9/ For those asking for sources—here are a few foundational references behind the key points in this thread.

Population health is complex. The data reflects that.

8/ Simple explanations spread faster than accurate ones.

Single-variable stories are easy.
Complex systems are harder.

Confidence beats nuance on social media.

Virality ≠ validity

8/ Simple explanations spread faster than accurate ones.

Single-variable stories are easy.
Complex systems are harder.

Confidence beats nuance on social media.

Virality ≠ validity

7/ If one variable explained longevity, epidemiology wouldn’t exist.

The Hong Kong example doesn’t disprove nutrition science.

It shows why context, measurement, and systems matter more than viral claims.

6/ There’s also a timing problem.

Meat intake in Hong Kong rose sharply after 1960.

Today’s elderly population largely grew up on very different diets.

Health outcomes reflect decades—not headlines.

5/ Longevity isn’t driven by a single variable.

Hong Kong benefits from:

• Strong primary care
• Prevention-focused public health
• Low smoking rates
• High daily movement

That’s what moves population outcomes.

4/ Even within countries, consumption isn’t evenly distributed.

In the U.S., ~12% of people consume ~50% of all meat.

Population averages hide who is actually exposed—and at what levels.

3/ Hong Kong is also a major transit hub.

Large volumes of meat pass through the city into mainland China.

That inflates per-capita estimates in ways that make simple comparisons unreliable.

2/ The headline number is misleading.

“Meat consumption” in Hong Kong is based on carcass availability—including bones, waste, and re-exported supply.

It does NOT equal what people actually eat.

Hong Kong consumes more meat per capita than almost anywhere—and has the world’s longest life expectancy.

This gets used as a “gotcha” against nutrition science.

But it’s a misunderstanding of how epidemiology actually works.

🧵

9/ Summary

8/ A lab leak would become more likely with multiple independent lines of affirmative evidence.

Absence of evidence ≠ evidence of origin.

7/ If early cases had clustered more clearly around lab-associated locations,
the lab leak hypothesis would be more likely.

Observed patterns matter.

6/ If records showed unexplained illness or exposure among lab personnel prior to the outbreak,
the lab leak hypothesis would be more likely.

Occupational records → critical evidence.

5/ If records revealed undisclosed research on related viruses,
the lab leak hypothesis would be more likely.

Missing records → incomplete evidence.

4/ If early genomes showed clear signatures of lab adaptation,
the lab leak hypothesis would be more likely.

This is testable in sequence data.

3/ If records showed a lab possessed a closely related precursor virus before the outbreak,
the lab leak hypothesis would be more likely.

No documented progenitor → no clear origin pathway.

2/ If early cases had clearly clustered around lab staff or lab-linked contacts,
the lab leak hypothesis would be more likely.

That would be a real epidemiologic signal.

If SARS-CoV-2 came from a lab leak, what evidence would make that theory more likely?

Not speculation.
Not politics.

Positive, testable evidence.

Here’s what that would look like 👇

11/ References

10/ Fair takeaway: Seheult’s framework is strongest on (1) circadian light, (2) plausible PBM mechanisms, (3) early human signals in select contexts. What’s needed next: replication, standardized dosing, and long-term endpoints.

9/ Practical nuance from “sun rules”: glass reliably blocks UVB (so indoor sun won’t do much for vitamin D). But IR/NIR transmission varies with window type/coatings—so one universal % claim won’t fit all situations.

8/ Important limit: these studies may show specific measured endpoint changes—they don’t prove broad claims (immunity, longevity, chronic disease prevention). That leap requires larger, long-duration human trials with hard outcomes.

8/ Important limit: these studies may show specific measured endpoint changes—they don’t prove broad claims (immunity, longevity, chronic disease prevention). That leap requires larger, long-duration human trials with hard outcomes.

7/ Systemic-effect claim: newer work suggests longer wavelengths can penetrate tissue and may produce distal effects. Some studies report measurable endpoint changes even when eyes are shielded—suggesting systemic signaling is possible.

6/ Clinical trials exist in specific contexts. Example: a randomized, triple-blind, sham-controlled ICU trial reported shorter ICU stay + improved mobility/strength with PBM. Promising—still needs replication across centers/protocols.

5/ A concrete human finding: 670-nm red light (15 min) reduced the glucose rise after a glucose challenge in healthy adults. Interesting acute physiology—NOT proof it treats diabetes or improves long-term metabolic outcomes yet.

4/ Key caution: mechanism ≠ broad clinical promise. PBM effects depend on wavelength, dose, timing, target tissue, and baseline health. “Works in cells” doesn’t guarantee “works for everyone, outdoors, daily” without outcome trials.

4/ Key caution: mechanism ≠ broad clinical promise. PBM effects depend on wavelength, dose, timing, target tissue, and baseline health. “Works in cells” doesn’t guarantee “works for everyone, outdoors, daily” without outcome trials.

3/ Mechanism: PBM biology is plausible and well described—red/NIR photons interact with mitochondrial chromophores (often cytochrome-c oxidase), shifting NO/ETC signaling → ↑ATP and downstream redox/inflammation signaling.