Your bone density at 45 vs. 55 — the window most women don't know is closing.

In the five to ten years surrounding the final menstrual period, women lose bone at a rate that dwarfs any other decade of life. The interventions that matter most work best before the damage compounds — and most women are acting too late, or not at all.

The acceleration nobody warned you about

Bone density is a slow-moving story for most of adult life. From your late 20s (peak bone mass) through your late 30s, you lose roughly 0.5–1% of bone mineral density per year from the spine and hip — gradual, manageable, largely irrelevant in the short term. Then the menopause transition happens, and everything changes.

In the two to three years immediately before and the two to three years immediately after the final menstrual period — a window of roughly four to six years — bone loss accelerates dramatically. The most frequently cited figures come from the SWAN bone density sub-study, which tracked lumbar spine and femoral neck density longitudinally in women through the transition. Rates of loss during the late perimenopausal and early postmenopausal period averaged 1.8–2.5% per year at the lumbar spine and 1–1.5% per year at the femoral neck. Some women lost 3–5% per year in peak-loss years.

Over a decade straddling menopause — from age 47 to 57, say — a woman might lose 15–25% of her spinal bone mineral density. To understand what that means in absolute terms: the difference between ostepoenia and osteoporosis at the spine is roughly 10–15% of bone mineral density. A woman who enters her menopause transition with excellent bone density can exit the peak-loss window in osteopenic territory without any symptoms, any fractures, or any warning.

When to get a DEXA scan — and what the T-score means

A DEXA (dual-energy X-ray absorptiometry) scan measures bone mineral density at the lumbar spine and hip and reports it as a T-score — the number of standard deviations above or below the mean for a healthy young adult. A T-score of 0 is average for a 30-year-old. A T-score of −1.0 to −2.5 is osteopenia. Below −2.5 is osteoporosis by WHO definition.

The standard screening age for DEXA in most national guidelines (including the US Preventive Services Task Force) is 65 — but this is a minimum, not an optimal, recommendation. Women with risk factors for accelerated bone loss should be scanned earlier. Those risk factors include: early menopause (before 45), low body weight (BMI under 19), significant smoking history, long-term corticosteroid use, family history of hip fracture, and — critically — current or recent perimenopause with significant vasomotor symptoms, which is associated with higher rates of bone turnover even before the final period.

If you're 47 and perimenopausal with any of the above risk factors, a baseline DEXA now gives you a starting point. A follow-up scan in two to three years tells you your rate of loss — which is arguably more useful than a single cross-sectional T-score, because rate of change predicts fracture risk better than absolute density in premenopausal and early postmenopausal women.

What actually moves bone density

Hormone replacement therapy is the most potent pharmacological tool available for preventing perimenopausal and postmenopausal bone loss, and its effect size is large enough to matter at a population level. The Women's Health Initiative, despite its much-discussed limitations, found that combined estrogen-progestogen HRT reduced hip fracture risk by 34% (HR 0.66, 95% CI 0.45–0.98) and vertebral fracture risk by 34% in postmenopausal women over 5.6 years of follow-up. Estrogen-only HRT in women without a uterus showed a 39% reduction in hip fracture risk.

The bone protection from HRT is present at standard doses and is maintained throughout the duration of treatment. When HRT is stopped, bone loss resumes — so the benefit is not permanent, but it is substantial during the years of highest loss rate. For women who start HRT in early perimenopause or within a few years of the final period, the effect is to dampen the accelerated loss curve substantially. This is the "timing hypothesis" applied to bone: starting earlier captures more of the window where the intervention matters most.

Resistance training is the non-pharmacological intervention with the best evidence for bone. Impact loading and mechanical loading stimulate osteoblast (bone-building cell) activity through mechanotransduction — the process by which bone cells sense and respond to physical force. A meta-analysis by Zhao et al. in Osteoporosis International (2015) analysed 22 RCTs and found that resistance training produced a statistically significant increase in lumbar spine BMD (mean difference 0.015 g/cm²) and femoral neck BMD in postmenopausal women. That's a modest absolute effect — the intervention isn't replacing lost bone at scale — but it consistently slows further loss and in some studies produces small gains, particularly at weight-bearing sites.

The key mechanism for exercise-induced bone preservation is specificity: the bone adapts to the loading pattern it experiences. Jogging preserves hip density better than cycling. Squatting and deadlifting load the lumbar spine and femoral neck — exactly the sites most vulnerable in postmenopausal osteoporosis. Yoga and Pilates are beneficial for balance and fall prevention (a separate and important variable in fracture risk), but they don't apply the mechanical loading needed to stimulate significant bone remodelling.

Calcium and vitamin D — the baseline that most people get wrong

Calcium and vitamin D are the most discussed bone nutrients and also the most frequently misapplied. The mechanistic logic is clear: calcium is the primary structural mineral in bone, and vitamin D is required for intestinal calcium absorption and for the mineralisation of new bone matrix. Without adequate calcium and D3, no other intervention — HRT, bisphosphonates, resistance training — can work properly.

The confusion arises because high-dose supplemental calcium has been linked in some (not all) studies with increased cardiovascular risk, particularly the 2010 Bolland et al. meta-analysis in the BMJ. The more nuanced interpretation, supported by a 2019 reanalysis and by the National Osteoporosis Foundation's position paper, is that the risk signal appears to be driven by supplemental calcium in women who already have adequate dietary calcium — not by calcium obtained from food. The current evidence does not support systematic avoidance of calcium supplementation in genuinely calcium-deficient women, but it does support preferring dietary sources and using supplements to close gaps rather than to achieve dramatically supraphysiological intakes.

Dietary calcium targets for women over 50 are 1,200 mg/day. The average dietary intake is around 700–800 mg/day, leaving a gap of 400–500 mg for many women. Vitamin D sufficiency (serum 25-hydroxyvitamin D above 50 nmol/L, ideally above 75 nmol/L) is required for calcium absorption to work efficiently — and vitamin D insufficiency is extraordinarily common in northern latitudes, particularly in winter. A 2022 randomized controlled trial in The New England Journal of Medicine (LeBoff et al., the VITAL trial) found that vitamin D3 supplementation in older adults did not significantly reduce fracture risk in the overall population — suggesting that D3 alone is not sufficient, but that its role in calcium metabolism and bone mineralisation remains a necessary foundation rather than a standalone intervention.

The window is real — and it's narrower than you think

Bisphosphonates (alendronate, risedronate, zoledronic acid) and newer agents like denosumab or romosozumab are available for women with established osteoporosis or very high fracture risk, and they work well in that context. But the goal is not to arrive at 60 with osteoporosis and then treat it. The goal is to protect bone density through the perimenopausal decade — the window of accelerated loss — with the tools that work before the structural damage compounds.

A baseline DEXA scan in your mid-40s, a FRAX risk calculation, a genuine conversation about HRT timing with a menopause-literate clinician, and a resistance training programme that loads the spine and hips: that combination, started at 46, does something fundamentally different than the same conversation at 58. The bone you preserve through the loss window is bone you don't have to rebuild — and rebuilding is harder, slower, and pharmacologically more aggressive than protection. That asymmetry is the reason this window matters.