HPV-related health problems are falling — but why?
Some antivaccine voices insist the vaccine deserves little credit, and claim changing sexual behavior explains the trend instead.
This short thread looks at the evidence in probabilistic terms:
• direct vaccine effects
• herd immunity
• behavioral change
The pattern is not random — and it strongly favors vaccination as the primary driver.
1/ Some antivaccine accounts are now arguing:
“HPV-related health problems are falling, but it’s not the vaccine. It’s changing sexual behavior.”
That explanation is possible in a weak, partial sense.
But as the main explanation?
It fails several basic causal tests.
The HPV vaccine explanation fits the evidence far better because the declines are:
• strongest for vaccine-covered HPV types • strongest in vaccinated birth cohorts • larger where vaccine coverage is higher • seen first in HPV infection and genital warts • later seen in cervical precancer and cervical cancer • also seen indirectly in some unvaccinated groups, consistent with herd effects
That is not the fingerprint of generic behavior change.
That is the fingerprint of vaccination.
2/ A useful way to think about this is probabilistic.
What best explains the widespread decline in HPV-related disease?
Direct HPV vaccine effect: very likely the dominant explanation.
Vaccine-driven herd immunity: very likely an important amplifier.
Changing sexual behavior: plausible as a minor contributor in some settings, but unlikely to be the primary driver.
Why?
Because behavior change would not selectively reduce the specific HPV types targeted by vaccines. It would not predict the sharpest declines in cohorts vaccinated before sexual exposure. And it would not naturally create a coverage-response relationship across countries and programs.
Vaccination predicts all of those things.
3/ The most important point: HPV vaccine impact follows the expected biological timeline.
First, vaccine-type HPV infections fall.
Then genital warts fall.
Then high-grade cervical lesions fall.
Then cervical cancer falls.
That is exactly what population-level studies have found.
This is what causal coherence looks like: the upstream infection declines first, followed by downstream disease outcomes years later.
The antivaccine argument often skips this timeline and pretends we are only observing a vague fall in “HPV problems.”
We are not.
We are observing the predicted downstream consequences of preventing infection with carcinogenic HPV types.
4/ Herd immunity is not an alternative to the HPV vaccine.
It is one of the ways successful vaccination programs work.
When enough people are protected, transmission of vaccine-type HPV falls. That means even some unvaccinated people face lower exposure risk.
This is why declines in HPV-related outcomes have been observed beyond the directly vaccinated group, including evidence of indirect protection in unvaccinated females and reductions in genital warts among males after girls-only vaccination programs.
That is not evidence against vaccination.
It is evidence that vaccination programs can reduce circulation of the virus.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
1/ A Substack is circulating the claim that a Norwegian study shows “74% of new myocarditis cases were related to COVID-19 vaccination, while only 4.2% were related to infection.”
That is not what the study shows.
The paper is important because it validates that vaccine-associated myocarditis is real, rare, and concentrated in young males.
But the Substack turns a post-vaccination myocarditis validation study into a population-wide causal claim about all myocarditis in Norway.
That is a denominator error.
2/ What did the Norwegian study actually find?
Researchers reviewed suspected myocarditis cases occurring after COVID-19 mRNA vaccination in Norway.
Out of 4.1 million vaccinated people, they identified 177 cases where vaccination was considered the most likely cause.
That equals about 4.5 cases per 100,000 vaccinated individuals.
That is clinically meaningful.
It is not “nothing.”
But it is also not evidence that vaccines caused 74% of Norway’s myocarditis burden.
The study question was not:
“Among all myocarditis cases in Norway, what percentage was caused by vaccines vs infection?”
The study question was closer to:
“Among reported post-vaccination myocarditis cases, how many are clinically validated and likely vaccine-associated?”
3/ The “only 4.2% due to infection” claim is also misleading.
The study identified 10 cases adjudicated as SARS-CoV-2 infection-associated myocarditis within its reviewed case set.
But this does not prove infection plays only a minor role in myocarditis.
Why?
Because infection-associated myocarditis is not captured the same way as post-vaccination myocarditis unless the study is specifically designed to ascertain infection timing, testing status, prior infection, variant era, vaccination status, and myocarditis onset across the whole population.
You cannot use a vaccine-safety validation study as if it were a complete causal map of myocarditis etiology.
1/ For centuries, the debate focused on a question we now know the answer to:
Did Vikings reach North America before Columbus?
Yes.
Not only did Norse sailors reach North America nearly 500 years earlier, archaeological evidence now suggests they may have returned repeatedly for generations.
The more interesting question is no longer whether they came.
It is why so little of their presence appears to have survived.
Recent work from archaeology, dendrochronology, and population genetics points toward one of the most unusual episodes in human migration history:
A sustained trans-Atlantic connection that left remarkably little biological legacy.
2/ The story begins in Greenland.
When Erik the Red established Norse settlements there around 985 CE, the colonists faced an ecological problem that threatened their survival.
Greenland had grazing land.
It had fjords.
It had walrus ivory.
What it did not have was timber.
And timber was the strategic resource of Viking civilization.
Ships, homes, churches, tools, furniture, agricultural equipment—everything depended on wood.
Yet Greenland possessed almost no forests capable of supporting shipbuilding.
The question becomes unavoidable:
How did a maritime civilization survive for centuries in a place with almost no trees?
3/ The answer emerged from an unlikely source:
Thousands of tiny fragments of wood recovered from Norse sites in Greenland.
Researchers examined more than 8,500 specimens microscopically, identifying species through cellular structure.
Most matched expectations:
• Siberian driftwood
• Scandinavian imports
But a small group did not.
Among them were oak, beech, hemlock, and jack pine.
These species could not have originated from Greenland.
They could not have come from Iceland.
And ocean currents could not plausibly have delivered them.
The most likely source was northeastern North America.
The wood itself became evidence of repeated voyages.
For decades, cortisol has been portrayed as the villain of modern health.
The hormone that makes us gain weight.
The hormone that causes anxiety.
The hormone that destroys sleep.
But that story is incomplete.
Cortisol is one of the body’s most important survival signals. It helps regulate energy, immunity, metabolism, cardiovascular function, behavior, and recovery. Without it, life is not possible.
The real question isn’t whether cortisol is “good” or “bad.”
The question is:
What is the body trying to accomplish when cortisol rises?
In this thread, we’ll move beyond the outdated “stress hormone” narrative and explore cortisol through the lens of modern neuroscience, predictive processing, allostasis, metabolism, exercise, obesity, food noise, depression, Long COVID, and aging.
Because cortisol isn’t simply a stress hormone.
It’s a resource-allocation hormone helping the body navigate uncertainty.
1/ Cortisol’s Original Job
Cortisol did not evolve to make people anxious, gain weight, or struggle with sleep.
Its original purpose was survival.
For most of human history, threats were tangible: hunger, injury, infection, predators, cold, and uncertainty about where the next meal would come from. When the brain detected those challenges, cortisol helped mobilize energy, maintain blood pressure, sharpen attention, and prioritize immediate survival.
The problem is that modern humans still possess the same biological alarm system.
Today, the “predator” may be financial stress, job insecurity, relationship conflict, social isolation, or constant digital stimulation. Yet the nervous system often responds as though survival itself is at stake.
Cortisol is not a design flaw. It is an ancient adaptation trying to solve modern problems.
2/ Cortisol Is a Budgeting Hormone
Many people think cortisol creates energy.
It doesn’t.
Cortisol reallocates energy.
When the brain perceives increased demands, cortisol shifts resources toward processes needed immediately and away from processes that can wait.
More energy becomes available for:
• Alertness
• Blood glucose maintenance
• Cardiovascular readiness
Meanwhile, less energy is devoted to:
• Growth
• Reproduction
• Long-term repair
• Tissue rebuilding
In that sense, cortisol functions less like a “stress hormone” and more like a biological budget officer.
The question isn’t whether cortisol is good or bad.
The question is what priorities your body is funding—and what it is postponing.
GLP-1 drugs may be the most consequential medications of the last decade.
Originally developed for diabetes, they are now showing effects on appetite, cardiovascular disease, inflammation, fatty liver disease, kidney health, sleep apnea, and even behaviors linked to addiction and reward.
But the deeper scientists look, the stranger the story becomes.
What exactly are these drugs doing—and what do they reveal about hunger, metabolism, and human desire itself?
A short 🧵
@IntegralAnswers
1/ One of the most striking ideas from Ezra Klein’s conversation with Julia Belluz is that GLP-1 drugs may be teaching us something profound about obesity:
Hunger is not simply a stomach problem. It is a brain process.
Many people describe living with constant “food noise”—persistent thoughts about food, cravings, and the mental effort required to resist them. When GLP-1 drugs work, that noise often becomes dramatically quieter.
For some patients, the experience is revelatory. The issue was never a lack of character or willpower. The biological drive itself has changed.
That observation challenges decades of cultural assumptions about obesity. If a medication can make resisting food feel effortless for one person while another must fight cravings every waking hour, then physiology may play a much larger role than society has been willing to acknowledge.
As Belluz notes, genetics, neurobiology, environment, sleep, stress, and food availability all interact to shape behavior.
The implication is uncomfortable but important:
Many people judged for lacking discipline may actually be fighting biological forces that others never experience.
Then researchers started seeing effects they didn’t expect.
Clinical trials found significant reductions in cardiovascular events. Evidence emerged for benefits in fatty liver disease, kidney disease, sleep apnea, and possibly other chronic conditions.
What surprised researchers most was that some of these benefits appear to be only partly explained by weight loss.
Julia Belluz describes three broad possibilities currently being explored:
• Weight loss itself
• Reduction of chronic inflammation
• Direct effects on organs such as the heart, liver, and kidneys
The problem?
Scientists still don’t fully understand the mechanisms.
This is why GLP-1s are so fascinating. The drugs are producing real clinical benefits, but in many cases researchers are still working backward to understand exactly why.
Medicine occasionally encounters discoveries that work before we fully understand them.
The problem is that scientific papers contain far more than conclusions. They contain assumptions, methodological choices, statistical decisions, limitations, competing explanations, and uncertainties that may dramatically affect how the findings should be interpreted.
Over the years I’ve found that many disagreements about science don’t arise because people are looking at different data. They arise because they’re evaluating the same paper through very different lenses.
To improve the quality of my own reviews, I began using a structured evidence-review framework inspired by principles of evidence-based medicine, peer review, critical appraisal, and scientific skepticism.
The objective is not to prove a study right.
The objective is not to prove a study wrong.
The objective is to determine what the evidence can legitimately support.
This framework forces me to separate:
• What was studied
• What was claimed
• What the methods can support
• What the results actually show
• What remains uncertain
The process also helps guard against some of the most common errors in science communication:
• Narrative over numbers
• Claim inflation
• Correlation presented as causation
• Ignoring alternative explanations
• Overgeneralization beyond the study population
• Confusing statistical significance with clinical significance
The following visual cards outline the exact framework I use when reviewing medical studies, preprints, essays, Substack articles, and scientific claims before creating public-facing content.
Strong conclusions require strong evidence.
And perhaps the most important question in all of science remains:
“What observation would convince you that your hypothesis is wrong?”
LONG COVID TREATMENTS:
Evidence, Management, and Uncertainty
Long COVID is not one disease with one treatment.
It is a heterogeneous, multi-system condition likely involving overlapping biological processes including immune dysregulation, autonomic dysfunction, viral persistence, endothelial injury, metabolic impairment, and neuroinflammation.
That complexity explains why:
• no universal cure exists
• responses vary dramatically
• many therapies remain investigational
What does appear increasingly clear:
The best outcomes usually come from individualized, multi-system, symptom-guided care rather than one-size-fits-all protocols.
This thread explores:
• what appears clinically useful
• what remains uncertain
• what is promising but still speculative
2/15
WHY IS TREATMENT SO DIFFICULT?
Long COVID is biologically complex and highly variable between patients.
Major challenges include:
• no validated diagnostic biomarker
• fluctuating symptoms
• multiple overlapping mechanisms
• phenotype/endotype variability
• different organ systems involved simultaneously
Two patients may both have “Long COVID” while sharing very little biologically.
What we do know:
• PEM (post-exertional malaise) is real and clinically important
• autonomic dysfunction is common
• multi-system involvement is frequent
• symptom burden can be disabling
• individualized care matters
The future likely depends on identifying biologic subtypes and matching therapies accordingly.
3/15
TREATMENT FRAMEWORK MAP
Most current Long COVID treatments target:
symptoms
suspected mechanisms
functional recovery
—or some combination of all three.