Calcium is the raw material of bone, and supplementing it barely changes fracture risk. Tension builds muscle, and a sixteen-fold change in tempo makes no difference. The reason is the same in both cases, and once you see it you cannot unsee it.
There is an assumption sitting underneath most health advice, and it is so intuitive that almost nobody states it out loud.
The assumption is that if something is involved in a process, more of it will make the process go better. Calcium is in bone, so more calcium means stronger bone. Tension builds muscle, so more time under tension means more muscle. Turmeric is anti-inflammatory, so more turmeric means less inflammation.
It is a reasonable way to think. It is also wrong most of the time, and the reason it is wrong is worth understanding once properly, because it explains a great many disappointing results in one go.
Any process built from multiple steps runs at the speed of its slowest step. Adding more of anything that is not the slowest step changes nothing at all.
Chemists call this the rate-limiting step. Manufacturers call it the bottleneck. Whatever you call it, the practical consequence is the same and it is deeply counterintuitive: the amount of an ingredient tells you nothing about whether adding more of it will help. What matters is whether that ingredient is the thing currently holding the process back.
Think of a production line with six stations. Station four takes twice as long as the others. You can pile up as much raw material at station one as you like — nothing comes out of the end any faster. All you have done is create a bigger pile at station one.
Most supplement advice, and a fair amount of training advice, is the pile at station one.
Four examples, each of which surprised me when I first worked through the evidence.
Bone is roughly 60% mineral by weight, and most of that mineral is calcium phosphate. Calcium is unambiguously the raw material. So more calcium ought to mean more bone.
In 2015 the BMJ published two systematic reviews from the same research group in Auckland, on the same day. The first, by Tai and colleagues, pooled 59 randomised trials of dietary calcium or calcium supplements in people over 50. It found increases in bone density of roughly one to two percent — real, measurable, and too small to translate into meaningfully fewer broken bones.
The second, by Bolland and colleagues, went straight at fractures. It found no association between dietary calcium intake and fracture risk, and described the benefits of calcium supplements on fracture prevention as meagre and inconsistent.
Calcium is the substrate. Calcium is not the limiter.
The limiter is mechanical load. Bone is a living tissue that responds to strain by remodelling itself. It adds material where it is repeatedly loaded and removes it where it is not — which is why astronauts lose bone in orbit while eating a perfectly adequate diet, and why the strongest predictor of bone density in older adults is not what they eat but what they do.
Put the demand in place and the body will source the calcium. Supply calcium to a skeleton that is not being asked for anything and it goes to the pile at station one.
The gym version of the same error. Muscles grow in response to mechanical tension, and time under tension is how long the muscle spends loaded during a set. So slower repetitions mean more time under tension, which ought to mean more muscle.
Schoenfeld, Ogborn and Krieger tested this in Sports Medicine in 2015, pooling eight studies that directly compared different repetition durations with volume and intensity matched. Repetition durations anywhere from half a second to eight seconds produced similar increases in muscle size.
Sixteen-fold variation in tempo. No meaningful difference in outcome.
They did find one edge: deliberately very slow repetitions, beyond about ten seconds each, appeared inferior — though on limited evidence. And the mechanism for that is telling. To move a weight that slowly you have to make it lighter, and making it lighter reduces the mechanical tension that was the point of the exercise.
The limiter is effort against a meaningful load. The clock is not the limiter, and manipulating it while everything else stays the same achieves nothing. Slowing down helps only when it makes a set genuinely harder, and hurts when it forces the weight down.
This one is different in kind, and it is the most commonly misread finding in the whole of health research.
Grip strength is a remarkably good predictor of death. The PURE study, published in The Lancet in 2015, followed nearly 140,000 people across 17 countries and found grip strength predicted all-cause mortality more strongly than systolic blood pressure did. The finding has been replicated many times.
From which people conclude, entirely understandably, that they should train their grip.
But grip strength is not doing anything to keep you alive. It is a marker, not a lever. It happens to be an unusually convenient readout of total muscle mass, nutritional status, neurological function, physical activity history and general physiological reserve — everything that actually determines whether an older body copes with illness. A dynamometer measures all of that in about four seconds for the price of a cheap instrument.
Squeezing a gripper builds forearm strength. It does not build the thing the grip test was reflecting.
The distinction between a marker and a lever is the most useful single idea here, and it applies well beyond grip. Plenty of health advice consists of chasing a number that was only ever a shadow of the thing that matters.
Curcumin has genuine biological activity. It is the compound behind almost every claim made for turmeric.
But turmeric root is only a few percent curcuminoids by weight, and curcumin on its own is very poorly absorbed — it dissolves badly in water, is metabolised quickly by the gut wall and liver, and is cleared fast. Swallow curcumin powder and remarkably little reaches your bloodstream.
Which means the entire question is one of delivery. Piperine from black pepper substantially slows curcumin’s breakdown. Phospholipid complexes, nanoparticle formulations and oil-based preparations all address the same bottleneck from different angles.
The limiter is absorption, not dose. A larger scoop of turmeric mostly produces more expensive stool. This is the one example on the list where the answer is not to abandon the intervention but to fix the actual bottleneck — and it is why two curcumin products with identical milligram counts on the label can differ enormously in whether they do anything at all.
If you take one practical thing from this article, make it that: for curcumin, the delivery system is the intervention. The number on the front of the bottle is close to meaningless without it.
Across all four, the same question does the work:
What is currently the slowest step — and is the thing I am about to add actually it?
For bone, the slowest step is usually load, not calcium. For muscle, it is effort, not tempo. For longevity, grip was never a step at all. For curcumin, it is absorption, not quantity.
I want to be careful here, because this reasoning can be pushed too far and become its own kind of nonsense.
If you are genuinely deficient in something, that something is the rate-limiting step, and supplying it works dramatically. Iron in iron-deficiency anaemia. Vitamin D in someone deficient. B12 in someone depleted. Thiamine in alcohol dependency. In those situations the substrate is the bottleneck, correcting it produces changes that look almost miraculous, and anybody telling you nutrients do not matter is as wrong as the person selling you a fifth bottle.
The point is not that inputs never matter. It is that inputs matter when they are the limiting step and not otherwise — and that whether they are is a question with an answer, which can often be measured, rather than something to assume from the fact that the ingredient appears somewhere in the process.
That is most of what testing is for. Not to generate a longer list of things to take, but to find out which step is actually slow.
Because in every one of the four examples above, the honest answer to "what should I take?" was a question about something else entirely.
This article is general information and not medical advice. If you have a diagnosed condition or take prescribed medication, discuss changes with your GP or specialist.
Sources
Tai V, Leung W, Grey A, Reid IR, Bolland MJ. Calcium intake and bone mineral density: systematic review and meta-analysis. BMJ 2015;351:h4183.
Bolland MJ, Leung W, Tai V, et al. Calcium intake and risk of fracture: systematic review. BMJ 2015;351:h4580.
Reid IR, Bristow SM, Bolland MJ. Calcium supplements: benefits and risks. Journal of Internal Medicine 2015;278(4):354-368.
Schoenfeld BJ, Ogborn DI, Krieger JW. Effect of repetition duration during resistance training on muscle hypertrophy: a systematic review and meta-analysis. Sports Medicine 2015;45(4):577-585.
Leong DP, Teo KK, Rangarajan S, et al. Prognostic value of grip strength: findings from the Prospective Urban Rural Epidemiology (PURE) study. The Lancet 2015;386(9990):266-273.
That is a question with an answer, and usually a measurable one. It is most of what testing is for.
Ask AIdan → See the TDG Programme