I found this video by Ivor Cummins quite informative. Of course, I’m open to being told why he might be, or even is, wrong.
The data on Sweden and “dry tinder” are particularly interesting. Economists Dan Klein, Joakim Book, and Christian Bjornskov have written about this and he quotes it.
One thing I think Cummins brushed by unconvincingly is the difference between the USA Midwest and the USA Northeast, at about the 23:40 point. He says the shape is similar. He and I have a different view of the word “similar.”
One of the most upsetting parts is about how the lockdowns during the summer lengthened the time to herd immunity and therefore might themselves create an increase in the deaths in the fall and winter.
READER COMMENTS
robc
Sep 10 2020 at 3:42pm
The NE and MW are both single humped, while the South and West are double humped. In that way they are “similar”.
I agree with you, but I get what he is saying.
David Henderson
Sep 10 2020 at 5:40pm
Yes, that’s probably what he’s saying. I thought, though, that the sharp upside-down V vs. the upside down U deserved a few sentences. But thanks for noting that.
Alan Goldhammer
Sep 10 2020 at 8:09pm
What evidence do you have for this statement? Do you even know what the % infected needs to be to obtain herd immunity? I’ve probably read almost all the technical papers on this topic and see ranges from 25-70%. Remember also that herd immunity is a stochastic process capable of disruption if the boundaries of the herd are broken. If one is a proponent of herd immunity the best place to move to these days is probably Queens or parts of Brooklyn that have high levels of sero-positive individuals. We’ll see how this plays out going forward.
For someone who is claiming to present the “best” science, this clip is woeful in that regard. Of course if you want to believe him, feel free to go out to bars and restaurants and don’t worry.
Mark Z
Sep 10 2020 at 10:06pm
“Herd immunity” is probably the wrong way to think about this. The rate at which the virus spreads is a continuous function of the susceptible population is. Even if herd immunity is 70%, it’s still the case that the virus will spread more slowly in a population that is 30% exposed than in one where the exposed fraction is only 5%. Unless the epidemic can be more or less suppressed, which seems unlikely, preventing spread today increases susceptibility tomorrow. I don’t know what’s behind David’s thinking, but if one thinks it’ll be worse in the winter (because of effects of humidity changes on the respiratory system and such, or cooccurrence with flu season), then reducing infection rates today at the expense of increasing them in the winter may be a difficult tradeoff.
If a vaccine may be available this winter, then that of course shifts the calculus away from worrying about susceptibility in the winter.
David Henderson
Sep 10 2020 at 10:19pm
You wrote:
Actually, Cummins makes that point. He basically says that NYC is at or close to herd immunity.
Michael Byrnes
Sep 13 2020 at 7:35am
Herd immunity is a terrible and misleading term.
The 70% figure comes from simple modeling of a pathogen with an R0 in the 2.5 to 3.0 range.
The lower estimates come from models that assume that there is variation in individual susceptibility to the virus. If models are run that assume that some people are more likely to become infected than others, herd immunity thresholds are much lower, possibly as low as the 25% that has been suggested. Part of the rationale for this is that over time, the people who get infected are those who are more susceptible, and most of them will ultimately survive and have some immunity afterward, meaning that the overall susceptibility of the whole population falls as the “more susceptible” groups get infected, recover, and then are less susceptible.
The problem with this, though, is that a large part of what affects susceptibility is behavioral. As a group, people who observe social distancing, avoid large/indoor gatherings, wear masks, engage in outdoor activities preferentially to indoor whenever possible, etc., have are less likely to become infected than those people who don’t take these measures. That results in differential susecpitbility between groups and a lower “herd immunity” threshold as the virus runs through those who are less risk observant.
But the diffential suspecptibility that leads to a lower herd immunity threshold will only hold up as long as the less susceptible groups maintain their risk minimizing behaviors.
In much of the country, we are headed towards 2 things that may increase susceptibility among the (behaviorally) least susceptible: the reopening of schools (putting students and staff together for long amounts of time in poorly ventilated indoor spaces) and the transition to cooler weather that makes outdoor actvities less feasible.
As a result, there are places where perhaps we are at herd immunity thresholds now but will cease to be once schools reopen and the weather cools.
Richard A.
Sep 10 2020 at 10:04pm
There is a strong correlation between vitamin d deficiency and dying from the coronavirus. Dr. John Campbell has a good video on the latest vitamin d study.
Ken P
Sep 10 2020 at 11:42pm
I agree. And adding to this, the excessive flattening of the curve by lockdowns extends time that the most susceptible must avoid exposure and increases the potential exposure events. The total number of trips to the grocery store for example is much larger for 10 months than for two months.
Loz
Sep 13 2020 at 9:26am
Absolutely correct – the virus doesn’t survive well in warm, humid weather, doesn’t transmit easily outdoors, and people are generally less susceptible to respiratory illness in the summer. So we should be ‘making hay while the sun shines’ – letting healthy people get exposed to relatively low viral doses during the summer months, in order to reduce the risk of outbreaks during the winter period when hospitals are busy with the normal winter respiratory viruses (influenza and many others).
We’ve lost the main part of summer, but still have relatively warm weather (compared to February) for another month or more, so now is definitely not the time to be imposing more and more restrictions on people.
Regarding herd immunity, there is no other option, whether the immunity comes from vaccination, infection or mild exposure (just enough to prime the immune system, without causing a real infection). We will get there eventually – we just have to decide if we want it to take 3 months or a year. As long as the health system isn’t overwhelmed, getting there quickly is less risky and less expensive.
In New Zealand, they are betting on a highly effective vaccine – otherwise, are they going to remain closed off for five years? In the UK, with the virus already having spread widely, natural immunity may be a better option.
Vivian Darkbloom
Sep 11 2020 at 3:34am
I found the case he presented convincing. In particular, I wish someone (Alan Goldhammer?) would address the issue (rebut Cummings) as to why in some areas we have recently had a significant increase in the number of “cases” but a much lower and more or less steady number of hospitalizations and deaths. Cummin’s explanation was based on a lot of relevant facts and made sense to me (i.e., it seems based on “science”).
I few months ago I asked these questions at Marginal Revolution: Is it possible that the positive Covid tests reflect not only the persons who are currently ill and/or susceptible of transmitting the disease (i.e., the virus is “active”, but also those who *have had* at some point in the past been exposed to the virus? How long does evidence of exposure stay in the system and therefore is picked up by the tests being used? I got no answer to those queries, but Cummins seems to suggest that the PCR test is “positive” some time after a person has been ill or is a danger to others. Not only do recent results seem to be driven by the extent of testing, but by the nature of those tests as well. My subsequent research suggests, as does Cummins, that a positive PCR test does not mean that a person is infectious: https://www.cebm.net/covid-19/infectious-positive-pcr-test-result-covid-19/
I live in France in a department currently listed as a “red zone” solely because of the increase in the number of detected “cases” (not hospitalizations or deaths)—this “increase” despite very strict rules on mask-wearing, etc. My neighbour will be subject to a 14 day quarantine when he returns to Belgium tomorrow because he will have been in this zone (while classified “red”) for one day. Belgians (and residents of other European countries) are now prohibited by their governments from travelling to this area. This is the “European Union”?! It strikes me that the bar is continuously being lowered for initiating these top-down restrictions on liberties.
Glenn Ammons
Sep 11 2020 at 10:47pm
Great video, but his explanation why the US has had a relatively high death rate this summer is unconvincing. If I understand correctly, he claims that the explanation is normal seasonality: many regions in the US follow a “Northern Tropical” pattern for respiratory diseases, including influenza and COVID-19, where the deaths are more spread out and last through the summer.
But I looked at the CDC’s FluView Interactive, and that doesn’t seem to be the case. Normally. places like Florida and Texas have flu outbreaks at the same time as places like New York.
Also, I don’t think his theory explains why, for example, Iowa had an increase in deaths this summer. Granted, Iowa’s death rate hasn’t been huge, but it did see a bump and it’s definitely way more Northern than Tropical!
Peter Brown
Sep 14 2020 at 4:47am
There are some strange assumptions and lacunae in this video. First, the suggestion that 80 per cent of us will already have some immunity because there are a lot of coronaviruses out there. Really? The common cold gives us immunity? There is no medical evidence for this theory. Second, he dismisses lockdowns. Well, I sympathise with an Irishman who finds himself barred from the pub. But he makes no mention of China or South Korea, where lockdowns were indisputably effective. A bit selective with the evidence, there, Ivor?
robc
Sep 14 2020 at 7:39am
I think the relative prevalence of SARS-1 in Asia may give some credence to the immunity point of his. I thought one of the weaknesses was the lack of any Asian data in the post, but it was very Euro-centric, so I get that.
I think for your latter point, you need to compare Asian countries with lockdown with Asian countries without lockdown. I don’t think Asia to Europe comparisons make much sense.
Robg
Sep 15 2020 at 2:11am
To say he has been selective with the data is to be generous. Never mind the bogus seasonal patterns. Off the top of my head, Sweden and Spain might be considered as climatically different as New York and Florida or Utah, and since when did it start getting cooler in August.
He then completely neglects Asia – which might actually give enough contrast for meaningful conclusions.
The idea of Asian immunity doesn’t quite make sense given the ‘virulence’ of the Wuhan outbreak. Even here in Korea our first outbreak via a Wuhan connection, was severe, and the city was under lockdown. I could be proven wrong but I’d need hard evidence, of which I am happy to let the Epidemiologist’s supply.
It is also notable that most people wear masks, we also have some social distancing, rapid testing (6hrs), tracing and isolation of new cases, and localized lockdowns of varying degrees as needed.
I can’t tell you which of these is the most important but when people break these rules we get outbreaks, which are rapidly brought under control by the response. It has been like this from the get go.
Raising questions is about expert conclusions is fine – but if this guy wants to draw and promote conclusions (and make a living at it via Patreon!) he should be held to the same standard of proof…on which he’s a long way short. Instead he just adds to the fog of disinformation…
robc
Sep 15 2020 at 9:27am
I agree that his seasonality stuff is by far the weakest part. It doesn’t really work when you start looking at US states individually.
I do think the “dry tinder” aspect is strong though. I would have liked to see those graphs for Asia and for individual US states.
Lisa
Sep 18 2020 at 10:36am
Does anyone dispute Ivor’s GOMPERTZ death rate curves?
Despite a “fog of misinformation”, it seems that the rate curves are the definitive measure of what’s happening.
JH
Sep 18 2020 at 5:55pm
Look at the Johns Hopkins site and the page showing deaths over time by state completely explains the differences between the NE US and Florida, Texas, California etc. It is unbelievably descriptive for each state.
https://coronavirus.jhu.edu/data/state-timeline/new-deaths/alabama/0
A.m.k.
Sep 15 2020 at 8:46am
I watched about half of the video. Then I looked him up. He’s the self-named Fat Emperor. He is an engineer by trade and according to his bio, started researching health when his own was suffering. Now he tells people how to diet. I fully support the idea of getting answers for your own health but until I hear from a majority of epidemiologists and actual medically trained specialists who are on the front lines, I’ll keep wearing a mask. It’s difficult for me to conceive that the entirety of the planet has had the wool pulled over their eyes.
Robert McConnell
Sep 25 2020 at 10:54am
Yes, The Fat Emperor seems like a weird name but it’s based on the now established principle, that (saturated) fat is one of the most metabolically health food for most humans.
Cummins himself calls the US data confusing.
I have followed this engineer Ivor Cummins on YOUTUBE for 1 1/2 years. He is revered by many health care low carb professionals. He has given great talks/interviews on improving metabolic health and reversing chronic diseases. He and another engineer, Dave Feldman, are highly respected by many leading healthcare professionals in getting society off drugs for “chronic” diseases that science shows can be addressed through life style and diet interventions.
He and the cardiologist, Jeffry Gerber wrote an amazing book demonstrating this science to improve health–Eat Rich Live Long.
He has interviewed a number of virologists, statisticians and immunologists. Here is a video with a Swiss immunologist. https://youtu.be/GBRcK-od50Q
Caralhos Mefodam
Sep 15 2020 at 9:44am
Just quoting the latest update (15SEP2020) but unsure how these would compare to the average hospitalization healthcare numbers for this time of the year.
These are being attributed to Covid, as there is a delay in increased positive cases and hospitalization, eventual death in some cases.
“U.K. COVID daily hospitalisation rates have TRIPLED in the last fortnight and the number of patients in hospitals has DOUBLED. https://coronavirus.data.gov.uk/healthcare
With exponential increases the time for aggressive lockdown is NOW. Wash hands, make space, cover your face. #COVIDuk@ @
https://twitter.com/juniordrblog/status/1305814815072452608?s=20
RobUK
Sep 25 2020 at 3:59pm
In 1918 Africa took the hardest hit from that Pandemic therefor the seasonal effect for the low covid deaths does not appear feasible and questions are at present being asked as to why Africa has so few Covid deaths ( Uganda 71 covid related deaths out of 47 million, Nigera 1,102 out of 207 million. In 1918 there was NO treatment for malaria in Africa, today there is and the cheapest and still most widely used medication is Chloroquine. In western country`
s this drug has been largely banned, but country
s that have sanctioned it`s use (Turkey and Greece ) to name two have seen low covid related deaths. There have also been few covid deaths in two large groups that use Hydroxychloroquine a derivative of this drug daily, Systemic Lupus and Rheumatoid Arthritis. The link is obvious here, countrys that use CQ/HCQ have few covid deaths and country`
s that don`t have high Covid deaths.Thomas Baerg
Sep 16 2020 at 11:34am
Just a courtesy, Mr. Cummin’s rebuttal.
https://www.youtube.com/watch?v=eKKIr425b40&ab_channel=IvorCummins
James EG
Sep 17 2020 at 11:46am
Some of the analysis by I.C. seems faulty.
Regarding the claim that the interventions (lockdowns, masks, business closings) were not effective.
Looking at the distributions of outbreaks and showing no change in them compared to other seasonal respiratory infections due to an interventions in just some countries is cherry picking data. There are countries that were very effective in preventing the spread of coronavirus. So it may be that the interventions were ineffective because of compliance, timing, or the scope of interventions. Basically poor implementation could be why it was ineffective. I think you need to look at countries that were effective in preventing the spread for an idea of which work and which do not.
Looking at distribution and not magnitude is also faulty. It shouldn’t be a shock that a respiratory illness follows the pattern of similar respiratory illnesses. The interventions aren’t going to change those distributions, they will impact the magnitude of them though.
Comparing COVID only to the magnitude of 2018 as a judge of ‘effectiveness’ is also disingenuous. 2018 is one of the worst years on record. With interventions 2020 is still 30% worse currently. The 2020 epidemic isn’t over yet, whereas 2018 is. Predicting how much worse 2020 would be with no interventions is probably impossible. But minimizing that ‘only 30% difference” is misleading. Cost/Benefit analysis is opinion, not science, as it depends on how much you value the lives of others. There’s a lot of bias in this opinion of course.
Regarding inaccuracy of predictions
There were several predictions made on what the spread can be, each with their own margins for error.
Picking one of the most dire predictions made early in the pandemic and bashing it months later is post-game quarterbacking.
Predictions are made off limited knowledge, limited analysis time and for specific purposes. They aren’t bad science necessarily.
Regarding the claim the pandemic didn’t hit ‘at-risk’ occupations.
No data was given to support this claim.
Health care workers were badly hit in many countries. Probably was to be expected.
Meat packing plants in several countries were badly hit. This was not widely expected. Due to working conditions.
Migrant workers in the USA were badly hit. Mainly due to bad living conditions.
Many poor communities were badly hit. This was due to living conditions and the types of jobs they hold.
Maybe if you cherry pick occupations maybe you can find some that weren’t hit, so I’d like to see data to support this.
Regarding Sweden
Swedes did change their behavior during the pandemic. It may not have been a lockdown, but it’s not apples-to-apples with every other year.
Again, the distributions in Scandinavian countries are going to be similar. But the magnitude with interventions is mitigated.
The immunological ‘dry-tinder’ argument is interesting as an explanation for regression to the mean. I’m not familiar with the data to judge the validity of the idea. Of course the entire point of interventions is to protect high-risk individuals. Calling it ‘dry-tinder’ may just be a dehumanizing term to avoid feeling responsible for doing nothing.
David Millar
Sep 26 2020 at 10:51am
If it appears that Covid 19 is going to be with us for sometime and all we can do is mitigate until a vaccine is found then protecting the vulnerable and not ruining the economy makes sense to me up to a point. Half of the deaths have been in care homes so they are surely (easily?) protected as they are locked up anyway. It would mean being careful with staff/visitors so they are not isolated. The rest will have to proceed with caution and minimize contact whenever they can. (I know this is difficult and we are an integrated society but wrecking an economy will wreck lives.) I have issues with long Covid and the desperate descriptions from people who have suffered this. The death rate is so much lower now from the disease and I feel that either a different strain is now prevalent or many of those susceptible have since passed away?
James Christian
Sep 28 2020 at 2:19pm
Africa is certainly worth looking closely at. Discrepancies this large are worth examination to better understand this disease. Obesity and diet related influences certainly must be making a larger impact in parts of the west and I would imagine that there are also correlations between the prevalence of new allergies such as nuts and gluten and compromised immune systems there. Africa On the other hand does not appear to have any of these immune related problems. Furthermore, the youth of its populations must also help it’s response.
Gareth Valentine
Sep 29 2020 at 7:10am
I think you’ll find Cummins is a bit of a mountebank – his arguments effortlessly unravelled here: https://youtu.be/DUDg5ossirU
Carol James
Oct 2 2020 at 11:48am
I found the Igor Cummins video very persuasive until I reached the part where he claims that lockdowns didn’t reduce the number of influenza cases. Of course the 2020 winter flu deaths are the same as usual in the U.S. because lockdowns occurred after peak flu season. Flu season peaks between December and February. Lockdowns didn’t start in any states until March. Had lockdowns and masks started in December I expect flu deaths would have been considerably less.
After noticing this, I fact checked his evidence about a study that claims masks did not significantly reduce influenza spread in lab tests. According to Politifact: “No, a CDC-WHO study does not prove that masks do not prevent spread of COVID-19”. The study was published by the CDC and supported by WHO, but it was done by the University of Hong Kong and one of the study’s authors told Politifact the claim is incorrect.
I also found a New York Times article claiming only 20% to 50% may have cross imunity, not 80%.
I also question his assumption that American COVID-19 death counts are accurate. I found a graph by Sam Ashoo, MD, FACEP indicating a huge spike of pneumonia deaths in winter 2020. Reuters published this: “The overall number of U.S. deaths for any cause tallied by the National Center for Health Statistics during March, April and May was 781,000, or 122,300 more than the historical average for the period, according to the study funded by the National Institutes of Health and private foundations published in JAMA Internal Medicine.” CNN said “US coronavirus death count likely an underestimate … Also, deaths due to Covid-19 my be misclassified as pneumonia deaths …” As well, a July 14, 2020 Washington Post article describes a new reporting protocol for hospitals and said “Public health experts say bypassing the CDC could harm the quality of data”.
Finally, when I thought more about his impressive dry tinder explanation, I looked at graphs of cases versus deaths for influenza and it looks to me like there are simply more deaths when there are more cases. There are many reasons besides dry tinder for more cases, e.g. a less effective flu vaccine that year or less people get the flu vaccine or we have a longer winter.
I’m not sure why people listen to Igor at all. When I’m sick I see a doctor of medicine. When the climate is sick I get information from climate doctors (climatologists). And during a pandemic I get information from infectious disease doctors – NOT a chemical engineer. We should follow the science — the infectious disease science!
Alexandra X
Oct 9 2020 at 8:21am
Very helpful statistics for me. Ivor Cummins takes into account a wider perspective which gives sense and explains a lot. Since it is already known that covid 19 like all other respiratory diseases are more dangerous for people living in more polluted areas (or for people with polluted lungs by e.g. smoking) as well as for people with obesity, this statistic could be extended by analysis of these factors. Maybe it could fill the gaps, why some countries or parts of countries had worse than “expected” statistics.
Comments are closed.