Health Nerd
@gidmk.bsky.social
Epidemiologist. Research Fellow. Doctor of Spreadsheets. Writer (Slate, TIME, Guardian, etc). PhD, MPH. Host of senscipod Email gidmk.healthnerd@gmail.com he/him. Find my writing on Substack and Medium.
There are also lots of other errors in the paper, but honestly who cares that much.
And this isn't some nothing of a study. Here it is being cited in the Endocrine Society guidelines which currently recommend vitamin D supplements for children to reduce the risk of respiratory disease.
And this isn't some nothing of a study. Here it is being cited in the Endocrine Society guidelines which currently recommend vitamin D supplements for children to reduce the risk of respiratory disease.
November 26, 2025 at 9:15 PM
There are also lots of other errors in the paper, but honestly who cares that much.
And this isn't some nothing of a study. Here it is being cited in the Endocrine Society guidelines which currently recommend vitamin D supplements for children to reduce the risk of respiratory disease.
And this isn't some nothing of a study. Here it is being cited in the Endocrine Society guidelines which currently recommend vitamin D supplements for children to reduce the risk of respiratory disease.
Problem is, the researchers didn't pre-register influenza A as their outcome. They registered doctor-diagnosed influenza, and didn't specify the type.
Is that important? Well...
Is that important? Well...
November 26, 2025 at 9:15 PM
Problem is, the researchers didn't pre-register influenza A as their outcome. They registered doctor-diagnosed influenza, and didn't specify the type.
Is that important? Well...
Is that important? Well...
The stuff you find when you actually read the RCTs in a systematic review...
This paper is one of the foundational studies on vitamin D to prevent respiratory infections in kids. Cited 1,400 times as per Google Scholar.
This paper is one of the foundational studies on vitamin D to prevent respiratory infections in kids. Cited 1,400 times as per Google Scholar.
November 26, 2025 at 9:15 PM
The stuff you find when you actually read the RCTs in a systematic review...
This paper is one of the foundational studies on vitamin D to prevent respiratory infections in kids. Cited 1,400 times as per Google Scholar.
This paper is one of the foundational studies on vitamin D to prevent respiratory infections in kids. Cited 1,400 times as per Google Scholar.
Scientific error-checking, difficulty: easy
Find some issues in this table of results from an RCT of vitamin D for COVID-19. There's a couple of obvious issues, and some that are less easy to spot.
Find some issues in this table of results from an RCT of vitamin D for COVID-19. There's a couple of obvious issues, and some that are less easy to spot.
November 25, 2025 at 4:59 AM
Scientific error-checking, difficulty: easy
Find some issues in this table of results from an RCT of vitamin D for COVID-19. There's a couple of obvious issues, and some that are less easy to spot.
Find some issues in this table of results from an RCT of vitamin D for COVID-19. There's a couple of obvious issues, and some that are less easy to spot.
Scientific error-checking, difficulty: easy
Find the problems in this table of results from an RCT. Two main things I've noticed so far.
Find the problems in this table of results from an RCT. Two main things I've noticed so far.
November 25, 2025 at 2:32 AM
Scientific error-checking, difficulty: easy
Find the problems in this table of results from an RCT. Two main things I've noticed so far.
Find the problems in this table of results from an RCT. Two main things I've noticed so far.
Very interesting data - desk rejections have risen hugely at Frontiers journals due largely to paper mill detections and other fraud checks.
Do any other publishers do this automatically?
Do any other publishers do this automatically?
November 18, 2025 at 4:20 AM
Very interesting data - desk rejections have risen hugely at Frontiers journals due largely to paper mill detections and other fraud checks.
Do any other publishers do this automatically?
Do any other publishers do this automatically?
One thing you can say about Sweden is that they kept the messaging consistent, and public trust remained quite high throughout the pandemic.
The same is true of Japan.
It's NOT true of the United States, where everything became very political.
The same is true of Japan.
It's NOT true of the United States, where everything became very political.
November 10, 2025 at 4:40 AM
One thing you can say about Sweden is that they kept the messaging consistent, and public trust remained quite high throughout the pandemic.
The same is true of Japan.
It's NOT true of the United States, where everything became very political.
The same is true of Japan.
It's NOT true of the United States, where everything became very political.
This is one reason why I think Sweden did reasonably well. If you look at deaths and economics, they were worse off than their peers in Scandinavia in 2022, but not drastically so.
Depending on where you end the "pandemic", you could say they did worse than or pretty similar to Norway.
Depending on where you end the "pandemic", you could say they did worse than or pretty similar to Norway.
November 10, 2025 at 4:40 AM
This is one reason why I think Sweden did reasonably well. If you look at deaths and economics, they were worse off than their peers in Scandinavia in 2022, but not drastically so.
Depending on where you end the "pandemic", you could say they did worse than or pretty similar to Norway.
Depending on where you end the "pandemic", you could say they did worse than or pretty similar to Norway.
We have very up to date information on excess deaths here in NSW. Long story short, we have not had any meaningful excess since Omicron in 2022. A modest excess during winter last year, and essentially no excess this year.
November 10, 2025 at 4:19 AM
We have very up to date information on excess deaths here in NSW. Long story short, we have not had any meaningful excess since Omicron in 2022. A modest excess during winter last year, and essentially no excess this year.
Someone asked me about ivermectin for COVID-19 recently and it's such a bizarre blast from the past.
Can't believe it's been more than 4 years since we proved much of the benefit was based on fraudulent research! Wild.
Can't believe it's been more than 4 years since we proved much of the benefit was based on fraudulent research! Wild.
November 9, 2025 at 10:57 PM
Someone asked me about ivermectin for COVID-19 recently and it's such a bizarre blast from the past.
Can't believe it's been more than 4 years since we proved much of the benefit was based on fraudulent research! Wild.
Can't believe it's been more than 4 years since we proved much of the benefit was based on fraudulent research! Wild.
We're running a trial of exercise in nature here in Sydney. Free support for exercise and wellbeing for those who participate. If you know anyone who might be interested, please share our link: www.powerlab.site/research/pan...
November 9, 2025 at 10:53 PM
We're running a trial of exercise in nature here in Sydney. Free support for exercise and wellbeing for those who participate. If you know anyone who might be interested, please share our link: www.powerlab.site/research/pan...
A very interesting comparison here.
2025 vs 2023 for my state of NSW. Testing for COVID-19 and influenza has not changed substantially that I'm aware of in this time.
The labs generally test for both on any swabs taken. The ratio of COVID-19:Flu
2025 - 0.49
2023 - 2.97 (!)
2025 vs 2023 for my state of NSW. Testing for COVID-19 and influenza has not changed substantially that I'm aware of in this time.
The labs generally test for both on any swabs taken. The ratio of COVID-19:Flu
2025 - 0.49
2023 - 2.97 (!)
November 6, 2025 at 10:58 PM
A very interesting comparison here.
2025 vs 2023 for my state of NSW. Testing for COVID-19 and influenza has not changed substantially that I'm aware of in this time.
The labs generally test for both on any swabs taken. The ratio of COVID-19:Flu
2025 - 0.49
2023 - 2.97 (!)
2025 vs 2023 for my state of NSW. Testing for COVID-19 and influenza has not changed substantially that I'm aware of in this time.
The labs generally test for both on any swabs taken. The ratio of COVID-19:Flu
2025 - 0.49
2023 - 2.97 (!)
A study I wrote to the journal about in May this year was just retracted: www.sciencedirect.com/science/arti...
The review paper claimed that the majority of benefits in clinical trials could be explained by placebo effects.
The review paper claimed that the majority of benefits in clinical trials could be explained by placebo effects.
November 6, 2025 at 12:19 AM
A study I wrote to the journal about in May this year was just retracted: www.sciencedirect.com/science/arti...
The review paper claimed that the majority of benefits in clinical trials could be explained by placebo effects.
The review paper claimed that the majority of benefits in clinical trials could be explained by placebo effects.
At a minimum, a Japanese or Dutch style government mandated and overseen insurance system would probably reduce the cost overall. In the Netherlands, for example, they set the cost of the standard cover and mandate that everyone gets at least this level of cover. You can choose your insurer.
November 4, 2025 at 11:18 PM
At a minimum, a Japanese or Dutch style government mandated and overseen insurance system would probably reduce the cost overall. In the Netherlands, for example, they set the cost of the standard cover and mandate that everyone gets at least this level of cover. You can choose your insurer.
It gets better/worse.
November 4, 2025 at 8:08 AM
It gets better/worse.
One thing that always identifies charlatans - they blame patients when their treatment fails.
November 3, 2025 at 6:14 AM
One thing that always identifies charlatans - they blame patients when their treatment fails.
7/n For both SARS-CoV-2 and influenza, the risks of heart attack and stroke were very high in the first week of infection (red) lower up to 14 days after (yellow) and declined almost entirely by 28 days post-infection (blue).
November 3, 2025 at 4:45 AM
7/n For both SARS-CoV-2 and influenza, the risks of heart attack and stroke were very high in the first week of infection (red) lower up to 14 days after (yellow) and declined almost entirely by 28 days post-infection (blue).
3/n When you aggregate together cohort studies, influenza and COVID-19 have similar elevated risks when compared to control cohorts over the long-term - about a 60% increased risk of heart attacks, strokes, and other heart disease.
November 3, 2025 at 4:45 AM
3/n When you aggregate together cohort studies, influenza and COVID-19 have similar elevated risks when compared to control cohorts over the long-term - about a 60% increased risk of heart attacks, strokes, and other heart disease.
Headlines are saying that COVID-19 and flu TRIPLE your risk of heart attack. This is just a painful story about pretty weak science.
If nothing else, the "triple" risk is only over the first 14 days post-infection and declines enormously after that!
1/n
If nothing else, the "triple" risk is only over the first 14 days post-infection and declines enormously after that!
1/n
November 3, 2025 at 4:45 AM
Headlines are saying that COVID-19 and flu TRIPLE your risk of heart attack. This is just a painful story about pretty weak science.
If nothing else, the "triple" risk is only over the first 14 days post-infection and declines enormously after that!
1/n
If nothing else, the "triple" risk is only over the first 14 days post-infection and declines enormously after that!
1/n
Not a bad run for a random Saturday morning.
October 31, 2025 at 10:04 PM
Not a bad run for a random Saturday morning.
7/n None of the things measured in this paper are mentioned in the pre-registration. The primary outcomes are completely different!
October 29, 2025 at 10:20 PM
7/n None of the things measured in this paper are mentioned in the pre-registration. The primary outcomes are completely different!
5/n There were also some minimal differences in a per-protocol analysis where the authors allocated individuals based on their self-reported sedentary time rather than the random allocation.
October 29, 2025 at 10:20 PM
5/n There were also some minimal differences in a per-protocol analysis where the authors allocated individuals based on their self-reported sedentary time rather than the random allocation.
4/n So why the headlines?
Well, in a secondary analysis at maximal exercise intensity there was a small but statistically significant difference on one marker. See if you can spot it!
Well, in a secondary analysis at maximal exercise intensity there was a small but statistically significant difference on one marker. See if you can spot it!
October 29, 2025 at 10:20 PM
4/n So why the headlines?
Well, in a secondary analysis at maximal exercise intensity there was a small but statistically significant difference on one marker. See if you can spot it!
Well, in a secondary analysis at maximal exercise intensity there was a small but statistically significant difference on one marker. See if you can spot it!
3/n The main outcomes reported in the study are a variety of metabolic markers. On these markers, there was no effect whatsoever for the intervention.
October 29, 2025 at 10:20 PM
3/n The main outcomes reported in the study are a variety of metabolic markers. On these markers, there was no effect whatsoever for the intervention.
A bunch of headlines are proclaiming that you can boost your metabolism and you don't even have to exercise!
Except, that's not really true. Let's look at the study. 1/n
Except, that's not really true. Let's look at the study. 1/n
October 29, 2025 at 10:20 PM
A bunch of headlines are proclaiming that you can boost your metabolism and you don't even have to exercise!
Except, that's not really true. Let's look at the study. 1/n
Except, that's not really true. Let's look at the study. 1/n