Longevity Digest — July 12–19, 2026: Heat training, OSK claims, and the measurement problem

The week in one screen

Peter Attia's new piece makes heat training look useful for one narrow purpose: raising hemoglobin mass. It does not yet make heat exposure a proven longevity or VO2 max intervention. David Sinclair spent the week pushing two familiar ideas more aggressively, that aging is reversible and that his lab's OSK reprogramming work is moving toward human findings, but his posts still contain no efficacy or safety data. Bryan Johnson's most concrete development is a female-health measurement project around Kate Tolo, not a new supplement dose. Rhonda Patrick supplied the most actionable protocol detail, along with two reminders that observational health signals are not treatment evidence.

Coverage window: July 12–19, 2026, Pacific Time.

Peter Attia: heat training has a signal, but not a finished protocol

Peter Attia, the physician behind peterattiamd.com, published "Does heat training build a better engine?" on July 18. The article reviews studies in mostly elite cyclists and other fit participants. The common setup was about five weeks of six 50-minute sessions per week at roughly 32°C (90°F), around 60% of VO2 max. Across studies, hemoglobin mass generally rose about 2–5%. 1

The most useful distinction is between the marker and the outcome. In one head-to-head study, altitude training increased hemoglobin mass by 3.5% ± 2.0%, while heat training increased it by 5.4% ± 3.9%; the difference was not statistically significant (p = 0.801). Attia notes that five of six trials did not show a clear improvement in lactate threshold, VO2 max, or sport performance. In the underlying five-week trial, heat exposure did raise hemoglobin mass by 2.4–2.6%, and the heat-suit group improved selected power measures, but that is a performance-training result in elite cyclists, not evidence that heat training slows aging. 2

Attia's conclusion is cautious optimism. Hemoglobin mass is a plausible intermediate measure, but it is still a surrogate. The evidence is young, much of it comes from one collaborative research group, and gains can move back toward baseline within roughly 10–14 days after normal training resumes. For a reader, heat training belongs in the category of an optional, supervised performance experiment. It should not be promoted as a general-purpose longevity intervention, and the published studies do not justify copying an elite-cyclist schedule without considering heat illness risk, fitness, medications, and hydration.

David Sinclair: stronger language on reversibility, no new human result

David Sinclair, a Harvard-based aging researcher and host of Lifespan, was active on X this week. On July 19 he posted a compact list of propositions ending with "Aging is a disease" and "Aging is reversible." The post is a statement of his research worldview, not a new study or clinical result. 3

Hours earlier, Sinclair wrote that "the race is on" and said Yuancheng in his lab discovered the benefits of OSKs almost a decade ago. He described a three-year path to publication and another six years before a human trial could begin, adding that he looked forward to reporting the findings. The important verb is report: this post announces an anticipated evidence milestone, not the result of one. It contains no human efficacy data, safety result, dose, or timeline that a reader can use to start an OSK protocol. 4

That distinction matters because Sinclair's language is now moving between established framing and translational promise in the same burst of posts. His July 19 statements support tracking the program; they do not support taking reprogramming compounds, buying an unvalidated product, or treating "reversible" as a demonstrated outcome in healthy adults.

Bryan Johnson: more measurement, not a new stack

Bryan Johnson, founder of Blueprint and Immortals, used a July 15 LinkedIn post to frame Kate Tolo's female-health project as a 90-day effort across three menstrual cycles. The post lists 1,900 biomarkers, 14.8 million data points, 100 tasks per day, 6–10 hours per day, and more than 50 devices. 5

A July 13 Wall Street Journal profile gives the project more texture: a four-hour morning routine beginning at 6:30 a.m., repeated temperature, heart-rate-variability, hormone, blood, stool, hearing, taste, and sleep-related measurements, plus a planned annual cost of about $2.6 million for equipment, testing, and staff. The article also quotes researchers warning that more data does not automatically produce better health outcomes, and that self-measurement should not be sold to other women as a proven path. 6

This is a research-participation and product-development story, not a consumer protocol. The number of biomarkers is not the same thing as validated clinical benefit; repeated measurements can describe a person without identifying which intervention caused a change. The week's concrete Blueprint-adjacent signal is therefore measurement and female-specific data collection, not a substantiated change to the supplement stack or a new dosage recommendation.

Rhonda Patrick: two personal doses and two evidence cautions

Rhonda Patrick, a biomedical scientist and host of FoundMyFitness, posted the week's clearest supplement protocol. On July 14 she wrote that she normally takes glutamine at 5 g/day and creatine at 5–10 g/day. During travel, sleep deprivation, or an exposure she thinks may threaten her health, she said she raises each to 15–20 g/day, split into doses. She described glutamine as fuel for immune cells and creatine as support for cellular energy, especially for the energy-demanding brain during jet lag. 7

These are Patrick's stated personal doses, not a clinical guideline or a trial-derived travel protocol. The post does not establish that increasing either supplement prevents infection, repairs sleep loss, or improves jet-lag recovery. Readers with kidney disease, pregnancy, medication use, or other relevant conditions should not treat a social post as a dosing clearance.

Patrick also shared a new microplastics finding on July 15. The European Society of Cardiology release describes a 61-person study comparing people with heart attacks, chronic ischemic heart disease, and normal coronary arteries. Micro- and nanoplastics were detected in 84% of the heart-attack group, versus 40% and 32% in the comparison groups; the release also mentions higher inflammatory markers including TNF-α and IL-6. The study was limited by sample size and does not prove that microplastics cause heart attacks. Patrick's own post used the same boundary, calling the association alarming while saying causation remains unknown. 8 9

On July 17, Patrick made a similar distinction about testosterone. She summarized an observational analysis in which off-label testosterone use was associated with a 51% higher relative risk of major cardiovascular events; among 227 men using testosterone, those without documented low testosterone had more mortality and cardiovascular events over 10 years. She explicitly said the analysis was observational, not causal, and separated it from the randomized TRAVERSE evidence in properly diagnosed hypogonadism. The practical takeaway is narrower than "testosterone is dangerous": avoid casual extrapolation from a prescribed therapy for diagnosed deficiency to liberal off-label use. 10

Where the four signals meet

Person	Intervention or claim discussed	What the evidence supports	What it does not support
Peter Attia	Heat training	A repeatable rise in hemoglobin mass in studied athletes	A reliable increase in VO2 max, sport performance, or lifespan
David Sinclair	Aging reversibility and OSK reprogramming	Continued monitoring of a research program	Starting an OSK protocol or claiming human benefit from this week's posts
Bryan Johnson	High-volume female-health measurement	A way to build a personal data record and generate hypotheses	Treating biomarker volume as proof of a treatment effect
Rhonda Patrick	Travel supplements, microplastics, and off-label testosterone	A personal protocol to evaluate cautiously; observational associations that need context	Universal dosing, causal claims, or replacing clinical assessment

There was no clean same-intervention convergence this week. The more defensible cross-expert agreement was methodological: Attia separated hemoglobin mass from performance, Patrick separated association from causation twice, and the Johnson project shows why measurement alone cannot tell readers what to do. Sinclair's posts sit at the other end of that spectrum, where the scientific claim is still ahead of the publicly reported human result.

What is reasonable to carry forward

If heat training interests you, treat it as a performance experiment with a clear endpoint such as hemoglobin mass or training output. Do not market it to yourself as an anti-aging therapy.
Treat Patrick's 15–20 g travel doses as her personal practice. Check the actual purpose, dose, product quality, and contraindications before considering any change to your own stack.
Read the microplastics and testosterone numbers as signals for follow-up, not as proof of cause. The study designs and the authors' own caveats matter more than the headline percentage.
Keep Sinclair's OSK claims on a watch list until human results, safety data, and a reproducible protocol are public.
Do not copy the scale of Tolo's measurement project. More measurements can be useful, but the decision-relevant question is whether a change improves a validated outcome and whether the intervention caused it.