The finding that travels furthest from a body of research is rarely its strongest. It is usually the one that answers the question people most want settled, regardless of how little evidence stands behind it.

The Synthesis: The Cochrane review of exercise for depression, updated in January 2026, pulled together 73 randomised controlled trials involving around 5,000 adults. Against no treatment, exercise reduced symptoms. But most of the trials behind that result were weak ones. Only 22 of the 73 properly concealed which treatment each participant would be given, and in every trial the participants obviously knew whether they were exercising. Seven trials met the highest standard on all counts. In those seven, the benefit shrank until it was barely distinguishable from none.

Against psychological therapies (i.e. talking treatments aimed at changing people's thoughts, emotions, or behaviours), across ten trials, exercise performed about the same, and that is the review's most reliable comparison. Against antidepressants it also performed about the same, but only five trials tested it, covering 330 people between them, which is too little to settle the question. Thirty-five new trials were added to this update and the overall picture did not move.

The Calibration: Coverage has compressed this into one claim: that exercise works as well as antidepressants. Five small trials cannot carry that. The comparison the review supports best is the one that gets the least attention: exercise and psychological therapy work about equally well. The advantage over doing nothing depends heavily on trial quality, and it shrinks to almost nothing in the best run studies. The review also found no difference in how many people finished each treatment, so the idea that exercise is easier to stick with is not established either. Exercise is one of the options worth weighing alongside standard depression treatment, not a replacement for it.

The Synthesis: Metformin's longevity reputation rests largely on Targeting Aging with Metformin (TAME), designed to randomise 3,000 adults aged 65 to 79 without diabetes to either metformin or a placebo. The design was published in Innovation in Aging in 2018. The American Federation for Aging Research, which manages the trial, is still seeking donors to launch it, so a decade of coverage rests on a study that has not begun.

What has been studied is metformin in people with prediabetes. A study published in the Journal of the American Medical Association in 2026 followed around 1,100 such adults for 21 years, after they had been assigned at random to intensive diet and exercise change, metformin, or a placebo. Multiple chronic conditions developed in 82% of the lifestyle group, 85% of the metformin group, and 87% of the placebo group. Only the lifestyle difference held up once age and other factors were accounted for. This was a long observational follow-up rather than a 21-year trial, and the placebo stopped when the original trial ended.

A second 2026 analysis, in eClinicalMedicine, pooled nine studies covering 827 adults with diabetes or related conditions. All of them followed the same kind of exercise program. Some also took metformin, and in that group the exercise did less for them. Their blood pressure fell less and their fitness improved less than in the group exercising alone. Blood sugar and cholesterol came out the same in both groups.

The Calibration: The case for metformin as a longevity drug still rests on animal work, a plausible mechanism, and comparisons among people already taking it for diabetes. The longest human evidence comes from adults with prediabetes, and there the diet and exercise program was linked to less chronic illness while metformin was not. Metformin and structured exercise are routinely prescribed together, on the assumption that each strengthens the other. In the trials that tested the combination, that did not hold, and whether the same applies to healthy adults taking metformin for longevity has not been tested.

The Synthesis: The Alcohol Intake and Health Study, published in the Journal of Studies on Alcohol and Drugs in 2026, drew on more than 7,200 articles to model lifetime risk against U.S. health data. Across cancer, cardiovascular disease, liver disease, and injury combined, no level of regular drinking showed a net protective effect. Small protective signals appeared below about three drinks a week, but they were too weak to rely on. Above that, risk climbs steadily. It passes 1 in 1,000 at roughly 6.5 drinks a week for men and 7 for women. At 8.5 drinks a week it reaches about 1 in 100 for both sexes, and at 14 drinks a week, about 1 in 25 for men. These are modeled estimates with wide margins. Researchers also still disagree about how much of the apparent benefit at low intake is an illusion created by the comparison groups, which can fill up with people who stopped drinking because they were already ill.

The Calibration: The belief that a daily glass of wine protects the heart is overestimated. Harder to see is how much risk builds at amounts guidelines long treated as ordinary. The 2025-2030 Dietary Guidelines for Americans dropped the daily ceilings of two drinks for men and one for women in favor of advice to drink less, and the Canadian Centre on Substance Use and Addiction placed low risk at two drinks a week or fewer in 2023. Those are different ways of communicating the same risk, not different biology.

Guideline reversals are often read as proof that health science cannot be trusted, but the pattern usually says less about failure than about how evidence accumulates. A single study, even a well-designed one, is a piece of evidence rather than a final answer. Its participants may be few or unrepresentative, and confidence grows as later studies point the same way in wider groups of people. Guidance often has to be issued before that happens, because clinical decisions cannot wait.

When a larger trial, a more diverse population, or longer follow-up later produces a different answer, updating the guidance is not an admission that the earlier recommendation was baseless. It may reflect a shift in certainty, in who the evidence applies to, or in the balance of benefit against harm. Media coverage, built around conflict and novelty, tends to flatten that process into a reversal.

The Grading of Recommendations Assessment, Development and Evaluation framework, known as GRADE and used by many guideline bodies, exists to make the certainty behind a recommendation explicit, so that recommendations can be strengthened or weakened as evidence arrives. The useful question when guidance moves is what specifically changed: a new trial, a different population, a longer follow-up, or a revised threshold. That question separates a field adjusting to better evidence from one discovering its previous approach was baseless.

The Claim: Cold plunge content across social media and wellness marketing presents daily ice baths as an immune booster.

The Distortion: A 2025 systematic review and meta-analysis in PLOS ONE, covering 11 randomised studies and more than 3,000 adults, found no change in immune measures in the hours after immersion. The favorable signals came from what participants reported about themselves, such as sick days off, not from immune function.

The Calibration: The physiological response to cold is real. The immune benefit marketed alongside it has not been shown.

The Claim: Telehealth clinics market testosterone therapy for tiredness, low sex drive, and the sense of slowing down in middle age. Some prescribe it after a short online questionnaire, without confirming that a man's testosterone is actually low.

The Distortion: Since 2015, testosterone labels have said that safety and effectiveness had not been established in men whose levels are low simply because of age. In June 2026, the Food and Drug Administration requested removal of that limitation after a broader evidence review that included TRAVERSE. That trial followed more than 5,200 men with symptoms and confirmed low testosterone and found no rise in heart attacks or strokes. A separate regulatory finding went the other way: blood pressure studies required by the FDA showed that testosterone products raise blood pressure, prompting a class-wide warning in 2025.

The Calibration: Testosterone therapy has an established role when a deficiency has been appropriately diagnosed. Neither finding establishes a case for testosterone as an anti-aging treatment in men whose levels are normal. The useful distinction is between replacing a hormone that is documented as low and boosting one that is not.

Stay calibrated.

If The Calibration brought you a clearer perspective today, forward it to someone who values the same.

Received this from a friend or colleague? Subscribe here to get it directly.

Exercise for Depression. Cochrane Database of Systematic Reviews, 2026.

Lifestyle and Metformin Interventions and Risk of Multimorbidity in Adults With Prediabetes. JAMA, 2026.

Trials of Geroscience-Based Therapeutics: The Targeting Aging With Metformin (TAME) Example. Innovation in Aging, 2018.

TAME - Targeting Aging With Metformin. American Federation for Aging Research, 2026.

The Effects of Metformin and Exercise Training on Cardiorespiratory, Blood Pressure, and Metabolic Adaptations Across the Spectrum of Glucose Dysregulation: A Systematic Review and Meta-Analysis. eClinicalMedicine, 2026.

Alcohol Intake and Health Study: No Protective Effect at Low Levels, With Mortality Increasing to 1 in 25 at 14 Drinks Per Week. Journal of Studies on Alcohol and Drugs, 2026.

Dietary Guidelines for Americans, 2025-2030. U.S. Department of Agriculture and U.S. Department of Health and Human Services, 2026.

Canada's Guidance on Alcohol and Health: Final Report. Canadian Centre on Substance Use and Addiction, 2023.

GRADE Guidelines: 1. Introduction - GRADE Evidence Profiles and Summary of Findings Tables. Journal of Clinical Epidemiology, 2011.

Effects of Cold-Water Immersion on Health and Wellbeing: A Systematic Review and Meta-Analysis. PLOS ONE, 2025.

HHS Announces Requested Updates to Testosterone Therapy Product Labels. U.S. Department of Health and Human Services, 2026.

FDA Issues Class-Wide Labeling Changes for Testosterone Products. U.S. Food and Drug Administration, 2025.

Cardiovascular Safety of Testosterone-Replacement Therapy. New England Journal of Medicine, 2023.

The Calibration is an editorially curated health briefing. Content is drawn from published research and credible sources for informational purposes only. It does not constitute medical advice, clinical guidance, diagnosis, or treatment, and is not a substitute for consultation with a qualified healthcare professional. Nothing in this briefing should be used to make health or treatment decisions or to delay seeking professional medical advice.

Sponsored content, when present, will always be clearly identified.