Bowel cancer screening cuts death risk 43%, estrogen leaves fewer Alzheimer's plaques, your marrow quietly resupplies your brain, and four other things worth your time this week.

Week of August 24th, 2026

The thread running through this week's research is size: not whether something works, but how much, and how much of the headline number survives contact with an honest analysis. We cover a Swedish screening program that separates the benefit of getting the invitation from the benefit of actually mailing the kit back, autopsy data that complicates twenty years of warnings about estrogen and dementia, a genomic detective story showing that immune cells from your blood migrate into your brain and become part of its immune system, a meta-analysis putting a hard number on stair climbing, a blood-pressure divergence between the two leading GLP-1 drugs that matters if you take a pill for hypertension, a million-person diet cohort whose effect sizes shrink dramatically under scrutiny, and a new way to read biological age off tissue architecture. Seven findings, graded honestly. Here's what's worth your attention.

1. The 43% Number Belongs to People Who Actually Mail the Kit Back

Colorectal cancer is among the most preventable cancers we have, and the barrier is almost never the science, it's the unopened envelope on the kitchen counter. Johannes Blom at Karolinska Institutet, with Lennarth Nyström and Håkan Jonsson at Umeå University, published an analysis on August 20th in JAMA Network Open of the Stockholm-Gotland fecal occult blood screening program, following 376,511 people for up to 14 years, during which 1,668 colorectal cancer deaths occurred. Being invited to screening was associated with a 26% lower risk of dying from colorectal cancer. Actually participating was associated with a 43% lower risk. The 17-point gap between those two numbers is, in effect, the price of non-participation, and it's the most useful thing in the paper.

Grade it as strong population-level evidence for a recommendation we already make, with the novelty sitting in the statistics rather than the biology. The authors adjusted for contamination bias (control-group members who get screened elsewhere) and for nonadherence, which is precisely why the participation figure runs so far above the intention-to-treat estimates most screening trials report. The honest caveat: people who return the kit are systematically more health-engaged than people who don't, so healthy-user bias plausibly inflates part of that 43%, and this is a program evaluation, not a randomized trial of participation. The practical version is unglamorous. Stool testing every one to two years from age 45, or colonoscopy on the schedule your risk warrants, and the test only counts if it goes back in the mail.

Blom J, Nyström L, Jonsson H. Fecal Occult Blood Screening Outcomes Adjusted for Contamination Bias and Nonadherence. JAMA Network Open. 2026. DOI: 10.1001/jamanetworkopen.2026.29856. https://doi.org/10.1001/jamanetworkopen.2026.29856

2. Estrogen-Only Therapy, and Fewer Alzheimer's Hallmarks in the Autopsied Brain

For two decades women have been told that menopausal hormone therapy raises dementia risk, a message that traces back to the Women's Health Initiative memory substudy and has shaped prescribing ever since. Jennifer Bruno, Jacob Shaw, and S.M. Hadi Hosseini at Stanford Medicine published a study in Neurology on August 12th that went looking in the tissue itself. Across 21,462 women, estrogen-only hormone therapy was associated with 39% lower odds of a clinical dementia diagnosis and 35% lower odds of Alzheimer's pathology at autopsy, scored on a composite of amyloid plaques, tau tangles, and neuritic plaque density. Among roughly 258 estrogen-only brains compared with about 2,701 from women who took no hormone therapy, 18% of the hormone users showed no Alzheimer's changes at all, versus 10% of non-users.

Grade it as the strongest neuropathological signal yet for estrogen, and still firmly observational. What makes it different from prior work is that it measures physical deposits rather than cognitive test scores, which removes one large layer of noise. What it cannot do is establish cause. Women prescribed estrogen decades ago differed from those who weren't in education, access, and baseline health, the formulations and timing bear little resemblance to current practice, and combination estrogen-progestogen therapy did not show the same association. This is not a reason to start hormone therapy for brain protection. It is a reason the blanket warning deserves revisiting, and it strengthens the timing hypothesis that when therapy starts matters more than whether it is taken at all.

Bruno J, Shaw JS, Hosseini SMH. Association Between Menopausal Hormone Therapy and Alzheimer Disease Neuropathology. Neurology. 2026. DOI: 10.1212/WNL.0000000000218413. https://doi.org/10.1212/WNL.0000000000218413

3. Your Bone Marrow Has Been Quietly Resupplying Your Brain

Every neuroscience textbook says the same thing: microglia, the brain's resident immune cells, seed the brain before birth and maintain themselves for life behind a blood-brain barrier that keeps circulating immune cells out. Julia Belk and colleagues in Siddhartha Jaiswal's lab at Stanford, with Howard Chang and a large collaboration, tested that assumption in Nature using an elegant trick: the somatic mutations each cell clone accumulates act as a natural barcode, so you can match cells in postmortem brain tissue to their lineage in the same person's blood. Across all 20 aged individuals examined, they found marrow-derived cells that had crossed into the brain, and single-cell mitochondrial DNA lineage tracing showed those infiltrating cells were nearly identical to microglia, sometimes making up a large share of the pool. Analysis of large sequencing cohorts then found that most forms of clonal hematopoiesis were associated with lower Alzheimer's risk.

Grade it as a textbook rewrite with nothing yet to do about it. This is human-only biology, absent in both mice and non-human primates, which is exactly why it went unnoticed for so long, and it also means decades of mouse microglia work may not transfer cleanly. The caveats are real: 20 brains for the lineage tracing, postmortem tissue, and the clonal-hematopoiesis-to-Alzheimer's link is association only. The significance is strategic. If the blood is continuously reinforcing the brain's immune cells through midlife and beyond, then a door we assumed was closed is open, and marrow health becomes brain health in a way nobody had budgeted for.

Belk JA, Zhang Y, Shi Q, et al. Somatic mutations reveal the ontogeny of microglia in human aging. Nature. 2026. DOI: 10.1038/s41586-026-10939-0. https://www.nature.com/articles/s41586-026-10939-0

4. Stairs: The Only Exercise You Never Have to Schedule

The hardest part of most exercise advice is the part where you have to find forty spare minutes. Stair climbing sidesteps that entirely, and a systematic review and meta-analysis by Sophie Paddock, Rahul Veerni, Bhalraam U and colleagues at the University of East Anglia and Norfolk and Norwich University Hospital, published in the American Journal of Cardiovascular Drugs, put numbers on it. Pooling nine cohort studies covering 480,479 participants followed for a median of 14 years, regular stair climbers had a relative risk of 0.61 for cardiovascular death (39% lower) and 0.76 for death from any cause (24% lower), with lower rates of heart attack, stroke, and heart failure as well. One large cohort inside the analysis suggested roughly six flights a day as the sweet spot, though the optimal dose is not settled.

Grade it as a meta-analysis of observational cohorts, which means the true effect is almost certainly smaller than 39%. Reverse causation is the obvious hazard: people who climb stairs routinely are people whose knees, lungs, and hearts already permit it, and no amount of statistical adjustment fully fixes that. The mechanism is nonetheless plausible and well characterized. Stair climbing is brief, repeated, genuinely vigorous work, the kind that raises VO2 max, which remains one of the strongest single predictors of all-cause mortality we can measure. If you take one thing from this: the elevator is a small, repeated, free decision, and the direction of the evidence is not ambiguous.

Paddock S, Veerni R, Bhalraam U, et al. Evaluating the Impact of Stair Climbing on Cardiovascular Risk Reduction: A Systematic Review and Meta-analysis. American Journal of Cardiovascular Drugs. 2026. DOI: 10.1007/s40256-026-00811-x. https://doi.org/10.1007/s40256-026-00811-x

5. Tirzepatide and Semaglutide Part Ways on Blood Pressure

Most GLP-1 conversations stop at weight and A1c, but blood pressure is where the two dominant drugs quietly diverge, and it matters most to the people already taking something for hypertension. Qing-Xin Chen, Man Guo, Fang-Yuan Teng and colleagues at Southwest Medical University pooled 32 randomized controlled trials covering 47,332 participants with type 2 diabetes or obesity, published August 15th in Endocrine. Tirzepatide cut hypertension-related adverse events sharply (relative risk 0.40, 95% CI 0.26 to 0.60) but roughly doubled hypotension-related events (RR 2.45, 95% CI 1.35 to 4.45), rising to 2.58 at higher doses. Semaglutide was essentially neutral on both counts, with neither the hypertension signal (RR 0.81) nor the hypotension signal (RR 1.39) reaching significance.

Grade it as a careful synthesis of randomized safety data with a narrow but genuinely actionable message. The limitation is that these are adverse-event reports collected across trials rather than blood pressure measured as a prespecified endpoint, so the precision is coarser than the confidence intervals suggest. Still, the clinical read is clear enough. If you are on an antihypertensive and you start or escalate tirzepatide, your blood pressure medication may need to come down, and lightheadedness when you stand up is worth reporting rather than shrugging off as dehydration. A home cuff and a scheduled check-in with whoever prescribes both drugs is the cheap insurance here.

Chen QX, Zhou XY, Wu Q, et al. Effect of tirzepatide and semaglutide on blood pressure: A systematic review and meta-analysis. Endocrine. 2026;91(1). DOI: 10.1007/s12020-026-04757-7. https://doi.org/10.1007/s12020-026-04757-7

6. A Million Americans, 38 Years, and Surprisingly Small Meat Effects

Few dietary questions generate more heat and less light than red meat, so a cohort this large deserves attention for its precision as much as its conclusion. Catherine Kelly, Joan Sabaté, Terryl Hartman, Alpa Patel and colleagues analyzed the American Cancer Society's Cancer Prevention Study-II in Current Developments in Nutrition: 988,378 adults, median baseline age 56, followed a mean of 27.5 years, with 626,017 deaths recorded through 2020. Age-adjusted analyses showed the expected dose-response pattern, but most of it dissolved once demographics, smoking, lifestyle, and BMI were accounted for. After full adjustment, low red and processed meat intake was the only category still associated with every mortality outcome, and the magnitudes were modest: about 4% lower stroke mortality and 2% to 5% lower cancer mortality. White-meat-only diets were tied to 7% lower all-cause mortality in men and 6% among those who reached age 80. Vegetarian diets showed few consistent associations.

Grade it as an enormous, long, well-executed observational study whose most valuable contribution is the size of the effect rather than its direction. The design limits are important: a single 28-item food questionnaire administered in 1982 and never repeated, only 0.9% of participants eating white meat only and 0.4% vegetarian, and the near-total collapse of the raw dose-response after adjustment, which is the classic signature of confounding rather than causation. The takeaway is not that meat doesn't matter. It's that cutting red and processed meat is a small lever, worth pulling, and no substitute for the large ones: training, blood pressure, glucose control, and showing up for screening.

Kelly CC, Sabaté J, Hartman TJ, et al. Associations between white meat-only and vegetarian diets with mortality from all causes, heart diseases, cancers, and stroke in the American Cancer Society's Cancer Prevention Study-II. Current Developments in Nutrition. 2026;10:109438. DOI: 10.1016/j.cdnut.2026.109438. https://doi.org/10.1016/j.cdnut.2026.109438

7. Reading Biological Age Off the Architecture of Your Tissue

Nearly every aging clock on the market reads chemical marks on DNA. Ernesto Abila, André Rendeiro and colleagues at CeMM in Vienna, with collaborators in Munich, Cologne, Leuven, and Melbourne, took an entirely different substrate. Publishing in Nature Medicine on August 14th, they trained deep learning models on 25,712 whole-slide histopathology images spanning 40 tissue types from 983 people in the GTEx cohort, producing tissue clocks that estimate biological age from the physical architecture of tissue. The clocks tracked established aging markers including telomere attrition, subclinical pathology, and comorbidity burden, and organ-specific age acceleration mapped onto modifiable demographic, lifestyle, and medical factors. The most consequential step: by pairing histology with transcriptomic data, the team predicted tissue-specific age gaps from blood samples, then validated the approach across independent cohorts for eight diseases including Alzheimer's, stroke, and Crohn's.

Grade it as a real methodological advance with no consumer product behind it and no clinical claim attached. This is a computational study built on an existing biobank, not a prospective trial, and predicting an organ's age gap is not the same as predicting what happens to the person carrying it. What it does well is move organ-specific aging from a molecular abstraction to something structural and visible, then show that a blood draw might eventually stand in for a biopsy. That's the piece worth watching. The whole field has been limited less by interventions than by the lack of a trustworthy measuring stick, and this is a serious attempt at building one.

Abila E, Buljan I, Zheng Y, et al. Histological aging signatures for monitoring tissue-specific aging and disease. Nature Medicine. 2026. DOI: 10.1038/s41591-026-04566-5. https://doi.org/10.1038/s41591-026-04566-5

Have a great week,

Dr. Smith and Winston

Medical disclaimer

The content above is for informational and educational purposes only and does not constitute medical, nutritional, or professional advice. It is not intended to diagnose, treat, cure, or prevent any disease and is not a substitute for personalized guidance from a licensed physician or other qualified healthcare provider. Individual needs and risks vary. Do not begin, stop, or modify any exercise regimen, dietary plan, medication (including tirzepatide, semaglutide, estrogen-based menopausal hormone therapy, or blood pressure medications), or supplement based on this article alone. Consult your doctor first — particularly if you have or suspect a cardiovascular, metabolic, kidney, gastrointestinal, or neurological condition, take prescription medications, or are pregnant or breastfeeding. Studies summarized here vary in design and quality; preclinical and animal findings often do not translate to humans, and associations reported in observational research do not prove cause and effect.