Bone Health & Resistance Training

The skeleton is the quietest tissue in the body — it does not ache as it weakens, and it changes on a scale of months, not weeks. This page is the loading case: what exercise can and cannot do for bone, and how to load the skeleton without loading it into danger.

Why Bone Needs Loading

Bone is one of the few tissues in the body that gets stronger when it is physically demanded of, and weaker when it is not. The mechanostat — Harold Frost's 1987 framing — describes a thermostat-like system: strain above a set point nudges bone formation, strain below it nudges resorption (Frost 1987). Bed rest and spaceflight make the point brutally: unload the skeleton and it sheds mass on a measurable schedule.

• 🦴 The strain window — classic loading studies associated the adaptive response with roughly 1,000–3,000 microstrain; below that range, resorption dominates (Rubin & Lanyon 1985). Everyday walking sits under the adaptive threshold.
• ⏳ The slow clock — one bone remodeling cycle runs 3–6 months. The skeleton's answer to a new training stimulus arrives on that schedule, not next week's.
• 📉 The midlife math — bone mass peaks around age 30, then declines; loss accelerates after menopause and with age. Loading in midlife is a preservation-and-possibly-gain strategy, not a repair.

The practical consequence: the skeleton rewards the same habits that build muscle — progressive overload — but it judges them on a much slower ledger.

How Bone Decides to Adapt

Not all loading is equal in bone's eyes. Charles Turner's three rules, drawn from decades of animal loading studies, still organize the field (Turner 1998):

• ⚡ Dynamic beats static — bone responds to changing strain, not held strain. A squat's loading and unloading cycle signals far more than standing under a load would.
• 📏 Magnitude matters — within limits, bigger strains produce bigger responses. This is the case for heavy resistance and for impact, and the reason walking alone underwhelms.
• 🎲 Novelty counts — the same stimulus, repeated forever, stops registering. Rotating exercises and progressions keeps the skeleton paying attention.
• 🔁 Bout separation — loading split into shorter bouts with rest between them appears to signal more than one long session; animal work found this pattern enhanced bone formation (Robling 2001).

What the Trials Actually Show

The strongest single study in the exercise-and-bone literature is LIFTMOR: 101 postmenopausal women with osteopenia or osteoporosis trained twice weekly for eight months, either supervised high-intensity resistance plus impact (five sets of five at over 85% of one-rep max) or a low-intensity home program (Watson 2018). The high-intensity group gained 2.9% at the lumbar spine while controls lost 1.2%; the femoral neck held nearly steady (+0.3%) while controls lost 1.9%. One minor adverse event — a back spasm — was reported.

• 📈 The spine moved — lumbar spine was the responsive site; a 2.9% gain reverses a year or more of expected postmenopausal loss, in women already below normal density.
• 🩺 Safety was tracked, not assumed — the program was supervised, screened, and deliberately progressed; that design is part of the result, not a footnote.
• 🧭 Site-specificity is the rule — reviews find spine responds to axial loading while hip effects are smaller and slower, and upper-extremity sites barely move (Martyn-St James & Carroll 2006; Howe 2011). Men show small but measurable benefits too (Kelley 2000).

The Loading Menu

The practical menu has four families, each loading bone differently — and they combine better than they compete. The dose framing comes from the ACSM position stand: impact on most days plus resistance on 2–3 days per week is the general pattern associated with bone health (Kohrt 2004).

Ground-reaction forces are approximate multiples of body weight across the biomechanics literature: walking ≈1.2, brisk walking ≈1.6, jogging ≈2.5, running ≈3.0, jump landing ≈4.5. The jump is the biggest signal the skeleton can get without a barbell — and the one that demands the most careful progression. The loading menu subtopic turns this table into a buildable routine.

Loading Rules That Protect the Spine

The same strain that builds bone can injure it when the spine is asked to bend and bear at once. The rules are simple, and they are the difference between loading and overloading.

• 🪜 Progress like everything else on this site — start light, add load in the smallest steps, and follow the double-progression pattern; the skeleton wants the same graduated asking as the muscle.
• 🧍 Keep the spine neutral under load — avoid end-range spinal flexion while lifting; the exercise selection page owns the movement-quality layer.
• 🚩 Sharp pain is a stop sign — joint pain or new back pain means stop that movement today, not the program; swap in a pain-free pattern. Pain that persists is a clinician question.
• 😴 Let the remodeling happen — loading sessions need recovery like muscle does; 48 hours between hard sessions for the same region is a sensible default.

Osteoporosis, Fracture History & the Safety Line

Everything above assumes an uncomplicated skeleton. That assumption is the line: known osteoporosis, fracture history, or bone-related conditions change the safety calculus — clinician or physiotherapist guidance is required. The epidemiology explains why the line exists: a prior fragility fracture roughly doubles the risk of another, independent of density (Kanis 2004), and a first vertebral fracture raises the risk of a new one within the following year about five-fold (Lindsay 2001).

None of these rows means "never train." Every one means "build the plan with guidance before loading" — the LIFTMOR trial's strongest evidence came from supervised, screened women with diagnosed low bone mass. The safety-line subtopic goes deeper into what that supervision looks like.

Falls: The Other Half of the Equation

Density gets the headlines, but most hip fractures follow a fall. A systematic review of 88 trials with nearly 20,000 community-dwelling older adults found exercise cut fall rates by about a fifth — and programs that challenged balance and ran more than three hours weekly cut them by nearly 40% (Sherrington 2017).

• 🧘 The exercises are cheap — single-leg stands, heel-to-toe walking, tai chi; the stability & mobility pillar page has the menu.
• 📊 The dose is the point — the 39% figure came from programs that challenged balance and totaled three-plus hours weekly; thirty seconds of standing is not the same intervention.
• 🦴 The bone link is direct — a skeleton you load, on legs that do not fall, is the whole program; each half protects the other.

The Decade-Long Plan

Bone work is measured in years, and the plan should be built to survive them. Density trends are read across scans, not snapshots — DXA re-test cadence is commonly about every two years, and screening typically begins at 65 for women unless risk factors push it earlier (Cosman 2014).

• 🏋️ Train the loading, keep it boring — two or three resistance sessions plus impact most weeks, progressed on the double-progression rules; novelty in exercise selection, consistency in the habit.
• 🥩 Feed the bricks — protein at the 1.2–1.6 g/kg working target, calcium 1,000–1,200 mg daily (the higher end for women 51+ and men 71+), vitamin D 600–800 IU; food first, and a D-status check via a clinician if unsure (Weaver 2016).
• 📅 Expect the schedule — first detectable spine changes around six months, hip changes on a 12–24-month horizon; a flat first-year scan is data, not failure.
• 🔄 Re-screen on the clinical rhythm — DXA every couple of years, with the results steering the plan; the long-term plan subtopic owns the full decade view.

Questions, Answered Briefly

• ❓ Can I build bone with bodyweight alone? — Partly. Bodyweight squats and push-ups load the skeleton, and jumps add impact, but the strain-magnitude rule favors progressive resistance for the spine and hip; the two combine better than either alone.
• ❓ Do I need to jump if I have knee problems? — No. Resistance work plus balance training covers most of the benefit without impact; joint limitations are clinician territory before any loading program.
• ❓ Will lifting make my bones denser? — The honest answer: loading is associated with gains at loaded sites in trials, and nobody can promise what your DXA will show. Density is judged on trends across scans, and the years between them are where the work happens.
• ❓ Is walking enough? — For density, likely not: walking's ground-reaction force sits near or below the adaptive range. Walking is excellent for everything else; bone wants the heavier and the faster ends of the menu.

This Page in One Workflow

1. Screen — before the first loaded rep: any red flag from the table, any fracture history, any bone-related condition? That decides whether the plan starts with a clinician or with a barbell.
2. Load — two or three resistance sessions weekly, spine neutral, progressed on double progression; add impact (jumps, hops) as the entry level allows.
3. Balance — a few minutes of single-leg and dynamic balance work most days; the falls data earn their place in the plan.
4. Feed — protein at the 1.2–1.6 g/kg target, calcium and vitamin D covered from food where possible.
5. Re-measure — DXA on the clinical rhythm, roughly every two years; judge the trend, not the snapshot.

Run the loop for a year before judging it — the skeleton answers on a remodeling clock, and the first scan after you start is often the baseline that later scans beat.

Go Deeper: Bone Health & Resistance Training

These five companion pages turn the topic into smaller, testable practices.

• 🔗 Bone Responds to Loading
• 🔗 What Trials Show for Bone Density
• 🔗 The Loading Menu
• 🔗 Osteoporosis, Fracture History & the Safety Line
• 🔗 The Long-Term Bone Plan

Related Topics