Name: Bone Health: Load It, Feed It Protein, and the Calcium Story Is Overrated
Creator: Realised
Published: 2026-06-01

Summary

The levers that actually build and preserve bone are, in order: load the skeleton (progressive resistance and impact training is the #1 modifiable lever, not an optional extra), eat enough protein (it HELPS bone, the "protein leaches calcium" claim is debunked), get calcium preferentially from food and correct vitamin D only if you're low; the popular "drink milk for strong bones" slogan oversimplifies, isolated high-dose calcium SUPPLEMENTS show weak fracture benefit with a possible cardiovascular-risk signal (food beats pills), and bisphosphonate drugs genuinely cut fractures in genuinely hi

Why Strong

Strong Evidence because the load-bearing claims rest on genuinely strong anchors: an RCT of loading in the exact target population (LIFTMOR), a debunk of the protein-harms-bone myth that is consistent across independent meta-analyses, and large registry-plus-RCT fracture-reduction data for bisphosphonates with a well-characterised harm/duration relationship. The core directional claims (load first, protein helps, food-calcium over pills, drugs work in high-risk people) are well supported.

NOT unconditionally Strong on every sub-claim. The loading evidence is bone-density- and function-based rather than fracture-endpoint-based (fracture RCTs for high-intensity loading are still limited); the protein-benefit MAGNITUDE is cohort-dominated; and the calcium-supplement cardiovascular signal is contested. The entry is tiered Strong on the strength of the whole, with those soft spots named explicitly.

NOT Emerging because this is not a handful of suggestive studies: loading is RCT-backed, the protein debunk is meta-analytic and replicated, and the drug fracture data is among the best-characterised in the field. K2 is the sole genuinely emerging lever, and it is scoped as supporting/optional.

The per-lever split (read this, not just the headline):
• Loading: Strong for bone density/function via an RCT in the target population; fracture-endpoint data still limited.
• Protein: Strong for the debunk of harm; Moderate for the benefit magnitude (cohort-heavy).
• Calcium supplements: Moderate and contested — weak fracture benefit, possible CV signal, food the safer default.
• Bisphosphonates: Strong for fracture reduction and the benefit-dominates-in-high-risk conclusion.
• Vitamin K2: Emerging — positive but heterogeneous, fracture endpoint unclear.

Practical takeaway

The framing to hold: bone strength is built by loading the skeleton and supplying it with protein and mineral substrate, in that order of leverage. Load first, feed protein, get calcium from food, correct vitamin D only if low. Reserve calcium pills and bone drugs for people whose risk actually justifies them.

Lever one — load the skeleton (the highest-yield move).
• Progressive resistance training plus impact/weight-bearing loading is the primary lever. The trial-supported model is supervised, heavy (compound lifts progressing toward high intensity) with an impact component, not light circuits. Mechanics are owned by resistance_training_and_body_composition.
• If you are older, frail, or have had a fragility/vertebral fracture, get screened and supervised before high-intensity loading. LIFTMOR's safety came WITH screening and supervision, not without.

Lever two — eat enough protein (it helps bone).
• Adequate-to-higher protein supports bone density and lower fracture risk when calcium is adequate. Do NOT restrict protein "to protect your bones"; that is the debunked myth. Intake specifics are owned by diet_protein_intake.

Lever three — calcium from food, not pills.
• Aim to get calcium preferentially from food (dairy, leafy greens, fortified foods, tinned fish with bones). Food calcium is the safer default and shows no cardiovascular signal.
• Do NOT reach for isolated high-dose calcium supplements as a bone strategy; the fracture benefit is weak and there is a contested cardiovascular signal. Supplement calcium only to close a genuine dietary gap under clinical guidance.
• For the practical calcium detail (which form, the ~500 mg per-dose absorption ceiling, the oxalate/spinach trap, best food sources, and eggshell powder), see calcium_food_vs_supplement_and_absorption.

Lever four — correct vitamin D only if you're low; K2 is optional.
• Vitamin D supports calcium handling and bone; the lever is REPLETION if deficient, not blanket high-dosing. Dosing is owned by vitamin_d3_high_dose_supplementation.
• Vitamin K2 is experimental/supporting at best; do not treat it as a primary lever. Cofactor logic is owned by vitamin_d_k2_magnesium_cofactor_stack.

Where drugs fit (clinician territory).
• Bisphosphonates are for genuinely high-risk people (established osteoporosis, prior fragility fracture), where the fracture math clearly favours treatment. This is a prescribing decision, not self-management.
• After several years of treatment, a drug holiday may be appropriate in lower-risk patients to limit the rare duration-dependent harms. Timing is individualised clinical territory, decided with the prescriber, not from a slogan.

Evidence detail

Why This Entry Exists

Ask most people how to keep their bones strong and you get one answer: calcium, usually via milk or a supplement. That instinct is not wrong so much as badly ordered and half-mythical. The best-supported lever is the one nobody markets, because you cannot bottle it: mechanically loading the skeleton with progressive resistance and impact. Bone is a living, adaptive tissue that lays down mineral where it is stressed and gives it up where it is not, so the signal that builds bone is force, not intake. Meanwhile two widely held beliefs run the wrong way. The first is that calcium PILLS are the fix; the fracture benefit of isolated high-dose calcium supplements is weak, and there is a contested cardiovascular-risk signal, whereas food calcium carries no such flag. The second is the old "acid-ash" idea that dietary protein leaches calcium out of bone; that has been debunked, and higher protein is actually associated with better bone and fewer fractures.

The honest read is not "supplements are useless and drugs are bad," though. Bisphosphonates genuinely reduce fractures in high-risk patients, by up to about half at the spine and hip, and the fracture-prevention math clearly favours treatment there. What complicates the drug story is a rare, real, duration-dependent set of harms (atypical femoral fracture, jaw osteonecrosis) that motivate a planned "drug holiday" after several years in lower-risk patients. So the whole topic is a study in getting the ORDER right: load first, feed protein, food-calcium and vitamin-D repletion as substrate, and reserve the pills and the drugs for the people whose fracture risk actually justifies them.

What bad advice this protects against, in all directions:
• "Calcium is the answer, take a supplement" → overstated. Isolated high-dose calcium supplements show weak fracture benefit and a possible cardiovascular signal (Bolland); food calcium is the safer default and loading matters far more.
• "Protein is bad for bone, it leaches calcium" → debunked. The acid-ash hypothesis was not supported; higher protein is associated with ~11% lower hip-fracture risk and higher BMD when calcium is adequate.
• "Drink milk for strong bones and you're covered" → too narrow. Dietary calcium alone shows no consistent fracture-prevention signal, and no amount of milk substitutes for loading the skeleton.
• "Bisphosphonates cause atypical fractures, avoid them" → backwards for high-risk people. In osteoporosis the fracture PREVENTED vastly outnumbers the atypical fracture caused (5-year NNT ~56 vs NNH ~1424, Abrahamsen 2024); the harm scales with duration, which is why holidays exist, not why you skip the drug.
• "Weightlifting is dangerous for fragile bones, just walk" → mostly wrong but with a real caveat. Supervised high-intensity resistance and impact training raised bone density safely in osteoporotic postmenopausal women (LIFTMOR), but the trials were SCREENED and SUPERVISED; frail or vertebral-fracture-prone individuals need screening first.

This entry owns the real levers (loading, protein, food-calcium, vitamin-D-if-low, optional K2), the honest read on bisphosphonates, the calcium-supplement caveat, and the protein-myth debunk. It defers vitamin-D dosing to vitamin_d3_high_dose_supplementation, the calcium/K2 cofactor logic to vitamin_d_k2_magnesium_cofactor_stack, and resistance-training mechanics to resistance_training_and_body_composition. It states those boundaries and routes there rather than re-arguing them.

Evidence

Organised by lever, with the tier signal inline. The headline rests on an RCT in the target population, converging cohort-and-meta evidence for protein, contested-but-directional supplement data, and large registry-plus-RCT fracture data for the drugs. Read the per-finding signals, not just the thesis.

Loading — the #1 modifiable lever, RCT-proven in the target population (Strong Evidence).

1. Supervised high-intensity resistance plus impact training raised bone density and function in osteoporotic postmenopausal women, safely. In the LIFTMOR randomised controlled trial, 8 months of twice-weekly supervised high-intensity resistance and impact training (a 5x5 protocol at greater than 85% of one-rep-max, plus jumping-style impact) significantly improved lumbar-spine and femoral-neck bone mineral density and functional performance in postmenopausal women with low bone mass (osteopenia/osteoporosis), with no serious injuries (one minor adverse event was reported: mild low-back pain in a participant with pre-existing spondylolisthesis). This is the finding that overturned the "fragile bones can't lift" default and established loading as the primary modifiable lever. (Watson SL, Weeks BK, Weis L, Harding AT, Horan SA, Beck BR. High-Intensity Resistance and Impact Training Improves Bone Mineral Density and Physical Function in Postmenopausal Women With Osteopenia and Osteoporosis: The LIFTMOR Randomized Controlled Trial. J Bone Miner Res. 2018;33(2):211-220. Strong — RCT in the exact target population, replicated in follow-up work (MEDEX-OP), but supervised and screened, so the safety result should not be read as licence for unsupervised loading of frail individuals.)

Protein — helps bone; the "leaches calcium" claim is debunked (Strong for the debunk, Moderate for the benefit magnitude).

2. Higher protein is associated with fewer hip fractures and better bone, and the acid-ash "protein leaches calcium" hypothesis was NOT supported. Groenendijk et al.'s pooled meta-analysis found higher dietary protein associated with a lower hip-fracture risk (hazard ratio roughly 0.89), and the Wallace and Frankenfeld review put the reduction in the region of 11% (an 11 to 16% range across intakes); higher-protein intakes also associate with higher bone mineral density and lower bone-loss rate when calcium is adequate. The old acid-ash model, that dietary protein acidifies the body and pulls calcium out of bone, was directly examined and not supported. Protein helps bone; it does not harm it. (Wallace TC, Frankenfeld CL. Dietary Protein Intake above the Current RDA and Bone Health: A Systematic Review and Meta-Analysis. J Am Coll Nutr. 2017;36(6):481-496; Groenendijk I, et al. High versus low dietary protein intake and bone health in older adults: a systematic review and meta-analysis (PMC6704341); National Osteoporosis Foundation position paper, Am J Clin Nutr 2017. Strong for the debunk of harm; Moderate for the magnitude of benefit — the fracture signal is cohort-dominated, and protein-supplement RCTs show lumbar-spine BMD gains but a weaker fracture endpoint.)

The calcium-supplement caveat — the part the "drink milk" slogan hides (Moderate Evidence, contested).

3. Isolated calcium supplements carry a weak fracture benefit and a possible cardiovascular signal; food calcium does not. A meta-analysis of RCTs found calcium supplements associated with roughly a 30% relative increase in myocardial infarction (hazard ratio 1.31, 95% CI 1.02 to 1.67), with no such signal from dietary calcium. Separately, calcium-supplement fracture benefit (relative risk about 0.89) disappeared in the trials at lowest risk of bias, and dietary calcium showed no fracture-prevention association at all. The direction is consistent: food beats isolated high-dose pills. (Bolland MJ, Avenell A, Baron JA, et al. Effect of calcium supplements on risk of myocardial infarction and cardiovascular events: meta-analysis. BMJ. 2010;341:c3691; Bolland MJ, Leung W, Tai V, et al. Calcium intake and risk of fracture: systematic review. BMJ. 2015;351:h4580. Moderate — the CV signal is genuinely contested (later analyses, e.g. Ann Intern Med. 2016, found calcium intake within tolerable limits not clearly cardiotoxic), while the weak-fracture-benefit finding is the sturdier of the two. Directionally: food > isolated high-dose supplements.)

Bisphosphonates — genuine fracture reduction in high-risk people, with rare duration-dependent harm (Strong Evidence).

4. In osteoporosis, bisphosphonates cut fractures up to about 50%, and the benefit vastly outweighs the rare atypical-fracture harm in high-risk patients. Bisphosphonates reduce vertebral, hip and nonvertebral fracture risk by up to roughly half in osteoporosis; the risk-benefit strongly favours treatment in high-risk patients. In a Danish case-cohort analysis quantifying both sides, the 5-year number-needed-to-treat to prevent one hip fracture was about 56, against a 5-year number-needed-to-harm of about 1424 for one atypical femoral fracture. Atypical femoral fractures and jaw osteonecrosis are real but rare, and their risk rises with treatment duration (atypical-fracture rate roughly 1.74 per 10,000 patient-years overall, rising to about 6.04 at 5 to under 8 years of use and about 13.10 at 8 or more years), which is why a 2-to-3-year drug holiday is advised after prolonged use in lower-risk patients. (Black DM, Geiger EJ, Eastell R, et al. Atypical Femur Fracture Risk versus Fragility Fracture Prevention with Bisphosphonates. N Engl J Med. 2020;383(8):743-753 (duration-dependent atypical-fracture rates); Abrahamsen B, et al. J Clin Endocrinol Metab. 2024;109(11):e2141 (5-year NNT ~56 / NNH ~1424); Schilcher J, Michaëlsson K, Aspenberg P. Bisphosphonate use and atypical fractures of the femoral shaft. N Engl J Med. 2011;364(18):1728-1737; drug-holiday review, PMC9189912. Strong — large registry plus RCT-derived fracture-reduction data; the harm/duration relationship and holiday rationale are well established.)

Vitamin K2 — supporting, not primary; the weakest lever (Emerging Evidence).

5. Vitamin K2 (MK-7) shows a positive but inconsistent effect on bone. A 3-year RCT of 375 micrograms/day MK-7 in 142 osteopenic postmenopausal women slowed vertebral height loss and improved some bone density parameters, while meta-analyses report conflicting bone-density and fracture outcomes. K2 is a supporting cofactor at best, not a primary lever, and its fracture endpoint is unclear. (Rønn SH, et al. Vitamin MK-7 supplementation and bone mineral density and microarchitecture: a 3-year randomized, placebo-controlled trial (PubMed 33030563); systematic reviews PMC9403798 and Front Public Health 2022. Emerging — RCTs are positive but heterogeneous, fracture endpoints unclear. Cofactor logic deferred to vitamin_d_k2_magnesium_cofactor_stack.)

Mechanism

Why loading is the master lever. Bone is not inert scaffolding; it is a tissue that continuously remodels in response to the mechanical strain placed on it (Wolff's law, operationalised through the mechanostat model). Osteocytes sense strain and orchestrate the balance between bone-building osteoblasts and bone-resorbing osteoclasts. High-magnitude, novel loading, the kind delivered by heavy resistance training and impact, produces the strain signal that tips remodelling toward net formation, especially at the loaded sites (spine and hip in a squat/deadlift/impact protocol). This is why a drug or a supplement can supply mineral substrate but cannot, by itself, tell the skeleton WHERE to put it: the localisation signal is force. It is also why the best lever is the least monetisable one.

Why protein helps rather than harms. Roughly half of bone by volume is protein (mostly type-I collagen matrix), and that matrix is what mineral crystallises onto, so protein is a structural building block, not a threat. Higher protein intake also raises IGF-1 and supports the muscle mass that loads bone in the first place, coupling the protein and loading levers. The discarded acid-ash hypothesis held that metabolising protein generates acid that the body neutralises by leaching alkaline calcium salts from bone; measured across trials, the compensatory rise in calcium absorption offsets any urinary calcium increase, and net bone balance is neutral-to-positive. When calcium intake is adequate, higher protein tracks with better bone, which is the opposite of the myth.

Why food calcium and supplemental calcium diverge. Calcium is the mineral substrate for bone, but the body regulates serum calcium tightly regardless of intake. Food delivers calcium slowly, in modest amounts, alongside other nutrients, and dietary calcium carries no cardiovascular signal. A single large-dose supplement produces a sharp, supraphysiological rise in blood calcium; the leading (contested) mechanistic hypothesis for the cardiovascular signal is that these transient spikes may promote vascular calcification over time. Whether or not that mechanism holds, the empirical pattern (food shows no signal, isolated high-dose pills show a possible one, and the pill's fracture benefit is weak) is enough to default to food.

Why bisphosphonate benefit and harm both scale with the drug's action. Bisphosphonates bind bone mineral and suppress osteoclast-driven resorption, which lowers bone turnover and lets density accrue, reducing fractures. The same durable turnover suppression is the mechanistic basis of the rare harms: over many years, profoundly low remodelling can leave micro-damage unrepaired in the femoral shaft (atypical fracture) or the jaw (osteonecrosis, especially after dental trauma). Benefit dominates early and in high-risk bone; harm accumulates slowly with duration. That mechanistic asymmetry is exactly why the rational move is time-limited treatment with a holiday, not avoidance.

Risks And Contraindications

• This entry is not medical advice, and NOT a reason to start or stop a prescribed bone drug. Osteoporosis-drug decisions (starting, stopping, holidays) belong with the prescribing clinician.
• Bisphosphonate benefit-dominance holds specifically in HIGH-RISK patients (osteoporotic, prior fragility fracture), not in low-risk osteopenia. Do NOT generalise the 5-year NNT of ~56 to everyone; in lower-risk people the balance is different, and that is precisely where holidays and non-drug levers come first.
• The calcium-supplement cardiovascular signal is genuinely contested. Frame it as "possible signal, food is the safer default," NOT "calcium pills cause heart attacks." Later analyses softened the claim; the sturdier finding is the weak fracture benefit.
• High-intensity impact and resistance training was supervised and screened. An unsupervised recommendation to load frail individuals, or those with prior vertebral fracture, could cause harm. Flag supervision and screening; the safety result does not transfer to unscreened self-loading.
• The protein benefit is cohort-heavy for fractures. State the debunk of HARM confidently (that is robust); claim the magnitude of BENEFIT more tentatively.
• Red-flag boundary. Sudden severe back pain (possible vertebral compression fracture), a fall resulting in inability to bear weight or a deformed/rotated limb (possible hip fracture), or new thigh/groin pain in someone on long-term bisphosphonates (possible impending atypical femoral fracture) warrant urgent medical assessment, not self-management.

Controversy

Nature: a small set of genuinely effective levers (loading, protein, food-calcium, vitamin-D-if-low) plus a genuinely effective drug class (bisphosphonates in high-risk people) sitting against a folk-wisdom default ("calcium/milk builds bone; protein is bad for bone; lifting is dangerous for fragile bones") that is variously oversimplified, backwards, or over-cautious, with the added twist that the food-calcium and supplemental-calcium evidence point in DIFFERENT directions.

Position A — "The genuine levers build and preserve bone, and the drugs work in high-risk people." The grounded take. Progressive resistance and impact loading is the #1 modifiable lever (LIFTMOR: 8 months of supervised high-intensity training raised spine and hip density in osteoporotic postmenopausal women, no serious injuries — one minor adverse event, mild low-back pain in a participant with pre-existing spondylolisthesis). Adequate protein helps bone (roughly 11% lower hip-fracture risk with higher intake), and the acid-ash "protein leaches calcium" claim is debunked. Food calcium and vitamin-D repletion supply the substrate. Bisphosphonates genuinely cut fractures by up to half in high-risk people (5-year NNT ~56 for one hip fracture). Where it goes wrong if overstated: reading LIFTMOR's safety as licence to load frail unscreened people, and generalising the drug NNT beyond high-risk patients.

Position B — "The pill story is oversold, and the drug harms are real." The corrective take. Dietary calcium shows no consistent fracture-prevention signal, and isolated calcium SUPPLEMENTS' fracture benefit collapses in the lowest-risk-of-bias RCTs while carrying a possible ~30% relative myocardial-infarction signal (Bolland). Vitamin K2 evidence is mixed and experimental. Bisphosphonates carry real, if rare, duration-dependent harms (atypical femoral fracture, jaw osteonecrosis), which is why drug holidays exist. Food beats pills. Its own nuance: even granting the harms, bisphosphonate benefit still dwarfs harm in high-risk patients (NNH ~1424 for one atypical fracture vs NNT ~56 for one hip fracture prevented, Abrahamsen 2024), and the calcium CV signal is contested, not settled. Where it goes wrong if overstated: sliding into "calcium doesn't matter" or "avoid bone drugs," both of which harm the people the drugs help most.

The funding/bias dimension — cui bono, both ways. Toward what sells: much positive calcium-supplement and MK-7 K2 literature is manufacturer-adjacent (NattoPharma/Gnosis behind branded MK-7; dairy councils behind "drink milk"), and the foundational bisphosphonate efficacy trials (FIT, HORIZON) were manufacturer-funded (Merck, Novartis). Toward the skeptic side: the Bolland/Reid Auckland group that produced the weak-fracture-benefit and CV-signal findings is independently funded and openly adversarial to the supplement industry, so their results are not pro-drug spin, but neither are they settled (later independent analyses softened the CV claim). The un-monetisable levers, loading and food, have essentially no sponsor pushing them, which is itself a tell.

Realised Position: Get the ORDER right. Load the skeleton first, because force is the signal that builds bone and it is the best-supported lever. Eat enough protein; it helps bone and the leaching myth is dead. Take calcium preferentially from food and correct vitamin D only if low. Reserve isolated calcium supplements and bisphosphonates for people whose risk genuinely justifies them, where the fracture math clearly favours the intervention; do not chase calcium pills or "more milk" as a strategy, and do not avoid a bone drug that would prevent a hip fracture out of fear of a much rarer atypical one. K2 is optional. The cleanest tell of honesty here is that the two best-supported levers, loading and food, are the two nobody can sell you.

Cross-Pillar Connections

This is a genuinely cross-pillar topic: bone is built by movement, fed by diet, and mineralised via supplements, and the bias read spans method.
• Movement / strength (resistance_training_and_body_composition): owns the mechanics of progressive resistance training; this entry holds only the stone-cold bone-specific claim that loading is the #1 modifiable bone lever and that it needs screening/supervision in fragile populations.
• Diet (diet_protein_intake): owns protein-intake specifics; this entry holds only the bone-relevant finding that protein HELPS bone and the acid-ash "leaching" claim is debunked.
• Supplements (vitamin_d3_high_dose_supplementation): owns vitamin-D dosing; this entry holds only the "correct D if low, don't blanket high-dose for bone" framing.
• Supplements (vitamin_d_k2_magnesium_cofactor_stack): owns the calcium/K2/magnesium cofactor logic; this entry holds only the bone-relevant food-calcium-over-pills split and the "K2 is optional" scoping.
• Movement / longevity (sarcopenia_prevention): the muscle-loss counterpart; muscle and bone decline together with age and respond to the same loading-plus-protein levers (the "muscle loads bone" coupling).
• Movement / strength (grip_strength_loaded_carries): loaded carries are a practical, low-tech way to deliver the axial and impact loading that bone responds to, and grip/strength track skeletal robustness.

What would change our mind

Falsifiability: explicit upgrade/downgrade criteria from source

• We'd rehabilitate the calcium pill and drop "food beats supplements" if a large, low-risk-of-bias RCT showed isolated high-dose calcium supplements meaningfully reduce fractures WITHOUT a cardiovascular signal.
• We'd escalate the calcium caveat from a soft "possible signal" to a warning if a well-powered independent RCT confirmed a hard cardiovascular endpoint from calcium supplements.
• We'd revisit the protein-helps-bone position if head-to-head trials showed dietary protein RESTRICTION improved bone outcomes. No current evidence points there.
• We'd downgrade the loading claim from surrogate-strong to uncertain if a large trial showed high-intensity resistance/impact training fails to reduce actual FRACTURES (not just bone density and function) in a well-powered sample. This is the honest soft spot in Position A: the loading evidence is currently bone-density- and function-based, and fracture-endpoint RCTs for high-intensity loading are still limited.
• We'd re-weight the drug recommendation if newer data showed the atypical-fracture/osteonecrosis harm rate is materially higher than the registry figures suggest, narrowing the population in whom benefit clearly dominates.

Industry bias note

Structural incentives the evidence base may reflect

Cui bono runs in both directions here, and the entry balances it explicitly.
• Toward over-claiming / what sells. Much of the positive calcium-supplement literature and the branded MK-7 K2 trials (e.g. NattoPharma/Gnosis-linked) are manufacturer-adjacent, with a clear pro-supplement incentive; treat single positive branded RCTs with caution. "Drink milk for strong bones" carries a dairy-industry tailwind. And the foundational bisphosphonate efficacy trials (FIT, HORIZON) were manufacturer-funded (Merck, Novartis), a pro-drug incentive worth naming, even though the efficacy result has held up in later independent analysis.
• Toward the skeptic counterweight. The Bolland/Reid Auckland group behind the weak-fracture-benefit and cardiovascular-signal findings is independently funded and deliberately adversarial to the supplement industry, a designed counterweight to supplement-funded positive trials. That means their findings should not be dismissed as pro-drug spin, but also should not be taken as settled: later independent analyses softened the cardiovascular claim. Crucially, the harm side of the drug story was quantified by the INDEPENDENT Black 2020 Kaiser cohort, not by industry, and it STILL concludes benefit dominates in high-risk groups.
• The net read. The best-supported lever, loading, and the safer mineral source, food, have essentially NO commercial sponsor pushing them, while the weakest levers (isolated calcium and K2 pills) have the loudest marketing. That inversion, best evidence for the least monetisable lever, is itself the signal. Net editorial stance: trust the un-monetised levers first, and read every pro-supplement or pro-drug claim against who funded it. (See cui_bono_industry_funding_bias for the general pattern.)

Bone Health: Load It, Feed It Protein, and the Calcium Story Is Overrated