Mike Johansen
@mikejohansenmd.medsky.social
Primary Care doc, likes research, teaches residents, #HSR, #MEPS, #FamilyMedicine
Pinned
Don't be a point-estimatologist.
Don't think you will beat the algorithms.
Don't do #quantifauxication.
Don't think you will beat the algorithms.
Don't do #quantifauxication.
Reposted by Mike Johansen
This is one of the most remarkable academic debacles I've ever seen.
A large RCT got published in BMJ. There are currently 44 Pubpeer comments, mostly about the data, including...well. Read for yourself.
pubpeer.com/publications...
A large RCT got published in BMJ. There are currently 44 Pubpeer comments, mostly about the data, including...well. Read for yourself.
pubpeer.com/publications...
PubPeer - Prevention of acute myocardial infarction induced heart fail...
There are comments on PubPeer for publication: Prevention of acute myocardial infarction induced heart failure by intracoronary infusion of mesenchymal stem cells: phase 3 randomised clinical trial (P...
pubpeer.com
November 11, 2025 at 1:19 AM
This is one of the most remarkable academic debacles I've ever seen.
A large RCT got published in BMJ. There are currently 44 Pubpeer comments, mostly about the data, including...well. Read for yourself.
pubpeer.com/publications...
A large RCT got published in BMJ. There are currently 44 Pubpeer comments, mostly about the data, including...well. Read for yourself.
pubpeer.com/publications...
Reposted by Mike Johansen
My son Seamus died 15 years ago today. He was killed by a careless driver in a crosswalk.
He walked at 9 months, almost as if he knew he wouldn’t have much time. His favorite thing to do was throw rocks into water. He loved the moon. He called butterflies “buddy buddies.” I miss him every day.
He walked at 9 months, almost as if he knew he wouldn’t have much time. His favorite thing to do was throw rocks into water. He loved the moon. He called butterflies “buddy buddies.” I miss him every day.
November 9, 2025 at 5:10 PM
My son Seamus died 15 years ago today. He was killed by a careless driver in a crosswalk.
He walked at 9 months, almost as if he knew he wouldn’t have much time. His favorite thing to do was throw rocks into water. He loved the moon. He called butterflies “buddy buddies.” I miss him every day.
He walked at 9 months, almost as if he knew he wouldn’t have much time. His favorite thing to do was throw rocks into water. He loved the moon. He called butterflies “buddy buddies.” I miss him every day.
Reposted by Mike Johansen
As the ~discourse~ seems to bend interminably towards Republicans trying to figure out how they can (further) HDHP-ify ACA coverage, it's worth revisiting what is probably our best (most rigorous) study on the effect of deductibles in health insurance.
academic.oup.com/qje/article-...
academic.oup.com/qje/article-...
November 10, 2025 at 9:51 PM
As the ~discourse~ seems to bend interminably towards Republicans trying to figure out how they can (further) HDHP-ify ACA coverage, it's worth revisiting what is probably our best (most rigorous) study on the effect of deductibles in health insurance.
academic.oup.com/qje/article-...
academic.oup.com/qje/article-...
This is just an amazing story.
LSU's initial offer was <1/2 what he was contractually owed.
www.espn.com/college-foot...
LSU's initial offer was <1/2 what he was contractually owed.
www.espn.com/college-foot...
Report: Kelly rejected LSU buyouts, seeks $54M
Former LSU football coach Brian Kelly reportedly rejected settlement offers of $25 million and $30 million dollars and is asking LSU for written confirmation, by Monday 6p ET, that he will get the $54...
www.espn.com
November 10, 2025 at 6:11 PM
This is just an amazing story.
LSU's initial offer was <1/2 what he was contractually owed.
www.espn.com/college-foot...
LSU's initial offer was <1/2 what he was contractually owed.
www.espn.com/college-foot...
Reposted by Mike Johansen
The editorial of this paper is kind of funny. They don't recommend use given "outsized results of early trials have failed to be replicated". Given the effect size is ginormous, the likelihood that any meta-analysis would not show that this is beneficial is very small.
www.nejm.org/doi/full/10....
www.nejm.org/doi/full/10....
Fish-Oil Supplementation and Cardiovascular Events in Patients Receiving Hemodialysis | NEJM
Cardiovascular disease is the leading cause of death in patients receiving hemodialysis,
yet effective preventive therapies remain limited. Supplementation with n−3 polyunsaturated
fatty acids, esp...
www.nejm.org
November 10, 2025 at 2:27 PM
The editorial of this paper is kind of funny. They don't recommend use given "outsized results of early trials have failed to be replicated". Given the effect size is ginormous, the likelihood that any meta-analysis would not show that this is beneficial is very small.
www.nejm.org/doi/full/10....
www.nejm.org/doi/full/10....
Reposted by Mike Johansen
We invented the most powerful weight loss drug ever and it might be an effective oncology drug too 😳🤯
"Low-dose RETA significantly reduced tumor volume by 3-fold compared to vehicle, demonstrating efficacy comparable to anti-PD-1 therapy, which achieved a 5-fold reduction in tumor volume"
"Low-dose RETA significantly reduced tumor volume by 3-fold compared to vehicle, demonstrating efficacy comparable to anti-PD-1 therapy, which achieved a 5-fold reduction in tumor volume"
November 9, 2025 at 10:03 PM
We invented the most powerful weight loss drug ever and it might be an effective oncology drug too 😳🤯
"Low-dose RETA significantly reduced tumor volume by 3-fold compared to vehicle, demonstrating efficacy comparable to anti-PD-1 therapy, which achieved a 5-fold reduction in tumor volume"
"Low-dose RETA significantly reduced tumor volume by 3-fold compared to vehicle, demonstrating efficacy comparable to anti-PD-1 therapy, which achieved a 5-fold reduction in tumor volume"
bahahahahahahhahahahahahha.
MORE RESEARCH LIKE THIS!
MORE RESEARCH LIKE THIS!
Who would've thunk?
In a multicenter randomized clinical trial including 200 patients with persistent AF undergoing cardioversion, the risk of recurrent AF was SIGNIFICANTLY ⬇️ in the group allocated to coffee consumption (47%) compared to abstinent group (64%).
#ScienceMatters
#DoctorsWhoLoveCoffee
In a multicenter randomized clinical trial including 200 patients with persistent AF undergoing cardioversion, the risk of recurrent AF was SIGNIFICANTLY ⬇️ in the group allocated to coffee consumption (47%) compared to abstinent group (64%).
#ScienceMatters
#DoctorsWhoLoveCoffee
Caffeinated Coffee Consumption or Abstinence to Reduce Atrial Fibrillation
This clinical trial compares the effect of caffeinated coffee consumption vs abstinence from coffee and caffeine on recurrent atrial fibrillation.
jamanetwork.com
November 9, 2025 at 9:03 PM
bahahahahahahhahahahahahha.
MORE RESEARCH LIKE THIS!
MORE RESEARCH LIKE THIS!
Reposted by Mike Johansen
When a man's livelihood depends on finding a spurious correlation. . .
What happened when I dug into some of the key studies and ideas quoted in The Tipping Point and its sequel… kucharski.substack.com/p/the-real-r...
November 9, 2025 at 4:05 PM
When a man's livelihood depends on finding a spurious correlation. . .
If someone constantly needs to tell you how good they are, they aren't that good.
“.. if anything, [OpenAI’s] endless barrage of statements, interviews, and communications .. is starting to become counterproductive (anyone, whether it’s a company or politician, that insists on saturating media cycles .. usually isn’t doing so from a place of confidence).”
- Vital Knowledge
- Vital Knowledge
November 9, 2025 at 12:26 PM
If someone constantly needs to tell you how good they are, they aren't that good.
The currents drugs are going to get cheaper b/c they are 2nd generation. 3rd generation drugs are approaching quickly. Who even knows how many generations there will be in the next 10 years.
Donald Trump has announced a deal to slash the cost of obesity drugs—but the biggest winners might be just a small number of patients, @nicholasflorko.bsky.social argues.
Trump’s Ozempic Deal Has a Major Flaw
Obesity drugs are still too expensive.
bit.ly
November 9, 2025 at 12:22 PM
The currents drugs are going to get cheaper b/c they are 2nd generation. 3rd generation drugs are approaching quickly. Who even knows how many generations there will be in the next 10 years.
November 9, 2025 at 12:16 PM
Pro tip: understand the small data before going to the big data regression mess.
November 9, 2025 at 11:24 AM
Pro tip: understand the small data before going to the big data regression mess.
Reposted by Mike Johansen
I appreciate @bmj.com follows a formal process, but just how much evidence do they need before adding an Expression of Concern.
Numerous PubPeer comments for stem cell for heart disease paper - which had huge media attention hailing it as a medical breakthrough.
pubpeer.com/publications...
Numerous PubPeer comments for stem cell for heart disease paper - which had huge media attention hailing it as a medical breakthrough.
pubpeer.com/publications...
PubPeer - Prevention of acute myocardial infarction induced heart fail...
There are comments on PubPeer for publication: Prevention of acute myocardial infarction induced heart failure by intracoronary infusion of mesenchymal stem cells: phase 3 randomised clinical trial (P...
pubpeer.com
November 9, 2025 at 10:42 AM
I appreciate @bmj.com follows a formal process, but just how much evidence do they need before adding an Expression of Concern.
Numerous PubPeer comments for stem cell for heart disease paper - which had huge media attention hailing it as a medical breakthrough.
pubpeer.com/publications...
Numerous PubPeer comments for stem cell for heart disease paper - which had huge media attention hailing it as a medical breakthrough.
pubpeer.com/publications...
Dumb statistical choices tend to beget dumb statistical choices.
November 9, 2025 at 10:49 AM
Dumb statistical choices tend to beget dumb statistical choices.
Reposted by Mike Johansen
This is, unsurprisingly, nonsensical. Is he suggesting eliminating health insurance and giving people a few thousand dollars instead? And then when they get a cancer diagnosis they just go bankrupt?
He is so unserious. That's why we are shut down and Americans know it.
He is so unserious. That's why we are shut down and Americans know it.
November 8, 2025 at 4:40 PM
This is, unsurprisingly, nonsensical. Is he suggesting eliminating health insurance and giving people a few thousand dollars instead? And then when they get a cancer diagnosis they just go bankrupt?
He is so unserious. That's why we are shut down and Americans know it.
He is so unserious. That's why we are shut down and Americans know it.
Reposted by Mike Johansen
Seems like a good time to do a special issue of @jhppl.bsky.social on the Politics of Private Health Insurance. Call for papers will be out soon, so stay tuned. There’s seem to be, uh, a few politically relevant topics to discuss. @mirandayaver.bsky.social @adrianna.bsky.social
Republicans waged a relentless, decades-long campaign to privatize Medicare and Medicaid
They fought even mild efforts to expand public insurance
Now they are objecting to extending extra ACA subsidies, supposedly because money flows to private insurers
We are supposed to take this seriously?
They fought even mild efforts to expand public insurance
Now they are objecting to extending extra ACA subsidies, supposedly because money flows to private insurers
We are supposed to take this seriously?
November 8, 2025 at 10:10 PM
Seems like a good time to do a special issue of @jhppl.bsky.social on the Politics of Private Health Insurance. Call for papers will be out soon, so stay tuned. There’s seem to be, uh, a few politically relevant topics to discuss. @mirandayaver.bsky.social @adrianna.bsky.social
Reposted by Mike Johansen
Senate update: Republicans are now trying to sneak a backdoor national abortion ban into their government funding bill. Republicans will stop at nothing to control women's health care decisions.
November 8, 2025 at 9:05 PM
Senate update: Republicans are now trying to sneak a backdoor national abortion ban into their government funding bill. Republicans will stop at nothing to control women's health care decisions.
Affordability week coming to a close.
WOLFERS: “.. I can’t tell you why the president lies, but I can tell you that everything he has said about prices being lower is a lie. .. I’m starting to think that he doesn’t actually care about the cost of living.”
@justinwolfers.bsky.social @svdate.bsky.social
www.huffpost.com/entry/trump-...
@justinwolfers.bsky.social @svdate.bsky.social
www.huffpost.com/entry/trump-...
November 8, 2025 at 5:38 PM
Affordability week coming to a close.
Reposted by Mike Johansen
initially thought that this was a shadow being projected on his face. didn’t realize he was in blackface
November 8, 2025 at 4:51 PM
initially thought that this was a shadow being projected on his face. didn’t realize he was in blackface
Heterogeneity of effect size!
The RRR is consistent in all these ascvd preventive treatments.
The RRR is consistent in all these ascvd preventive treatments.
Among patients with atherosclerosis or diabetes but no previous myocardial infarction or stroke, evolocumab led to a lower risk of major adverse cardiovascular events than placebo. Full VESALIUS-CV trial results: nej.md/4oWrIN6
#AHA25
#AHA25
November 8, 2025 at 3:47 PM
Heterogeneity of effect size!
The RRR is consistent in all these ascvd preventive treatments.
The RRR is consistent in all these ascvd preventive treatments.
Naming things is important
I guess Rubin choosing "Missing at Random" for the scenario in which it is evident that the data are not missing at random is a close contender.
November 8, 2025 at 1:47 PM
Naming things is important
Reposted by Mike Johansen
I feel like psychometrics is uniquely bad at naming things. Parallel, congeneric, tau-equivalent, essentially tau-equivalent measures? Configural, metric, scalar, residual invariance? Item difficulty defined so that the higher the difficulty, the easier the item???
November 7, 2025 at 5:45 PM
I feel like psychometrics is uniquely bad at naming things. Parallel, congeneric, tau-equivalent, essentially tau-equivalent measures? Configural, metric, scalar, residual invariance? Item difficulty defined so that the higher the difficulty, the easier the item???