Her Bones, Her Rules
In 2021, a systematic review of military load carriage research by Walsh and Low catalogued every study examining how the human body responds to carrying weight during walking. The list was long. The evidence was detailed. The sex breakdown was not complicated: nearly all of the studies had been conducted on young men. Not middle-aged men. Not older men. Young men, predominantly soldiers in their early twenties. Walsh and Low did not bury this finding in a footnote. They named it plainly as a gap in the literature, a systematic blind spot that had been producing load carriage protocols—training standards, injury thresholds, safe progression guidelines—for a population that constituted perhaps half the people who might actually use them. The woman carrying a pack along a trail in her late forties has been following rules written for a twenty-two-year-old male infantryman. She was never included in the study.
This chapter is about her.
The central fact is not complicated, though its implications take some unwinding. The female body responds to mechanical loading differently than the male body. Not worse, and in some important respects better, but differently. That difference matters enormously when you are designing a protocol intended to build bone, protect the patellofemoral joint, and preserve functional independence into the seventh and eighth decades of a woman’s life. The difference cannot be read off a scale or corrected by dividing the load by body weight. It is structural. It is biomechanical. And it is, for most women who have encountered fitness advice about weight training and load carriage, almost entirely invisible.
What follows is the evidence.
The Patellofemoral Question
In 2025, Richard Willy and colleagues published what may be the most consequential study ever conducted on female load carriage biomechanics. The study was not large by epidemiological standards—twenty males and twenty females, all healthy adults, walking at a controlled speed of 1.35 metres per second with three load conditions: no external load, a 20.4-kilogram pack, and a 34.0-kilogram pack. The researchers instrumented each participant with motion capture and force plates and used a validated musculoskeletal model to estimate patellofemoral joint stress at every point in the gait cycle. They were interested in a specific question: does the patellofemoral joint respond to load carriage differently in women than in men, and if so, why?
The answer was unambiguous. Females exhibited significantly greater peak patellofemoral joint stress, greater impulse per step, and greater cumulative patellofemoral joint stress across all load conditions compared to males. The sex-by-load interaction was statistically robust, with p-values between 0.002 and 0.005 and partial eta-squared values of 0.12 to 0.13—effect sizes that are large enough to be clinically meaningful, not merely statistically detectable. As load increased from zero to 20.4 kilograms to 34.0 kilograms, the divergence between male and female patellofemoral stress widened. The additional load hit female knees harder, and it hit them in a way that accelerated disproportionately with the load added.
The researchers then did something important. They controlled for everything they could think of: body mass, height, and quadriceps strength. The sex difference persisted. After accounting for all three variables, females still demonstrated substantially greater patellofemoral joint stress during loaded walking than males of comparable size and strength. The gap was not an artifact of smaller bodies or weaker muscles. It was something more fundamental.
What that something is requires a brief anatomical detour. The patellofemoral joint—the articulation between the kneecap and the groove of the femur it rides in—is exquisitely sensitive to the alignment geometry of the quadriceps mechanism. The Q angle, which describes the angle between the line of quadriceps pull and the patellar tendon, is on average larger in women than in men, a consequence of the broader female pelvis relative to femoral length. A larger Q angle creates a greater lateral vector on the patella with every quadriceps contraction, increasing the compressive and shear forces across the patellofemoral joint surface. This is not a defect. It is an anatomical reality that evolved alongside the same pelvic architecture that enables childbirth. But it means that when you add external load to the walking gait—which increases quadriceps demand throughout the stance phase—the female patellofemoral joint absorbs more stress per step than the male patellofemoral joint under the same conditions.
FIGURE: The Q-Angle and Patellofemoral Stress
The Q angle describes the angle between the line of quadriceps pull (ASIS to patella centre) and the patellar tendon (patella centre to tibial tubercle).
Female average: ~17° (wider pelvis → greater lateral vector on patella).
Male average: ~12°.
Under load carriage, every quadriceps contraction pushes the patella laterally. The larger the Q angle, the greater the compressive and shear forces on the patellofemoral joint surface.
Why 110–120 spm cadence matters: Higher cadence = shorter stride = lower peak quadriceps force per step = reduced patellofemoral stress per cycle. This is not a preference. It is mechanical necessity for the female knee under load.
Data: Willy et al. 2025 — sex × load interaction p = 0.002–0.005.
Willy and colleagues drew the practical implication directly. Female recruits, they wrote, may benefit from slower progressions of load carriage training, either in amount carried or distance trained, to reduce their risk of patellofemoral pain. This is not a caution born of fragility. It is a caution born of data. The appropriate response to understanding that women’s knees experience greater patellofemoral stress during loaded walking is not to remove the load. It is to build the load more carefully and to build the knee’s capacity to handle it more deliberately before the load arrives.
The Willy et al. finding did not emerge in isolation. Xu and colleagues had established foundational context in 2016, using an integrated musculoskeletal finite-element model—built on a female subject’s data, which itself was a rarity—to evaluate the effects of graded load carriage on tibial stress during walking. In that model, a 66-kilogram woman walking at 1.3 metres per second with progressively heavier loads generated disproportionate increases in internal joint reaction forces as load increased. At 30 percent of body weight, knee joint reaction forces had increased by 26.2 percent from unloaded baseline. Ankle joint reaction forces had increased by 16.4 percent. Hip joint reaction forces had increased by 19.0 percent. The model revealed what Xu and colleagues described as coordinated adjustments in lower extremity musculature—the nervous system recruiting additional muscle force to manage the additional load, and that additional muscle force itself driving joint compression. External ground reaction forces understated the true internal loading story. The bones and cartilage were experiencing more than the force plates could see.
The implications for protocol design are specific. Women beginning a rucking program are not beginning a male program at lower weights. They are beginning a program appropriate to their own biomechanical landscape, which requires conservative load progression not as a compromise but as a strategic imperative. The patellofemoral joint needs time to adapt. The articular cartilage, the subchondral bone, the vastus medialis oblique that controls patellar tracking—all of these structures respond to mechanical loading by remodelling toward greater capacity, but only if the loading stimulus is introduced gradually enough that remodelling can keep pace with demand. Women who progress load carriage too quickly do not break some arbitrary rule. They exceed the adaptation rate of a joint that is already working harder than its male counterpart.
What the Data Was Missing
Walsh and Low’s 2021 systematic review established something that the exercise physiology community has been slow to absorb. The entire evidence base for military load carriage—injury thresholds, safe progression protocols, return-to-activity guidelines—was built almost exclusively on young men. This is not a criticism of the researchers involved. Military populations are predominantly young and predominantly male, and research follows the population available. But the consequence is that the protocols derived from that research carry an implicit assumption: that the biomechanical responses documented in young male soldiers generalize to every other population that might carry a pack.
They do not.
The age gap matters as much as the sex gap. No published study has directly examined load carriage biomechanics in women aged 40 to 60. No study exists for women over 60. The entire population of postmenopausal women—the population with the most to gain from the osteogenic effects of loaded walking and the most to lose from injury—is unrepresented in the data from which we derive our understanding of how the body responds to carrying weight. When a 54-year-old woman with early osteopenia straps on a pack for the first time, she is following protocols extrapolated from studies of 23-year-old male soldiers. The extrapolation might be defensible. It might even be approximately correct for many outcomes. But it is an extrapolation, and the confidence interval around that extrapolation is much wider than any published protocol acknowledges.
What we can say with confidence is that load carriage magnifies whatever biomechanical characteristics a person brings to the activity. If a middle-aged woman has reduced quadriceps strength relative to her peak, which is common across the menopausal transition, the patellofemoral joint stress per step will be higher than in a younger woman of the same load and speed. If she has age-related changes in articular cartilage, the consequences of elevated patellofemoral stress are different than they would be in younger cartilage. If she has accumulated years of sedentary posture—hip flexors shortened, gluteus medius weakened, thoracic spine stiffened—the compensatory demands of load carriage on her already-stressed joints begin from a worse starting position.
The recommendation is therefore not to avoid load carriage. The recommendation is to begin with more conservative loads than a young man would need, progress more slowly, and invest the early months of a rucking program in building the musculature that protects the patellofemoral joint before the loads become substantial. Specifically: quadriceps strength with emphasis on terminal knee extension and single-leg stability, gluteus medius activation for pelvic control, and hip flexor mobility that prevents the anterior pelvic tilt that increases patellofemoral stress during walking. These are not corrective exercises appended to a real program. They are the preparation that makes the real program safe and productive.
There is a further implication that the data suggest but the literature has not yet stated directly. Tibial stress fractures are the injury most consistently associated with female military load carriage, occurring at higher rates in female trainees than male trainees and appearing to be driven at least in part by the elevated loading rates that Johnson et al. documented—higher vertical loading rates during ruck marching in female Army trainees compared to male trainees. Tibial stress fractures are not random events. They occur when accumulated bone fatigue exceeds the repair capacity of the tissue at the loaded site. The repair capacity depends on training history, nutrition, hormonal status, and the rate at which load is progressed. A conservative progression protocol directly reduces the rate of fatigue accumulation. Adequate energy intake preserves the hormonal environment for repair. Strength work in the hip and ankle musculature distributes forces more evenly across the tibial cross-section. None of this is speculative. The mechanism is understood. The intervention is available. It requires only that the protocol respect the reality of the female musculoskeletal system rather than approximate the assumptions that served a different population.
The Perimenopause Imperative
There is a window. It opens somewhere in the mid-forties for most women and closes, not dramatically but definitively, somewhere in the decade after the final menstrual period. During this window, the biological conditions for bone building are deteriorating—estrogen is declining, bone resorption is accelerating, the protective hormonal environment that maintained skeletal integrity across the premenopausal years is withdrawing. And during this same window, the skeleton’s responsiveness to mechanical loading remains intact, or nearly so. The mechanosensory apparatus in bone—the osteocytes that detect strain and signal for new matrix deposition—continues to respond to load. The window is not locked. But the conditions on the other side of it are different, and waiting to act until menopause is complete means starting from a lower baseline with a less responsive system.
This is the perimenopause imperative.
The Sipilä et al. study of 1,393 women aged 47 to 55, published in 2020 in the Journal of Cachexia, Sarcopenia and Muscle, provided some of the most precisely rendered evidence available for understanding what happens to bone and muscle during the menopausal transition. The study measured femoral neck bone mineral density and appendicular lean mass across four menopausal stages: premenopausal, early perimenopausal, late perimenopausal, and postmenopausal. The decline was not linear. It accelerated. The transition from late perimenopause to postmenopause produced the steepest losses. Women who had maintained high physical activity levels across all menopausal stages showed significantly greater appendicular lean mass at every stage—but the protective association between physical activity and femoral neck bone mineral density was strongest specifically during late perimenopause, the last period before the final hormonal threshold is crossed.
The practical translation is uncomfortable but important. A woman who begins a serious mechanical loading program at 55, two years after menopause, is not too late. The evidence is clear that postmenopausal women derive meaningful BMD benefit from weight-bearing exercise, and the meta-analytic data from Sánchez-Trigo et al. are compelling on this point. But a woman who begins at 47, during early perimenopause, is working with the hormonal environment rather than against it. She is loading bone that is still partially defended by estrogen. She is depositing skeletal capital at a moment when the biological machinery for accumulation is still running, before it shifts decisively toward withdrawal.
The Sánchez-Trigo et al. 2022 meta-analysis of non-supervised exercise interventions in adult women found standardised mean differences of 0.73 to 0.85 for bone mineral density changes at the lumbar spine and femoral neck in women who already had osteopenia or osteoporosis. Those effect sizes are large by the standards of bone biology research. Exercise was doing meaningful work on bone even after the window had partially closed, even in women whose BMD had already declined to pathological thresholds. But the key phrase is non-supervised exercise—the kind of thing a motivated woman does on her own, with a pack on her back, three mornings a week on trails near her home. This is not a finding that requires a gymnasium or a physiotherapist or a prescription. It requires a pack, a progressive protocol, and the understanding that this work matters.
The menopausal transition also imposes a simultaneous loss of muscle mass—sarcopenia proceeding in lockstep with bone loss—that compounds the clinical stakes of inaction. The biomechanical coupling theory holds that bone adapts to the mechanical forces generated both by gravitational loading and by muscle contraction. When muscle mass declines, the contractile forces that load bone decline with it. Sarcopenia and osteoporosis do not merely coexist; they are mechanistically linked. A study of 119 elderly women found that sarcopenic women had more than four times the odds of osteopenia and osteoporosis compared to non-sarcopenic women. Physical inactivity alone carried an odds ratio of 5.5. These numbers are not describing two separate problems. They are describing one compound problem: a body that stopped receiving mechanical loading signals, and bones and muscles that responded to that absence in the way biology designed them to respond.
Rucking addresses both simultaneously. The load carried provides the gravitational and mechanical stimulus to bone. The posterior chain—gluteus maximus, erector spinae, deep multifidus—works continuously under that load to maintain posture and propulsion, generating the contractile forces that bone uses as its primary remodelling signal. The quadriceps, which decline disproportionately in perimenopause, are loaded through every step of the stance phase. The hip extensors are firing to resist the forward pitch the pack creates. This is not incidental to the osteogenic effect. It is central to it. Loaded walking is not bone stimulus plus muscle stimulus. It is a single stimulus that works on both tissues through the same mechanical pathway. That is the case for beginning before the window closes.
Building Bone That Lasts
The biomechanical mechanism through which loaded walking builds bone is Wolff’s Law, which states that bone remodels in response to the mechanical forces imposed on it. The cellular machinery implementing this law begins with osteocytes—the most abundant bone cells, embedded in the mineralized matrix, connected to each other and to the bone surface by a network of fine processes threading through channels called canaliculi. When bone deforms under load, fluid flows through the canalicular network and the osteocytes detect this flow as a mechanical signal. They respond by releasing signalling molecules that activate osteoblasts—the bone-building cells—and suppress osteoclasts—the bone-resorbing cells. The net result, sustained over weeks and months, is increased bone mineral density at the loaded sites.
The magnitude of this response depends on the magnitude and rate of loading. Walking alone, without additional weight, is insufficient to drive meaningful bone accrual in postmenopausal women. The Honisett et al. 2016 randomised controlled trial demonstrated that moderate walking without added load did not alter bone formation or resorption indices in postmenopausal women. The mechanical stimulus was inadequate. What the skeleton needed was more force per step.
Snow et al. addressed this directly. Their five-year randomised controlled trial, published in 2000, assigned postmenopausal women to a weighted vest plus jumping program. At five years, women in the intervention group had preserved femoral neck bone mineral density while control subjects had declined. The load delivered the signal the skeleton required. The effect was durable. It persisted across years of follow-up. The bones that were loaded continued to differ from the bones that were not.
For rucking specifically, the relevant physics are not complicated. A 65-kilogram woman carrying a 20-kilogram pack walks with approximately 85 kilograms of weight pushing down through her lower limbs with every step. The ground reaction force at heel strike and midstance is elevated proportionally. The Xu et al. model predicts a 26-percent increase in knee joint reaction forces and a 19-percent increase at the hip at 30 percent of body weight. These are not forces that simply pass through the body harmlessly. They are forces that the skeleton detects, interprets as evidence of mechanical demand, and responds to by building more bone. The canalicular flow is occurring. The osteocytes are signalling. The osteoblasts are depositing matrix.
The cumulative stress is as important as the peak. A single heavy step does relatively little. Ten thousand steps with a pack—the rough yield of an hour of walking at moderate pace—delivers an osteogenic signal at every site in the loading chain: the calcaneus, the tibia, the femoral neck, the vertebral bodies. The lumbar spine, which is a particularly consequential site for osteoporotic fracture and a site that unloaded walking does not effectively stimulate, receives both the direct compressive load of the pack resting against it via the hip belt and the muscle-generated forces of the spinal extensors firing to maintain upright posture under load. The weighted-backpack back extension study cited by Daly et al. showed improved spinal BMD and reduced vertebral fractures eight years after a two-year intervention. The bones remembered what was asked of them.
The Pelvic Floor: Evidence Over Anxiety
The question of pelvic floor risk during load carriage generates more anxiety in women’s fitness conversations than the evidence warrants. This is worth addressing directly, with the same rigour applied to every other topic in this book.
The concern has a logical structure: physical activity increases intra-abdominal pressure; elevated intra-abdominal pressure places stress on pelvic floor structures; therefore physical activity, including load carriage, may cause or worsen pelvic organ prolapse. Each step in this chain sounds plausible. The overall conclusion is not well-supported by the evidence.
Bø and colleagues published a comprehensive narrative scoping review in the International Urogynecology Journal in 2023 examining the relationship between strenuous physical activity, exercise, and pelvic organ prolapse. One of the most important findings came from an IAP measurement study by Weir and colleagues, referenced within that review, which measured intra-abdominal pressure during a range of activities in 30 women and found that abdominal crunches, climbing stairs, walking on a treadmill, and many lifting activities did not increase intra-abdominal pressure significantly more than standing up from a chair. This finding was confirmed independently by O’Dell et al. Walking-based load carriage generates intra-abdominal pressure comparable to standing from a chair. This is the activity that was causing concern.
The pelvic organ prolapse prevalence data tell a similar story. Across eight prevalence studies of physically active women reviewed by Bø et al., rates of symptomatic prolapse varied from zero percent in several sport-specific populations to 23 percent in Olympic weightlifters and powerlifters. One factor was consistently associated with prolapse across studies. It was not exercise modality. It was parity—the number of vaginal deliveries. Women who had carried and delivered children had higher rates of prolapse. Women who had exercised heavily but not delivered vaginally were not at elevated risk from the exercise itself.
This is an important distinction that has been obscured in popular fitness communication. The woman who worries that rucking will cause prolapse is carrying an anxiety that the evidence does not support. The primary risk factor for pelvic organ prolapse is vaginal birth, specifically the mechanical strain of descent and delivery on the levator ani and its fascial attachments. Exercise that does not generate IAP substantially above standing from a chair—and walking with a pack does not—adds negligible risk to whatever parity-related vulnerability the woman already has.
There is a further consideration. Rucking, because it requires continuous activation of the trunk and pelvic stabilizers throughout the gait cycle under load, may provide a form of endurance training for the pelvic floor. Masroor et al., examining diaphragmatic breathing and core stabilisation under load, confirmed that the diaphragm, transversus abdominis, and pelvic floor function as integrated pressure management components during loaded movement. This is not three separate muscles doing three separate jobs. It is one system managing intra-abdominal pressure as a unified hydraulic unit. When the diaphragm descends on inhalation during nasal breathing under load, the transversus abdominis and pelvic floor respond with coordinated co-contraction—a reflex that has been observed consistently across EMG studies of loaded walking, even when participants were not instructed to engage any of these muscles. The transversus abdominis simply fires. The pelvic floor simply responds. The co-contraction is built into the loading pattern. Sixty minutes of loaded walking is sixty minutes of sub-maximal endurance work for the entire lumbo-pelvic cylinder. The clinical fitness community calls this kind of work core training. Ruckers simply call it a walk.
The clinical guidance that flows from this evidence is straightforward. Women with existing pelvic floor dysfunction, or with significant parity-related vulnerability, should work with a pelvic floor physiotherapist when initiating any loading program. That consultation is appropriate not because rucking is high-risk but because individual assessment always produces better protocol design than population-level generalisation. For women without existing symptoms, there is no evidence-based reason to avoid load carriage, and substantial evidence-based reason to undertake it.
The RED-S Line
There is one genuine metabolic risk in a rucking program for women, and it demands direct acknowledgment. It is not pelvic floor damage. It is Relative Energy Deficiency in Sport.
RED-S—or, in its earlier formulation, the Female Athlete Triad—describes the cascade of physiological disruption that follows inadequate energy availability in exercising women. The cascade begins with insufficient caloric intake relative to exercise energy expenditure, which depresses energy availability below the threshold the hypothalamic-pituitary-gonadal axis requires to maintain normal function. The HPG axis is metabolically expensive to run. When energy availability drops below approximately 30 kilocalories per kilogram of fat-free mass per day, hypothalamic GnRH pulse frequency decreases. Luteinizing hormone pulsatility follows. Estrogen declines. The skeleton, stripped of its hormonal defence, accelerates resorption.
The irony is precise: a woman who undertakes a rucking program for bone health, and simultaneously restricts her dietary intake in pursuit of body composition goals, may be accelerating the very bone loss she is trying to prevent. The mechanical loading creates the osteogenic signal. The hormonal environment created by energy deficiency overrides it. The osteocytes are signalling for bone formation. The estrogen deficiency is commanding bone resorption. The estrogen wins.
Holtzman and Ackerman, publishing in 2021 in Sports Medicine, established that an energy availability target of approximately 45 kilocalories per kilogram of fat-free mass per day is likely ideal for endurance athletes to maintain body mass and support training. This is not a ceiling. It is a floor. Below it, LH pulsatility is disrupted. Above it, the HPG axis maintains normal function and the bones receive both the mechanical signal and the hormonal permission to build.
For a 65-kilogram woman with 25 percent body fat—fat-free mass of approximately 49 kilograms—this translates to roughly 2,200 kilocalories per day before accounting for exercise energy expenditure. A 60-minute ruck with a 15-kilogram pack at moderate walking speed expends approximately 400 to 500 kilocalories above resting metabolic rate, depending on terrain and pace. That expenditure must be added. Not half of it, not a rough approximation—all of it. The architecture of bone health depends on this accounting.
Holtzman and Ackerman further established that protein intake should be at least 1.6 grams per kilogram of body weight per day, with evidence suggesting higher intakes during the follicular phase. For the 65-kilogram woman, this is at minimum 104 grams of protein daily, distributed across meals to maximise muscle protein synthesis—and bone matrix synthesis, which depends on the same amino acid substrates. Calcium and vitamin D are not optional accessories to a bone health protocol. They are raw materials. The mechanical signal demands matrix deposition. Matrix deposition requires mineral. Without adequate calcium intake and vitamin D status to support absorption, the osteoblast arrives at the construction site to find no building supplies.
The practical monitoring tool for HPG axis status is one that women already have access to: menstrual cycle regularity. A woman who begins a rucking program and finds her cycle shortening, lengthening, or disappearing has a direct readout of hypothalamic function. She is in energy deficit. The appropriate response is not to push through. It is to eat more, train less, or both, until cycle regularity is restored. This applies regardless of whether she is 35 or 45. It applies with particular force in the years around perimenopause, when estrogen is already declining and the HPG axis is already under hormonal stress. An energy deficit on top of a declining estrogen environment is not a difficult equation to solve. The answer is not more rucking. The answer is more food.
A Protocol Built for Her Body
What does a female-specific rucking protocol actually look like in practice?
The starting position is conservative by the standards of military load carriage—not because the female body is fragile, but because the patellofemoral joint needs time to adapt, the supporting musculature needs to be prepared before loads become substantial, and the evidence base for safe progression rates in women does not yet exist in the detail that a more aggressive protocol would require. Conservative is not a synonym for ineffective. It is a synonym for sustainable.
A 2025 review by Gai and colleagues, examining the effects of weighted walking on gait stability in older adults, recommended that external loads remain below five percent of body weight for older populations—substantially less than the fifteen to thirty percent range that military-derived protocols suggest (Gai et al., 2025). Their concern was gait instability and fall risk, not joint damage per se, but the recommendation underscores the gap between what young soldiers tolerate and what older civilians should attempt initially. This protocol begins conservatively for the same reason.
Start with five percent of body weight for the first four weeks. This is not a training load; it is an adaptation stimulus and an orientation period. The body is learning how to carry again—how to activate the posterior chain under axial load, how to manage the slightly altered centre of mass, how the hip belt and shoulder straps interact with posture and gait. For a 65-kilogram woman, five percent of body weight is 3.25 kilograms. This is not impressive. It is exactly right.
At week five, if the knees are symptom-free and the gait has stabilised—no excessive trunk lean, no pelvic drop, no knee valgus under load—the load increases to ten percent of body weight. This increase is held for four weeks. Patellofemoral joint adaptation proceeds at the cellular timescale, not the motivational timescale. The articular cartilage and subchondral bone are remodelling toward the new mechanical demand. This takes weeks, not days.
From ten percent, progression follows the same pattern: four weeks at each increment, no increase until the current level is symptom-free and biomechanically clean. The targets are meaningful: twenty percent of body weight provides substantial osteogenic stimulus and reaches the range where joint reaction forces are producing the bone-loading signals the Xu et al. model described. Twenty-five percent is where the effects on femoral neck and tibial bone geometry become robustly osteogenic based on the available data. Thirty percent is a significant load and may be the upper boundary for most recreational female ruckers who are using the activity for health rather than military preparation.
The symptom-free criterion deserves elaboration, because it is easy to dismiss and consequential to ignore. Anterior knee pain during or after a session is patellofemoral joint stress exceeding adaptation capacity. It is not delayed onset muscle soreness. It is not acceptable discomfort that builds character. It is a signal that the load or the pace or the terrain has moved ahead of the tissue’s capacity to adapt, and the appropriate response is to reduce load, not to push through. The Willy et al. data give this clinical guidance a biomechanical foundation: female patellofemoral joint stress diverges from male stress under load, it diverges more at higher loads, and the articular cartilage of the patellofemoral joint does not announce its distress early. Pain is a late indicator of a loading problem that preceded it by weeks. The conservative protocol is designed to keep the training stimulus below the threshold that produces pain, which means the tissue is adapting continuously without ever receiving the signal that it is being overtaxed.
The parallel work is not optional. Quadriceps strengthening—particularly single-leg terminal knee extension, which specifically targets vastus medialis oblique function and patellar tracking—should accompany every phase of the progression. Gluteus medius loading through lateral band walks, single-leg deadlifts, and step-ups trains the pelvic stabilisation that controls Q angle dynamics under load. Hip flexor stretching and thoracic mobility work counter the postural tendencies that increase anterior pelvic tilt during loaded walking. Calf raises and ankle mobility work support the dynamic stiffness adaptations that Santos et al. documented at the ankle under load—the ankle and knee becoming more mechanically efficient at managing the additional mass. These are not accessories bolted onto a rucking program. They are the program’s structural underpinning, and they double as a strength training stimulus for the same muscles whose loss drives sarcopenia and whose preservation drives independence.
Duration matters as much as load. The osteogenic stimulus of loaded walking is cumulative across steps, not concentrated in any single peak force event. The Xu et al. model showed that cumulative tibial stress during a single walking cycle was substantial and distributed across the medioposterior aspect of the tibia—a site at risk for stress fractures but also a site that responds to that cumulative loading with increased bone mineral density when training is progressive. Thirty minutes three times per week is a defensible starting point for bone benefit. Forty-five minutes accumulates considerably more osteogenic signal. Sixty minutes, sustained at Zone 2 intensity—identifiable by the ability to maintain nasal-only breathing throughout, a concept developed in detail in the next chapter—provides robust cardiovascular adaptation alongside the skeletal loading. The bone and the heart are being trained simultaneously. This is the hybrid benefit of the activity: it occupies the metabolic niche of Zone 2 aerobic training and the mechanical niche of resistance exercise at the same time, in a single session, requiring nothing more than a pack, a path, and appropriate footwear.
Terrain is an underappreciated variable. Walking on uneven surfaces, mild inclines, and varied gradients increases the multi-planar demands on the stabilising musculature and introduces the loading novelty that the bone mechanosensing literature identifies as a more potent osteogenic signal than repetitive, predictable loading. The osteocyte response is sensitive to strain magnitude and to strain novelty—bone adapts more strongly to loads it has not encountered before than to loads it has accommodated. A loop walked three times a week on the same flat path eventually delivers a diminishing signal. Adding a hill, reversing the direction, crossing a grassed area, or walking on a slight camber reintroduces novelty and maintains the adaptation stimulus. This is not a prescription for extreme terrain. It is a prescription for the mild unpredictability of ordinary outdoor walking as opposed to treadmill monotony.
Seen by the Science
The woman who picks up this book having spent years navigating fitness advice that was not designed for her deserves to be addressed directly.
The current scientific understanding of female load carriage biomechanics is incomplete. The Walsh and Low gap is real. The absence of load carriage biomechanics data in women over forty is a genuine failure of the research enterprise, not a hypothetical concern. The protocols we can offer are evidence-informed extrapolations from inadequate data, and that limitation should be stated plainly rather than hidden behind false confidence.
But incomplete is not empty.
What we know is this. The female patellofemoral joint experiences greater stress during loaded walking than the male patellofemoral joint, and the divergence grows with load. This requires conservative progression. The postmenopausal skeleton responds to mechanical loading with increased bone mineral density, particularly in women who already have osteopenia, with effect sizes large enough to be clinically meaningful. The perimenopause window is real and time-sensitive, and the most powerful intervention available to a woman in her forties who is watching her bone density scan numbers with growing concern is not pharmaceutical. It is mechanical. It is a pack and a progressive protocol and the discipline to eat enough to sustain it.
The pelvic floor is not the vulnerability it has been portrayed as. The energy balance is the vulnerability that has been underweighted. The patellofemoral joint requires specific attention and specific preparation. The timeline is more conservative than the male equivalent. None of this constitutes a lesser version of a male protocol. It constitutes the correct version of a female protocol—one derived from the evidence available, transparent about the evidence that does not yet exist, and designed around the biology that actually governs the outcome.
The woman on the trail at 49 with a twelve-kilogram pack and a training heart rate of 126 beats per minute, breathing through her nose, building bone with every step—she is not following a modified male program. She is following her body’s own protocol, implemented correctly. Her bones are adapting to the load they are receiving. Her patellofemoral joint is managing forces her protocol has prepared it to manage. Her HPG axis is intact because she ate breakfast before she left and will eat a real meal when she returns. Her pelvic floor is doing exactly what it evolved to do during sustained sub-maximal loaded locomotion: generating IAP for spinal stabilisation, co-activating with the diaphragm and transversus abdominis, working within its capacity.
She was born to carry.
The science, imperfect and incomplete and built mostly on data from young men she has never met, is finally beginning to see her clearly enough to serve her.
Her bones, her rules.
Chapter Notes
Willy et al. 2025 — The study recruited 20 males and 20 females (balanced for health status) and used three load conditions (0 kg, 20.4 kg, 34.0 kg) at 1.35 m/s. Patellofemoral joint stress was estimated via musculoskeletal modelling. Sex by load interaction: p = 0.002–0.005, partial eta-squared = 0.12–0.13. Controlling for body mass, height, and quadriceps strength did not eliminate the sex difference. Full citation: Willy R, Simon J, Hanser B, et al. Females exhibit greater peak and cumulative patellofemoral joint stress with moderate and heavy load carriage compared with males. European Journal of Sport Science. 2025. doi:10.1002/ejsc.70046.
Xu et al. 2016 — Female subject musculoskeletal-finite-element model (66 kg, 27 years, height 1.7 m). Load conditions: 0%, 10%, 20%, 30% BW via weight vest at 1.3 m/s. Knee JRF increase at 30% BW: 26.2%. Ankle: 16.4%. Hip: 19.0%. Citation: Xu C, Silder A, Ju Z, et al. An integrated musculoskeletal-finite-element model to evaluate effects of load carriage on the tibia during walking. Journal of Biomechanical Engineering. 2016;138(10). doi:10.1115/1.4034216.
Sánchez-Trigo et al. 2022 — Systematic review and meta-analysis, adult women only (n = 668), non-supervised exercise interventions. Overall BMD effects: lumbar spine SMD = 0.40, femoral neck SMD = 0.51. In women with osteopenia/osteoporosis: lumbar spine SMD = 0.73 (95% CI: 0.13–1.33), femoral neck SMD = 0.85 (95% CI: 0.33–1.37). Citation: Sánchez-Trigo H, Rittweger J, Sañudo B. Effects of non-supervised exercise interventions on bone mineral density in adult women: a systematic review and meta-analysis. Osteoporosis International. 2022;33(7):1415–1427. doi:10.1007/s00198-022-06357-3.
Walsh and Low 2021 — Systematic review of military load carriage effects on gait. All studies male participants (or predominantly male with one exception). Average participant age 20–31 years. Citation: Walsh G, Low D. Military load carriage effects on the gait of military personnel: a systematic review. Applied Ergonomics. 2021;93:103376. doi:10.1016/j.apergo.2021.103376.
Holtzman and Ackerman 2021 — Energy availability target: approximately 45 kcal/kg FFM/day. Protein minimum: 1.6 g/kg/day during follicular phase. LH pulsatility disruption below energy availability threshold. Citation: Holtzman B, Ackerman K. Recommendations and nutritional considerations for female athletes: health and performance. Sports Medicine. 2021;51(S1):43–57. doi:10.1007/s40279-021-01508-8.
Bø et al. 2023 — Narrative scoping review: POP prevalence in physically active women 0–23%. Parity consistently the only associated factor. IAP during walking comparable to standing from a chair (Weir et al., cited within). Citation: Bø K, Anglès-Acedo S, Batra A, et al. Strenuous physical activity, exercise, and pelvic organ prolapse: a narrative scoping review. International Urogynecology Journal. 2023;34(6):1153–1164. doi:10.1007/s00192-023-05450-3.
Snow et al. 2000 — Five-year RCT in postmenopausal women. Weighted vest plus jumping program preserved femoral neck BMD versus control decline. Direct evidence for osteogenic effect of external loading in postmenopausal women. Citation: Snow CM, Shaw JM, Winters KM, Witzke KA. Long-term exercise using weighted vests prevents hip bone loss in postmenopausal women. Journal of Gerontology: Medical Sciences. 2000;55(9):M489–M491.
Sipilä et al. 2020 — Cross-sectional study, n = 1,393 women aged 47–55. Significant linear declining trend in appendicular lean mass and femoral neck BMD across menopausal stages. Physical activity–BMD association strongest during late perimenopause. Citation: Sipilä S, Törmäkangas T, Sillanpää E, et al. Muscle and bone mass in middle-aged women: role of menopausal status and physical activity. Journal of Cachexia, Sarcopenia and Muscle. 2020;11(3):698–709. doi:10.1002/jcsm.12547.