockdowns (p.11). I gravitated toward it because it had a lot of numbers. I’m open to the claims that lockdowns were net negative, with severe and diverse harms.
Here’s the first paragraph:
We’ll go sentence-by-sentence.
I couldn’t find the source for this. When I search “300 million” and lockdowns, I see it referenced when saying that 300 million people were under lockdown in the United States and (separately) China. So maybe he just means people under lockdown?
The AP article and the second sentence above is based on this Lancet comment that derives the 10,000 number using predictions from a different Lancet comment (Headey et al). The predictions from Headey et al assume an 8% drop in GNI in LMICs and also that governments do nothing to prevent these issues. So it’s certainly wrong to say “died”--10,000 per month was a projection under an unrealistic counterfactual.
Also, in practice, according to Our World in Data (OWID), there doesn’t seem to have been an increase in malnutrition and I could not find any studies saying that the increase in child mortality actually happened. However, I’m not sure that we’re able to measure malnutrition well in 2020-2021, so wide confidence intervals on that one.
The source for this is a BBC article which cites this UNICEF report. It's a shaky modeling exercise predicting a 15% increase in neonatal and an 18% increase in maternal deaths in India, for example. The same models predict that 400,000+ people would die without lockdowns. So taken literally, the cited report is saying that lockdowns saved 200,000 lives.
Further, here’s OWID child mortality and maternal mortality in South Asia over that period. As with malnutrition, I can't see any increases (although COVID could have made measurement harder).
The cardiovascular disease concern is in this Time article (cited in the Tablet piece), extrapolating from increased alcohol and decreased exercise and citing a few medical articles. Doctors were also saying this about cancer. For example: “The COVID-19 pandemic will result in nearly 10,000 additional deaths due to breast cancer and colorectal cancer over the next decade, according to NCI projections” and see this paper for a review.
I couldn’t find anything quantitative on kidney failure and diabetes, although surely disruptions harmed these kinds of patients to some degree. Here’s an anecdotal report from India.
I don't think an increase has happened although it was predicted by many medical experts. Here's the age-adjusted death rate for neoplasms from CDC data. There are more careful ways to check, but this seems to rule out large impacts of deferred care.
The source for this is an economics paper using old-school macro techniques to measure the correlation between life expectancy and the unemployment rate. Hard to disprove and perhaps reasonable as a first pass, but I would bet that these numbers would be disproven by any study using person-level data to measure the impact of COVID-related layoffs on mortality. For reference, the numbers imply around 60,000 deaths per year from unemployment shock over the next 15 years. Diabetes kills 100,000 per year.
Here’s the second paragraph:
Sentence by sentence again:
No source for this although some of the claims are covered below. But all seems plausible?
The book only cites an article about a study of 571 Appalachian Ohio teens here? But seems plausible and larger studies say the same thing.
This is…a huge effect. The text is wrong, the reference cites the CDC, not Ohio State. And the CDC study does show a 50% increase in ER visits for suicide attempts among girls age 12-17 in February to March 2021. They’re talking about the red box here, which is just a portion of the lockdown period.
Concerning, but pretty different from what’s stated in the book. I looked for a study on the impact of lockdowns on child suicides. The best one estimates an increase of 4%.
The 22-point IQ difference should sound a little extreme on its face. This is a 1.5 ( = 22/15) standard deviation drop, which would move a median infant down to the 7th percentile.
Here’s the paper, on 700 infants in Rhode Island. I couldn’t find the version using Baylor (or more accurately Bayley) assessments. They do report a drop with a smaller effect size. Other papers find the same thing. I doubt that lockdown babies will look that different a few years out, but I’m not sure.
Yep.
No cites. Schools still seem messed up.
The DTP Vaccine
Chapter 10 (“More harm than good”) goes through several vaccines that RFK Jr thinks are killing people on net. I worked through the first section called “DTP Vaccine: African Genocide” (p.137). Diving right into the sentences. I skip stuff about WHO decisions and Bill Gates since it’s harder for me to sort through the evidence there.
No source given. I don’t think it’s true, as you'll see below.
He’s referencing this article by a sole researcher, Gordon Stewart, who was the UK’s RFK Jr. The effectiveness of the pertussis jab seems pretty uncontroversial so I'm holding off on a deep dive into pertussis for now. Here’s a figure.
The case-control study referenced is Baraff et al, which points out that SIDS tends to happen around DTP vaccinations. It’s based on 145 SIDS deaths in LA County (out of a total of 382 that they tried to contact).
The core evidence is in the histogram below, which shows days since DTP appointments in the blue bars and days since doctor appointments without vaccinations in the green bars (I made this from two tables in the paper).
They think an eerie number of SIDS deaths happened right after the DTP shots. It’s true that there are more SIDS deaths within 24 hours of the DTP shot. And the two distributions are significantly different. But the evidence here seems marginal and probably shouldn't convince you of anything.
Also, I read the paper and could not find the 1 out of 300 number. All the reported risks are lower.
It wasn’t randomized, it was just arguably random whether babies got the vaccine at 3 or 5 months.
The table below has the core result from the paper. They did find that vaccinated infants had higher mortality. Five out of 651 infants died in the unvaccinated group compared to 11 out of 462 in the vaccinated group. So, there were 7 excess deaths in the vaccine group. Still, the p-value for the differences in proportions is 0.026—a significant difference. The 10x estimate for girls is based on a total of just 4 deaths among girls and insignificant at conventional levels.
Overall this seems like results worth reporting. But the 10x estimate for girls is obviously bonkers.
Further, after the publication above, co-author Peter Aaby (a hero of RFK Jr’s story) published a 2x larger study on the effect of DTP. Their findings are the opposite. The DTP group fared better. Main results below.
16 of 1,795 infants in the DTP group died compared to 13 of 603 of the no-DTP group, so the control group had 2x higher mortality. (These numbers come from summing the Ns in the column headings and All Deaths in the bottom row.)
If DTP really caused a 5x increase in death, about 180 DTP infants should have died! Of course, a 10x increase in mortality would have been even more crazy. (This study uses control variables, but a 10x difference should be obvious without. )
The cited review has this section on the DTP vaccine. None of the studies are randomized and the effect is indistinguishable from zero if you remove the three studies from the Guinea-Bissau team (which includes the 2017 study above).
Gates and his WHO vassals continue to bully African nations into taking their lethal DTP vaccines by threatening to withdraw financial aid to their health departments and HIV programs if the government fails to achieve national uptake targets (90 percent).
Can’t find evidence of this.
Notes: Updated 11/23/2024 to add CDC cancer deaths and a review paper on expected cancer deaths due to deferred care.