LUVO Blogs
What the research actually says.
No hype, no borrowed authority. Every claim links to the primary study so you can read it yourself — including the studies that complicate our own products.
Chronic fatigue is not a supplement problem, and we will not pretend it is.
Fourteen trials, 809 people, and a high risk of bias in most of them. This is the article where the honest recommendation is a clinician who knows the criteria, not a bottle.
Nine years of magnesium and anxiety reviews, all ending in the same sentence.
Half the trials in anxious samples were positive, not one used a validated stress measure, and the largest magnesium-for-stress study had no placebo group. Every review since has asked for the trial nobody has run.
Sea moss: iodine, heavy metals, and almost no human trials.
An NIH database record says plainly that no medicinal value has been proven, and that sea moss carries arsenic, lead, cadmium and mercury from the water it grew in. We sell it anyway. Here is why that is complicated.
Sleep loss cost 7.56% of performance. Adding sleep back is harder to prove.
A meta-analysis of 227 outcomes measured what losing sleep does. The famous study on gaining it had eleven basketball players and no control group. The asymmetry is worth understanding.
Do sleep supplements stop working? Usually it was never working that hard.
Tolerance to diphenhydramine was complete in three days. Melatonin showed none over twelve months. And 63.6% of the drug response in insomnia trials shows up in the placebo arm.
If you had to halve your supplement spend, cut these first.
A ranked list of what to drop, written by a company that sells some of it. Nothing here is unsafe. It is just where the evidence stops justifying a standing monthly charge.
Brain training makes you better at the game. That is the whole finding.
Seven researchers reviewed every study the brain-training companies themselves cite. They found strong evidence for improvement on the trained task and little for anything else.
If you only fund three longevity habits, none of them is a capsule.
661,137 people for exercise. 1,382,999 for sleep. 1,863 for protein and training. Then the supplement aisle, where the largest mortality dataset for any ingredient is zero people.
Caffeine is the best-evidenced legal ergogenic, and the dose is per kilogram of you.
Three to six milligrams per kilogram is where the effect is consistent. In one trial of 101 athletes, 4 mg/kg also made one genotype 13.7% slower over 10 km.
Heat and humidity break supplements down faster than time does.
The date on the bottle is a manufacturer's estimate under ideal conditions. Where you keep it matters more than how long you have had it, and the bathroom cabinet is the worst option in the house.
Protein is the only macronutrient with a satiety effect that keeps showing up.
Raising protein to 30% of calories cut spontaneous intake by 441 calories a day for twelve weeks. But acute trials show it does not reduce what you eat at the very next meal, which is the interesting part.
Shift work is a clock problem. No amount of discipline reschedules a clock.
Melatonin after a night shift bought about 24 minutes of daytime sleep in pooled Cochrane data. That is the size of the best-evidenced tool for one of the hardest sleep problems there is.
Four ingredients carry a prenatal vitamin. The rest is mostly packaging.
Your care provider decides what you take and when — this is background reading for that conversation, covering what the evidence supports and what analyses found actually inside the bottles.
In one study of overtrained athletes, the difference was not how much they trained.
Thirty-nine athletes, fourteen of them overtrained, and their training patterns matched. What independently predicted the syndrome was carbohydrate, protein and total calorie intake.
How to stack supplements without wasting your money.
Start eight things at once and you will never know which one worked. Four rules on sequencing, timing windows and the mineral pairs that genuinely compete for absorption.
Resveratrol was going to be the longevity drug. Reading why it wasn't is a skill.
A 2003 yeast experiment became a global supplement category. The chain of inference that carried it there broke in three separate places, and knowing where is a skill you can reuse.
Saffron's mood results show up on the scales patients fill in, not the ones clinicians do.
Thirty-four trials, 1,769 people, moderate-certainty evidence on self-report inventories — and no significant effect on the clinician-rated scales measured in the same literature. That gap is the whole story.
The gut-brain axis is real. It is not yet a product category.
One 44-person trial cut depression scores and changed amygdala activity on fMRI. Pooled across 70 trials the mood effect is moderate, and heterogeneous enough to hold loosely.
Waking at 3am is a different problem from not being able to fall asleep.
In a survey of 8,937 US adults, 35.5% woke at least three nights a week. The things that help you drop off are not the things that reliably hold the second half of the night.
No supplement does what a GLP-1 drug does. Including ours.
Semaglutide moved body weight 14.9% in 68 weeks in a trial of 1,961 people. The best-run supplement trials in the same category top out near 5 kg. That gap is not a marketing problem.
Six hours a night for two weeks costs you two full nights of sleep.
In a laboratory dose-response study, fourteen nights at six hours produced deficits equal to two nights of no sleep at all. The people in the study did not notice it happening.
Creatine and hair loss: what the one study everyone cites actually measured.
A single 2009 rugby study built a fifteen-year rumour. It measured hormones, not hair. Here is what it found, what it never looked at, and what the research has done since.
The postpartum comeback: keep going beats doing it perfectly.
Broken sleep, higher nutrient requirements than before you were pregnant, and a warning about one of our own products. The best-evidenced thing on this list isn't a supplement.
A prebiotic capsule is a rounding error next to the fibre you are not eating.
Around 5% of Americans hit the fibre target, and most believe they already do. Supplemental fructans reliably raise two bacterial genera — and reliably fail to change microbiome diversity.
Panax ginseng for fatigue: modest, real, and routinely oversold.
Pooling 19 randomised trials, ginseng did not beat control overall. In the subgroup that matches most buyers it did, by an effect size the reviewers themselves call small.
Iodine is the mineral where too much looks like too little.
Excess iodine and deficient iodine can produce the same thyroid picture, and sea vegetables are the easiest way to overshoot without noticing. Both ends of the range are worth respecting.
Progressive overload is the only part of training a supplement cannot do for you.
A meta-analysis of 140 studies and 1,433 effect sizes mapped what actually drives strength. None of the variables it identified appear on any supplement label.
NAD+ and NMN: what the human trials show versus what the internet claims.
NAD precursors reliably do the thing on the label — they raise blood NAD. Whether that produces anything you would notice is a separate question, and the human answer is still modest.
The pill changes how long your coffee lasts. Here are the actual numbers.
Combined oral contraceptives cut caffeine clearance by roughly half, and pregnancy slows it further. Cycle phase does something much smaller. None of it is a reason to give up coffee.
A massage gun is a range-of-motion tool that people buy as a repair tool.
One percussive treatment added 5.4 degrees of ankle range in a controlled trial. A twelve-trial meta-analysis found no clear effect on soreness, and longer treatments scored worse.
Supplements are not ADHD treatment. Here is what the trials show anyway.
ADHD is a medical diagnosis with treatments that work. The supplement literature has one small effect that survives blinding, one trial that split down the middle, and a lot of unblinded parent ratings.
Melatonin is a timing signal. Most bottles sell it as a sedative.
The trial that compared doses head to head found 0.3mg restored sleep efficiency, while 3mg left melatonin circulating into the next morning. The shelf mostly starts at 5mg.
We sell shilajit. The human evidence is thinner than the marketing.
Two small trials, one reporting a p-value and almost nothing else. Here is what has actually been shown in humans, what has not, and what would have to be true for it to earn your money.
Nothing we sell competes with your treadmill time or your grip strength.
122,007 stress tests, 139,691 people squeezing a dynamometer, half a million UK Biobank participants. The two best-evidenced longevity markers are things you train, not things you buy.
The K2-with-D3 argument is mechanistic, and the outcome trials have not caught up.
A coherent biological story, a supportive observational cohort, and then a 365-man randomised trial that found nothing. This is what a promising idea looks like before it is proven.
Healthy people spike too. A CGM shows you that and then leaves you to interpret it.
In 153 healthy non-diabetic people wearing a blinded sensor, glucose sat above 140 mg/dL for a median of 30 minutes a day. The number the app flags as an event is, for most people, Tuesday.
Bitters are traditional, plausible, and barely studied in humans.
Six hundred milligrams of quinine delivered into the duodenum measurably slows gastric emptying. Two drops of tincture on your tongue before dinner is a different experiment entirely.
Should you cycle adaptogens? Nobody has ever run that trial.
Five days on, two days off. Eight weeks on, two weeks off. Every schedule you have read was invented by someone with something to sell. Here is what the trials actually did, and what that honestly leaves you with.
Enzymes or probiotics? One is a missing tool, the other is a missing ecosystem.
Lactase against lactose is one of the cleanest results in digestion. Broad-spectrum enzyme blends for people with no diagnosed insufficiency are among the weakest evidence in the aisle.
GlyNAC rests on one 24-person trial from one lab. That is worth saying out loud.
The results look extraordinary: oxidative stress, mitochondrial function, strength and gait all improved in sixteen weeks. Every human trial so far comes from the same research group.
Lion's mane and cognition: reading the human trials without the enthusiasm.
Thirty people in 2009 are carrying an enormous amount of internet weight. Here is every human trial worth knowing, what each one measured, and where the evidence gets thin.
Maca: the libido claim has trials behind it, the energy claim mostly does not.
Four randomised trials, one systematic review, and a consistent finding that maca does nothing to testosterone. The gap between what was studied and what gets sold is the whole story.
Most of a pre-workout scoop is doing nothing. Here is the line-by-line audit.
An analysis of the 100 top-selling pre-workouts found 18.4 ingredients per serving and 44.3% of them hidden inside proprietary blends. Only a few arrive at a dose worth paying for.
Two grams of curcumin left serum levels undetectable. That is the whole story.
Curcumin's activity was never really the question. Absorption was, and it is why two meta-analyses of the same ingredient reached opposite conclusions about muscle soreness.
Gotu kola's best evidence is about legs, not memory.
The meta-analysis found no significant effect on any cognitive domain. It did find higher alertness and lower anger an hour after a dose — and the circulation trials are stronger than either.
Tart cherry juice has real melatonin in it. The trials are still tiny.
Twenty people, fifteen people, eight completers. The direction is encouraging and the sample sizes are the reason we are not going to sell you a cherry supplement.
The blood panel worth buying before you buy a single supplement.
You are forty-three and something has slipped. The most useful thing you can spend money on is not on our shelf — and the testosterone number you want is the one most men get measured wrong.
Triglyceride, ethyl ester, krill and algal: what actually differs.
Four omega-3 formats, one head-to-head absorption trial, and a 51-point spread between the best and worst. Here is where the difference is real and where it is a price premium.
Biological age tests disagree with each other. That is the most useful thing about them.
Eleven measures of biological ageing were run on the same 964 people and agreed poorly. Six prominent epigenetic clocks deviated by up to nine years between replicates of the same sample.
Ox bile is for a specific person, and it is probably not you.
The human literature on supplemental ox bile is close to a single 1992 case report. Meanwhile 38% of people with functional diarrhoea have too much bile reaching the colon, not too little.
Burnout is a description of your job. Fatigue is a description of your body.
The WHO classifies burn-out as an occupational phenomenon and specifically not as a medical condition. A review of 182 studies found at least 142 different definitions of it. Neither fact is an argument for a supplement.
How much protein you actually need, according to the dose-response data.
The RDA is 0.8 g/kg. The plateau for people lifting weights sits near 1.6. Both numbers are real, they answer different questions, and most of the argument is people confusing the two.
Does supplement timing matter? Mostly less than you have been told.
Four timing rules have real human data behind them. Most of the rest is inference dressed up as protocol. Here is the short list worth following and the long list you can drop.
The leucine threshold is why a plant protein scoop has to be bigger.
Oat and lupin isolates are about 21% essential amino acids. Whey is 43%. Leucine content across plant isolates ranges from 5.1% to 13.5%. The scoop has to follow the chemistry.
Creatine's cognitive effect turns up when the brain is under strain, not otherwise.
A meta-analysis found a memory effect concentrated in adults aged 66 to 76 and none at all in the young. Europe's food regulator read 21 studies and said cause and effect was not established.
Time-restricted eating works the way any diet works: by making you eat less.
A twelve-week trial found a 0.26 kg difference against normal eating. A twelve-month one found 1.8 kg, not significant. The clock is a compliance tool, not a metabolic switch.
Beetroot works better the less fit you are, which is not how it gets sold.
Across 21 people spanning a VO2peak range of 28 to 82, the benefit from dietary nitrate tracked inversely with fitness. That inversion is the most useful fact about the ingredient.
Cortisol is not the villain, and your at-home test is probably measuring noise.
Cortisol is meant to surge in the 30 to 45 minutes after you wake up. Here is what a normal day looks like, why the popular tests are so unreliable, and the short list of times a real workup is warranted.
Omega-3 cut soreness by less than the amount researchers say you can notice.
Twelve trials produced a statistically significant reduction in muscle soreness of 0.93 points. The threshold for a change a person can actually perceive is 1.4. That gap is the article.
Valerian: people feel it working, and the instruments do not agree.
An umbrella review of eight systematic reviews found no evidence of efficacy for insomnia — alongside consistent subjective improvement. That gap is the whole story of this herb.
Cycle syncing is everywhere. The training evidence behind it is thin.
A meta-analysis of 78 studies found a trivial performance difference across the menstrual cycle, graded low quality. That is not the same as saying your experience of it is imaginary.
Most testosterone boosters do not raise testosterone. Here is what does.
A review of 52 trials across 27 marketed ingredients found that most of them fail. The levers that do show up in the data are less profitable, and the biggest one is free.
Calcium from food and calcium from a pill are not the same intervention.
The same total intake produced opposite associations depending on where it came from. If you are taking calcium for your bones, this is the trade-off worth seeing first.
The word adaptogen was coined in 1947 and no regulator has accepted it since.
It was invented to describe a substance that raises non-specific resistance to stress. Europe's herbal medicines committee examined the concept in 2008 and declined to accept the term at all.
Vitamin D trials keep failing. The genetics suggest they enrolled the wrong people.
Two trials totalling 47,000 adults found no mortality benefit. A Mendelian randomisation study in 307,601 people found the benefit sitting almost entirely below 50 nmol/L. Both results fit.
Colostrum and the gut barrier: two trials pointing opposite ways.
Sixty grams a day for eight weeks raised gut permeability by 251% in runners. Fourteen days of a smaller dose cut the exercise-induced rise by 70%. Both are real randomised trials.
Your metabolism did not slow down at 30. It holds flat from 20 to 60.
Doubly labelled water in 6,421 people aged eight days to 95 years found size-adjusted daily energy expenditure is stable across the whole of adult life. The decline starts at 63.
Under-recovering is a training load problem before it is a cortisol problem.
A review of 38 studies found resting hormone levels were mostly normal in athletes who were genuinely overreached. The monitoring tool that outperformed the blood test costs nothing.
Muscle starts drifting in your thirties. Strength leaves two to five times faster.
Whole-body MRI in 468 adults put the first decline in relative muscle mass in the third decade. What falls faster is force — and that is the number tied to independence.
Thermogenics raise your heart rate far more reliably than they lower your weight.
The strongest fat burner ever sold produced 0.9 kg a month and was pulled from the market. What replaced it is caffeine, and the cardiovascular effect is the best-measured thing about it.
Do B vitamins actually give you energy? Not unless you were short of them.
The word 'energy' on a B-complex label is doing two jobs at once. In biochemistry it means a co-enzyme in metabolism. On the front of a tub it means a lift. They are not the same thing.
The stress intervention with the most believable numbers is the one that is free.
Breathwork's pooled effect on stress is a modest g of −0.35 across twelve trials. That is smaller than the headline supplement figures, and it is the number we would trust more. Here is why.
CoQ10, statins and fatigue: the premise is right and the conclusion is not.
Statins really do lower blood CoQ10. Whether putting it back does anything for muscle pain or tiredness is where the evidence stops cooperating with the marketing.
L-theanine answers a calm question, not a sleep question.
Thirteen trials in 550 people point at 200-450 mg a day for sleep support. The largest meta-analysis could not pool a single sleep outcome. Both facts belong on the label.
The omega-3 trials that found something were studying people who were short.
A Cochrane review of three trials, a 3,501-person cognitive study inside AREDS2, and a meta-analysis of 24 trials that found exactly one effect — in people with low omega-3 status.
Vinegar really does blunt a glucose spike. It does not do the other thing.
Adding vinegar to a white-bread meal pulled its glycaemic index down to 64 in one trial. Then a 2024 study claimed 7 kg of weight loss in twelve weeks, and three researchers took it apart.
Creatine's recovery evidence is thinner and stranger than its strength evidence.
Two meta-analyses asked whether creatine speeds recovery and reached different answers. One of them found damage markers higher in the creatine group. Here is the honest read.
The supplement and drug interactions worth knowing before you add anything.
Most interaction warnings are theoretical. A handful are documented, common, and genuinely consequential — including one that can make hormonal contraception less reliable.
Intestinal permeability is real. Leaky gut, as sold, is a different thing.
Barrier dysfunction is measurable and it accompanies real disease. What has never been shown is that repairing it fixes anything — and no proposed treatment has been proven to repair it.
The omega-3 index predicts mortality. Taking fish oil mostly hasn't changed it.
Across 42,466 people in 17 cohorts, the highest blood omega-3 levels carried 15 to 18% lower mortality. Then 77,917 people were randomised to supplements and nothing happened. Both are true.
Caffeine, adenosine, and why the afternoon crash keeps arriving on time.
Caffeine does not make energy. It blocks the signal that you are tired — and once you drink it daily, the trial data suggests most of what you feel is your own baseline being restored.
Inositol has the best trial record on the shelf, and it still isn't high-quality evidence.
Twenty-six randomised trials in 1,691 women. A 1.79 times higher chance of a regular cycle than placebo. And an umbrella review that graded 85 outcomes and found not one at high quality.
Ashwagandha in men: the cortisol data is stronger than the testosterone data.
Twenty-three trials, 1,706 people, and a hormone effect that only appears in men. Also a trial where the hormones moved and the men could not feel any difference.
Brain fog is a symptom, not a diagnosis. Here is what to rule out first.
Researchers scraped 717 posts about brain fog and found at least seven different experiences under one phrase. Before a nootropic: sleep, iron, thyroid, B12, glucose, your medicine cabinet and your mood.
Most people do not need an electrolyte sachet. Some genuinely do.
Adding electrolytes alone to water did not reliably beat plain water for fluid retention in a controlled trial. Here is the short list of people for whom the sachet earns its place.
Tart cherry is sold for sleep and has better evidence for strength.
The most-cited tart cherry sleep study had fifteen people and beat placebo on one measure out of four. Its recovery file, across nineteen trials, is considerably stronger than that.
Glycine before bed: small trials, one funder, and a mechanism worth knowing.
Three grams of glycine at bedtime has a consistent direction in the human data and a plausible cooling mechanism. Nearly every study came from the same company's lab.
Training to failure is optional. What it is not is free.
Two meta-analyses covering thirty studies found failure no better than stopping short for strength, and at best trivially better for size. The cost shows up somewhere else.
Berberine outside PCOS: real numbers, poor absorption, and a trial base that isn't you.
Pooled trials show genuine drops in fasting glucose and LDL. Almost all of them recruited people already diagnosed with something — and 0.36% of an oral dose reached the blood in rats.
Ginkgo and ginseng have been tested for fifty years. The results are mostly flat.
We sell a ginkgo-and-ginseng capsule. Three JAMA trials, a Cochrane review and a meta-analysis of 1,132 people say it will not sharpen a healthy adult's memory.
Peppermint oil has the IBS trials. Most of the shelf does not.
Ten randomised trials in 1,030 people put peppermint oil ahead of placebo, with a number needed to treat of 4 and more side effects. Very little else in the aisle can show you that.
A single cortisol reading has to nearly triple before it means anything.
Twenty people spat into tubes six times a day for five days. The day-to-day swing was so wide that a second result must differ by 96 to 245 percent before it counts as a real change.
Most of what you do for sore muscles treats the feeling, not the damage.
Twenty-nine systematic reviews, 863 trials, and one uncomfortable finding: the cold plunge that makes the soreness better may also be taxing the adaptation you trained for.
The caffeine and l-theanine stack, and the part the caffeine is doing.
The famous 1:2 ratio comes from two industry trials in 2008. A meta-analysis found the effect tracked the caffeine dose, and a later study at tea-strength doses found theanine cancelled caffeine out.
Senolytics are a real idea being sold about a decade before the evidence.
Clearing senescent cells extended healthy life in mice and the biology is sound. The human trials are pilots of fourteen and nine people, and one of the two drugs is prescription chemotherapy.
CFU count is the least useful number on a probiotic bottle.
In one 362-person trial, 100 million CFU beat placebo and 10 billion CFU of the same organism did not. The strain designation is the number that carries the evidence.
Folate, folic acid and MTHFR: what is settled and what is being sold.
The trial that made folic acid a public health measure found a 72% reduction in neural tube defects. The gene test being sold alongside methylfolate has a professional guideline telling doctors not to order it.
The anabolic window is wide enough to walk through. The urgency was always the product.
A meta-regression of 23 hypertrophy studies found total daily protein, not timing, predicted the result. A 10-week trial then fed protein before or after training and found no difference.
Do you actually need a multivitamin? For most people, probably not.
Two of the largest datasets ever assembled on multivitamin use point in unglamorous directions. Here is what they found, and the short list of people the evidence genuinely supports.
Before you buy anything for energy, get your thyroid checked.
Nearly one in ten people at a Colorado health fair had an elevated TSH. It is a cheap test, symptoms are a poor substitute for it, and no supplement is a substitute for either.
Ten minutes beat thirty. Nap length decides what you get.
In the trial that compared five, ten, twenty and thirty-minute naps head to head, ten minutes won outright and thirty produced a stretch of impaired alertness first.
Collagen: the trials are real, the effects are modest, and we don't sell it.
Twenty-six randomised trials say oral collagen does something measurable to skin. The same review says half of them had missing outcome data. Both halves of that sentence matter.
Cordyceps and endurance: three trials, three different answers.
Trained cyclists got nothing out of five weeks of it. Older adults moved their lactate threshold without moving VO2 max. Here is how to read a genuinely mixed file.
Reishi has two Cochrane reviews, and neither one is about stress.
The mushroom sold for calm has been formally reviewed for cancer and for cardiovascular risk. Its reputation as a stress herb rests on almost nothing measured in a healthy, stressed adult.
Spermidine has beautiful mechanism data and one trial that came back flat.
Autophagy in yeast, flies and worms. A 20-year Austrian cohort where high intake matched being 5.7 years younger. And a 12-month randomised trial in 100 people that found nothing.
The cold plunge works. In a lifting block, that is the problem.
Pooling eight controlled studies, regular cold-water immersion had a deleterious effect on resistance training adaptations and no effect on endurance ones. It still makes tomorrow feel better.
A probiotic after antibiotics may delay your own microbiome coming back.
We sell probiotics. A 2018 Cell paper found that after a course of antibiotics they produced a markedly delayed and persistently incomplete return of a person's native gut bacteria.
Berberine, insulin sensitivity, and the PCOS research nobody mentions.
It gets sold as “nature's Ozempic,” which is both wrong and a shame — the actual trial data in women is more specific and more interesting than the slogan.
Citicoline has better trials than most of this shelf. Read who ran them.
Two positive randomised trials, a Cochrane review that separated memory from attention, and a 2,298-patient stroke trial stopped for futility. The pattern is consistent and so is the funding.
Green tea extract raises energy expenditure by about 100 calories. Caffeine alone did the same.
A meta-analysis of respiration-chamber studies put catechin-caffeine mixtures at 428 kJ a day. Caffeine on its own came in at 429. The Cochrane weight-loss result was minus 0.04 kg.
If you already eat enough protein, BCAAs have almost nothing left to do.
BCAAs alone raised muscle protein synthesis 22% in one trial. A complete protein raises it more, and its amino acids are already in your dinner. Here is who is actually left over.
Zinc has a narrow window between useful and counterproductive.
The dose that shortens a cold is roughly twice the daily upper limit, which is fine for four days and a problem for four months. The distinction is the whole article.
Iron helps when you are deficient and does harm when you are not.
Iron is the one supplement where guessing has a real downside, because the body has no route to excrete the surplus. Test first, then read this for how to take it.
Which magnesium should you take? An honest comparison of the forms.
Glycinate, citrate, oxide, threonate, malate. What the absorption data actually shows, where the marketing gets ahead of it, and why the form matters less than the dose you keep taking.
A nightcap sells you the first half of the night and bills you for the second.
At every dose studied, alcohol shortens sleep onset, deepens the first half, then fragments the back half. It also pushes the apnea index up by about four events an hour.
Why oral L-arginine mostly fails on the way in, and L-citrulline does not.
Your gut destroys much of the arginine you swallow before it reaches your blood. Citrulline takes the back door. Here is the pharmacokinetic data, and what it is honestly worth in a set.
The taurine longevity study was in mice. Two follow-ups say the premise was wrong.
A 2023 Science paper made taurine a longevity category overnight. Two 2025 papers could not find the human decline the whole hypothesis rests on. Species is the entire story here.
Nobody has run the deload trial you want. The tapering data is the next best thing.
Surveyed strength athletes cut both volume and intensity in a deload. A meta-analysis of 27 tapering studies found the effective move is to cut volume by 41 to 60% and leave intensity alone.
Low-FODMAP is a diagnostic protocol, not a diet you stay on.
Restriction is stage one of three. In one controlled trial it cut faecal Actinobacteria and butyrate, and the researchers concluded that strict long-term use should not be advised.
Perimenopause brain fog is real, documented, and usually temporary.
The word that won't come. The reason you walked into the room. The most useful thing anyone can give you is the correct name for it — and the timeline.
L-theanine and caffeine have real data. It is data about attention, not calm.
Two recent meta-analyses land on the same narrow finding: the pairing sharpens attention for an hour or two. The stress effect is smaller, shakier, and driven by trials at high risk of bias.
Chromium moves glucose in people who already had a problem. That is the whole finding.
Two systematic reviews reached the same split verdict: a measurable effect in people with type 2 diabetes, and nothing at all in people without it. Nobody can reliably measure your chromium status either.
HMB has been sold on strong claims since the nineties. The trials have not kept up.
A meta-analysis of 394 lifters found HMB's strength effect trivial in trained men and small in untrained ones. The clearest signal sits in a group that is not chasing a personal best.
B12 deficiency is common, cheap to test for, and routinely missed.
In the paper that changed how it is diagnosed, 28% of patients with neurological damage from B12 deficiency had entirely normal blood counts. A normal full blood count does not rule it out.
Bacopa runs on a twelve-week clock, or it does not run at all.
Every trial in the main review dosed for twelve weeks. Nine of seventeen memory tests improved, a 72-person trial found nothing at all, and the most common side effect is an upset stomach.
Vitamin D dosing: why more stopped being the answer.
Three large trials tested the assumption that a bigger dose is a better one. A 500,000 IU annual dose raised fractures, 10,000 IU a day lowered bone density, and 2,000 IU a day changed nothing in healthy adults.
Blue-light glasses: the Cochrane review could not pool a single outcome.
Seventeen randomised trials, sample sizes from five to 156 people, and findings so inconsistent that no meta-analysis was possible. Here is what the individual trials did find.
Muscle is not about how you look. It is about who dresses you at 85.
Low thigh muscle nearly doubled the risk of losing mobility in 3,075 older adults. But the honest reading is that quality beats quantity, and only one intervention reliably builds both.
The cool-down you do after training is not the active recovery with evidence.
One review concludes active cool-downs are largely ineffective. Another finds active recovery reduces soreness. They are not contradicting each other — they are studying two different things.
For constipation, the type of fibre matters more than the amount.
Across seven trials, 77% responded to fibre against 44% on placebo. But prunes beat psyllium head to head, and in a three-way comparison kiwifruit caused the fewest side effects.
Cinnamon lowered fasting glucose by 24 mg/dL in one review and did nothing in another.
Two systematic reviews, published a year apart, read the same small trials and disagreed. The measure that integrates months of glucose did not move in either of them.
Being tired is not always about being busy. Ask for a ferritin test.
Iron deficiency without anaemia is common in menstruating women, and a standard blood count will not find it. This is the article where we tell you to buy nothing and go get tested.
Sleep is a training variable, and it is the only one you can change tonight.
Across 69 studies and 227 performance measures — 89% of them in men — sleep loss cost an average of 7.6% of physical performance. The pattern of the loss matters more than the total.
Every published tulsi trial found a benefit. That is the problem.
A systematic review located 24 human studies of holy basil and 24 favourable outcomes. A literature with no negative results in it is telling you something about publishing before it tells you anything about the plant.
What a certificate of analysis actually proves, and what it does not.
A COA is a photograph of one batch on one day, taken by a lab someone paid. That is genuinely useful and much narrower than the way it gets marketed.
Beta-alanine works in a window of about one to four minutes. Most sets are not in it.
The tingle is not the mechanism. Two meta-analyses put the effect at its clearest in efforts lasting 60 to 240 seconds and at nothing measurable under a minute.
A 2pm coffee is still working at 8pm for most people.
Plasma half-lives ranged from 2.7 to 9.9 hours across ten healthy men in one clean crossover. That spread is why your afternoon espresso is harmless and your colleague's is not.
Why bloating hits women harder — and it isn't a discipline problem.
If your jeans fit in the morning and don't by evening, the epidemiology is on your side. IBS affects roughly 14–24% of women versus 5–15% of men.
VO2 max is the best mortality predictor you can actually train.
Across 750,302 veterans, the least fit had four times the death rate of the fittest. Here is what the number is, how to get yours without a lab, and how far it moves with work.
Nootropic is a marketing word. Here is what was actually in the bottles.
The term was coined in 1972 for a narrow pharmacological idea. Today it has no legal definition, and when chemists tested ten memory supplements they found five unapproved drugs.
Sleep hygiene is a weak treatment and a decent habit. Both are true.
Sleep hygiene education moved sleep efficiency about 5% and lost to CBT-I by a medium-to-large margin. The individual pieces hold up better than the bundle does.
Bloating has a differential diagnosis, and most of it is not a supplement.
In a US telephone survey of 2,510 adults, 15.9% reported bloating or distension in the previous month. Here is the list a gastroenterologist works through before anyone reaches for a capsule.
Your HRV is a trend line, not a verdict on your morning.
A meta-analysis of 24 studies found resting heart rate variability was largely unaffected by overreaching. The number is still useful. It is just not the number most people think they are reading.
How to read a supplement label, including ours.
Proprietary blends, elemental weight, and the serving size that quietly doubles the price. Five things on the back of a bottle that decide whether you bought a dose or a decoration.
Three hundred and fifteen weight-loss supplement trials, and sixteen results.
A 2021 systematic review sorted every randomised trial of fourteen weight-loss ingredients. Fifty-two were well enough run to judge. Sixteen of those found a difference, from 0.3 kg to 4.93 kg.
Do you need supplements at all? Most people need far fewer than they are sold.
We sell supplements, so read this with that in mind. There is a short list of people the evidence genuinely supports, and a long list of people who are buying reassurance.
Rhodiola's most glowing results come from the trials with no placebo group.
Two small controlled studies found something narrow and real. The enthusiastic burnout numbers everyone quotes come from single-arm trials where every participant knew exactly what they were taking.
The 3pm dip has four common causes, and only one of them is caffeine.
It shows up on schedule whether or not you ate lunch, which is the first clue that it is not really about lunch. Here is what is actually driving it, in the order we would check.
Creatine loading is a shortcut, not a requirement. Muscle ends up in the same place.
Six days at 20 g and 28 days at 3 g both raised muscle total creatine by about 20% in the same 1996 study. What loading buys you is three weeks, and it charges for them.
Magnesium and sleep: what the evidence supports, and what it doesn't.
Seventeen minutes faster to sleep, and a total sleep time result that wasn't statistically significant. This is the article where we argue against our own marketing department.
Why the ashwagandha dose on the label matters more than the brand on the front.
The trial everyone cites used 600mg a day. A 150mg capsule is not a smaller version of that study — it is a different intervention that nobody has tested.
Creatine isn't just for men. The female evidence is stronger than the marketing.
Women store 70–80% less creatine than men to begin with. That single fact reframes most of what the sports-nutrition aisle has told you for twenty years.
No articles in that combination yet. Try another filter.
