Today’s Top 10 Health News Stories—Translated Into Plain English.

That 93%–94% figure is real—it comes from the 176-week SURMOUNT-1 trial in the New England Journal of Medicine (PMID 39536238, N = 1,032)—but it is a Relative Risk Reduction (RRR). Here is the plain-English reality check: in the placebo group (people who took a dummy shot plus basic diet/exercise coaching for more than three years), only 13.3% developed Type 2 diabetes. That means 86.7% of people on placebo did NOT get diabetes anyway. Tirzepatide lowered that 13.3% rate down to 1.3%—which is a 12.1% Absolute Risk Reduction (ARR) and a Number Needed to Treat (NNT) of 9. Even more important: when the researchers stopped the injections for just 17 weeks at the end of the trial, weight and blood sugar immediately started bouncing back.

Here is the hard law of chemistry from our audit of the NIH Dietary Supplement Label Database (DSLD): pure magnesium bisglycinate has a molecular weight of 172.31 g/mol, and two big glycine amino acid molecules make up 85.9% of that weight. Only 14.1% of the powder is actual elemental magnesium . That means to get 400 mg of real elemental magnesium, you need 2,837 mg of powder—which physically takes 3 to 4 large Size-00 capsules to hold! Any bottle claiming 400–500 mg in a single capsule is either quoting the total weight of the powder (giving you only 70 mg of actual magnesium) OR secretly spiking the capsule with cheap Magnesium Oxide (which absorbs at ~4% in human trials, PMID 11794633, and acts like a laxative) .

In the primary New England Journal of Medicine trial (PMID 37351564, N = 272) and companion Lancet diabetes trial (PMID 37364590, N = 383), daily oral orforglipron produced 9.4% to 14.7% body weight loss at 36 weeks (compared to 2.3% on placebo) and dropped HbA1c by up to 2.1 percentage points . That matches injectable semaglutide (Ozempic/Wegovy)! However, here is the honest catch: 10% to 17% of participants stopped taking the pill early because of nausea or vomiting, especially when the dose was stepped up too fast .

In the landmark TIME trial published in The Lancet (PMID 36240838, N = 21,104), researchers randomly assigned 21,104 adults with high blood pressure to take their daily pills either in the morning (6:00–10:00 A.M.) or at night (8:00 P.M.–midnight) and followed them for 5.2 years . The result? 3.4% of the bedtime group had a heart attack, stroke, or vascular death compared to 3.7% of the morning group—a statistical tie (0.3% absolute difference, Hazard Ratio 0.95, p = 0.53) . The Canadian BedMed trial in JAMA confirmed the exact same tie .

In head-to-head sleep lab validation studies published in Sleep (PMID 32345045), researchers hooked adults up to gold-standard polysomnography (scalp EEG wires that directly measure brain waves, eye movements, and muscle tone) while they wore leading consumer smartwatches on their wrists . The verdict: wrist wearables are very good (~88%–93% accurate) at telling when you fell asleep and when you woke up, and 97%+ accurate at measuring overnight Resting Heart Rate (RHR) . However, because a watch on your wrist cannot read brain waves, its guesses at "Deep Sleep vs. Light Sleep vs. REM" disagreed with hospital EEG brain-wave recordings 30% to 45% of the time !

Here is the missing biochemical link proven in a landmark Vanderbilt University double-blind randomized controlled trial published in the American Journal of Clinical Nutrition (PMID 30541089, N = 180): the Vitamin D3 you swallow in a pill (cholecalciferol) is 100% biologically inactive when it enters your mouth . To turn it into the form your blood test measures (25-hydroxyvitamin D) and the active hormone your cells use (calcitriol), your liver enzyme (CYP2R1) and kidney enzyme (CYP27B1) require intracellular magnesium (Mg-ATP) at every single step . And according to official CDC NHANES dietary surveys, roughly 48% of Americans do not get enough magnesium from their food .

Let’s honor both halves of the science. First, in the NEJM SAMSON trial (PMID 33196154), researchers gave patients who had quit statins because of severe muscle aches 12 unmarked monthly bottles (4 with Atorvastatin 20 mg, 4 with an identical sugar pill, and 4 completely empty bottles) . Patients felt real pain—their pain scores jumped from 8.0 (no pill) to 16.3 on the statin—but their pain also jumped to 15.4 on the sugar pill (a 0.90 nocebo ratio) . Second, when you do have true pharmacological muscle soreness from a high-dose statin (like 40 mg or 80 mg), The Lancet RACING trial (PMID 35858640, N = 3,780) proved that stepping down to a moderate statin dose + generic ezetimibe 10 mg got more patients to an LDL under 55 mg/dL (72% vs. 58%) while cutting intolerance quitting by 44% (4.8% vs. 8.2%) !

In a landmark randomized crossover trial published in Diabetes Care (PMID 23761134) and confirmed across 7 randomized trials in Sports Medicine (PMID 35268055), researchers tested older adults with prediabetes in a metabolic chamber . When participants took a 15-minute brisk walk 30 minutes after meals instead of one uninterrupted 45-minute morning walk, their 3-hour post-dinner glucose spike dropped by 22 to 35 mg/dL (p < 0.01) and their 24-hour average blood sugar improved significantly —all with the exact same total minutes of walking!

In the 13-year NIH Diabetes Prevention Program Outcomes Study (DPPOS, PMID 26900641, N = 2,155) and a 4.3-year randomized controlled trial in the BMJ (PMID 20488910, N = 390, NNT for deficiency = 14), long-term Metformin cut blood Vitamin B12 levels by an average of 19% and doubled the risk of clinical or borderline B12 deficiency . Even more striking: patients on Metformin who developed low B12 had significantly higher rates of peripheral neuropathy (numbness and tingling in the feet)—which can look identical to diabetic nerve damage unless somebody orders a simple $20 B12 blood test !

According to the 2026 CMS Medicaid National Average Drug Acquisition Cost (NADAC) federal database, a U.S. pharmacy pays $0.03 per tablet ($0.90 for a 30-day bottle) for generic Pantoprazole 40 mg, $0.04 per tablet ($1.20/month) for Rosuvastatin 20 mg, and $0.03 per tablet ($0.90/month) for Losartan 50 mg . Yet a landmark study in JAMA Internal Medicine (PMID 29532078, N = 9.5 million claims) proved that insured Americans paying a routine $10–$20 generic copay overpaid on 23% of all prescriptions because PBMs pocket the spread between your copay and the 90-cent wholesale cost !