RightRep EN · PT Open the app
BLOG

Can You Train With a Stress Fracture? The Bone-Stress Audit: Where Fredericson's MRI Ladder Runs from Periosteal Edema Through Progressive Marrow Involvement to Frank Cortical Fracture, Where Bennell's Twelve-Month Track Census Counts Stress Fractures in 21.1% of Competitive Athletes at 0.70 per 1,000 Training Hours with the Tibia at 46% and the Navicular at 15%, Where Arendt's 74 College Athletes Prove the Grade Predicts the Return Calendar, Where Taunton's 2,002-Injury Clinic File Puts a BMI Under 21 in Women Behind Tibial Stress Fractures and Spinal Injuries, Where Tenforde's 748 Adolescent Runners Make Prior Fracture the One Robust Predictor in Both Sexes, Where Nattiv's 30 kcal/kg Fat-Free-Mass Line, De Souza's Return-to-Play Point System and the IOC's RED-S Consensus Open the Under-Fuel Door, Where Moraes' Brazilian-Led 15-Study Meta Ties Vitamin D to Stress Fractures at −5.82 nmol/L While Admitting the Athlete Subgroup Found Nothing, Where Buist's 532-Runner RCT Retires the 10% Rule as Immunity and Warden's Dials — Softer Landings, Higher Stride Rate, Calf and Foot-Intrinsic Strength — Carry the Prevention File: the Bone Asked for a Smaller Dose on a Rising Schedule, Not the Couch, and This Is the One File in the Family Where 'Not on That Bone' Is the Honest Answer

By RightRep AI Assistant · · 23 min read

There is a fence at the top of this post and it outranks everything below it: groin, hip or thigh pain in a runner that is creeping, focal and worse with weight-bearing is the femoral-neck door, and it is the family's one true emergency — a tension-side femoral-neck stress fracture can become a complete fracture and displace, which converts a twelve-week ladder into a hip surgery, so it wants a clinician and imaging today, not a loading question [2][3]. The second fence: pain in the midfoot — the navicular, the base of the second metatarsal, the arch — is the high-risk-site door: these bones heal poorly on loading programs alone, they hide on early X-rays, and Bennell's own census put the navicular second only to the tibia at 15% of track-athlete stress fractures [2][4]. The third: a spine that hurts at a point, a night that wakes you, pain that has stopped caring whether you trained — infection, tumor, the rare files — is a clinician-now door in every audit this library has written. The fourth: bone stress that keeps coming back in an athlete who is restricting food — the female runner with low body weight and irregular cycles above all, but the male lifter cutting hard too — is the RED-S/triad door, where the bone keeps losing the arithmetic fight because the fuel line is cut, and the treatment is energy, not a running ladder [13][14][15]. And the fifth, the family border: diffuse pain along five centimeters or more of the posteromedial shin that aches after runs but not with a single fingertip is the shin-splints audit's territory — the early end of this same continuum — and this file is the far end, where the pain has a fingertip, a hop test fails, and walking itself hurts [1]. One honesty clause before the fun part: this library has opened the orthopedic family one file at a time — the ACL, the meniscus, the runner's knee, the jumper's knee, the hamstring, the shin, the Achilles tendon, the foot's fascia, the ankle ligaments, the frozen shoulder, the hip bursa — and every one of those files ends with "keep training, modify the dose." This is the file where the honest answer is different, and the difference is the whole point: the bone that is cracking is the one tissue in the family that must come off impact while it heals — and everything else you own keeps training. This is that file.

The short answer: not on that bone — and a graded ladder back, not the couch

Strip the file to its spine. A stress fracture is the far end of a continuum Warden's rulebook names precisely: bone stress injury is the inability of bone to withstand repetitive loading — it begins as a stress reaction, can progress to a stress fracture, and ultimately to a complete bone fracture [3]. It happens when the homeostasis between microdamage formation and its removal is disrupted — the bone received more impacts than it could remodel in time [3]. Most of these injuries heal, and the rulebook says so plainly: most bone stress injuries readily heal following a period of modified loading and a progressive return to running activities [3]. So the shape of the answer is three sentences. One: the affected bone comes off impact now — this is the family's only real stop sign, because the difference between a stress reaction and a complete fracture is, mechanically, just the number of additional load cycles you feed it [1][3]. Two: nothing else stops — the upper body trains on day one, the other leg carries the training frequency, the zone-2 bike and the pool hold the engine, and the affected limb trains whatever the injury site leaves available [9]. Three: the comeback is a ladder with rungs set by the MRI grade and the site's risk class, not a calendar you negotiate with — Arendt's college file found a significant correlation between the grade of injury and the time to full return to activity, which is why you grade before you program [7]. The couch, for the record, is not on the ladder either: bone remodels in response to loading signals, which is why every modern guideline phases from load control to pain-free progression rather than prescribing six weeks of nothing [3][9].

The continuum door: what a stress fracture actually is

The resting script ("it's a bone bruise — six weeks off, then start over") and the shopping script ("boots, braces, laser, the creams") both miss what the tissue is doing. Fredericson's runners drew the map: fourteen runners with eighteen symptomatic legs, imaged within ten days, and magnetic resonance imaging suggested a progression of injury that starts with periosteal edema, moves through progressive marrow involvement, and ends in frank cortical stress fracture [1]. That ladder is the file's central image — one road, mile markers, and your position on it is measurable [1]. It is also why the diagnosis question and the grading question are the same question: clinical symptoms such as pain with daily ambulation, and examination findings including localized tibial tenderness and pain with direct or indirect percussion, correlated with more severe tibial stress injuries [1] — the body reports its own mile marker if you ask it in the right order (fingertip, hop, walk, night). Matcuk's radiology file adds the taxonomy that keeps the term honest: fatigue fractures — abnormal cyclical loading on normal bone, common in runners and athletes — are only half the word's meaning; insufficiency fractures are normal or trivial loading on abnormal bone, the elderly and osteoporotic file [8]. The gym-and-track reader almost always holds the first kind: a normal bone that was asked, repetitively, for more remodeling than the week allowed [3][8]. And the Brazilian endocrinology review puts the presentation on one line worth memorizing: localized pain, with or without swelling, and tenderness on palpation of bony structures that begins earlier in training and progressively worsens over two to three weeks [10] — the creep, not the crack. Nothing in that sentence describes a snap; the snap is a different file entirely.

The census door: who gets these, and where

The numbers file is unusually clean because the armed forces and the track kept meticulous records. Bennell's twelve-month prospective census of competitive track and field athletes — 53 women and 58 men, aged 17 to 26 — counted twenty athletes sustaining twenty-six stress fractures, an overall incidence of 21.1%, at 0.70 stress fractures per 1,000 hours of training; those fractures were 20% of the 130 musculoskeletal injuries the cohort sustained all year, with no difference between male and female rates [4]. The site table in the same census is the programming door's atlas: the tibia led with 12 injuries (46%), followed by the navicular with 4 (15%) and the fibula with 3 (12%), and the event you run predicts the site — sprints, hurdles and jumps were associated with significantly more foot fractures, while middle- and long-distance running produced more long-bone and pelvic fractures [4]. Arendt's college file repaints the same map at higher resolution: of seventy-four athletes with lower-extremity symptoms consistent with stress injury to bone, 68 were confirmed — 61 by MRI, 6 by positive radiographs alone, 1 by bone scan alone — and distance runners accounted for the most stress injuries to bone for both men and women; the tibia (37%) was the most frequently involved single bone, while the foot as a region (44%) was the busiest address on the body [7]. Taunton's clinic census of 2,002 running injuries adds the body-composition door the family keeps rediscovering: women with a body mass index under 21 kg/m² were at significantly higher risk of tibial stress fractures and spinal injuries [6]. And the scale of the problem in any sports-medicine clinic, per the Brazilian review: stress fractures represent 10–20% of all injuries in sports medicine [10]. Read together, the census says: this is a distance-runner's and a load-spike athlete's file, it lives in the tibia and the foot, and the thinnest athletes carry a disproportionate share of it [4][6][7][10].

The grading door: the MRI ladder and the return calendar

The diagnostic sequencing is settled, and it is shorter than the internet's version. Radiographs come first: Matcuk's file has them revealing linear sclerosis and periosteal reaction prior to the development of a frank fracture, with MRI highly sensitive — findings ranging from periosteal edema to bone marrow and intracortical signal abnormality — and preferred over CT and bone scintigraphy when the X-ray is quiet and the finger still finds a spot [8]. Fredericson's original argument for MRI over bone scan is the one that still holds: MRI correlated with an established technetium bone-scan grading system in fourteen of the eighteen symptomatic legs while more precisely defining the anatomic location and extent of injury, without ionizing radiation and in significantly less imaging time [1]. Knobloch's 2025 review states the current standard flat out: MRI should be utilized as the gold standard for diagnosis, grading, and return-to-sport timing prognosis when available, with plain radiographs used as first-line imaging [9]. And the reason the grade matters more than any other number in this file is Arendt's: a ten-year Division I review where every injured athlete followed the same treatment program, and there was a significant correlation between the grade of injury and time to full return to activity [7]. That correlation is why the fence says grade before you program: two athletes with identical fingertip pain and identical X-rays can be standing on different rungs of the same ladder, and the MRI is what tells you which [1][7][9].

The high-risk/low-risk door: the site decides the rules

The single most practical classification in this file is Diehl's, and it is maddeningly simple: stress fracture management takes into consideration the injury site — low risk versus high risk — the grade, the extent of microdamage accumulation, and the individual's competitive situation [2]. The low-risk sites — the shaft of the tibia and fibula, the ribs, the posterior elements of the pars in some classifications — heal on the two-phase program this file walks: mechanical load control first, then pain-free progression [9]. The high-risk sites are the ones where the natural-history file is ugly and the conservative program carries real failure risk — the femoral neck (tension-side especially: the cortex on top is loaded in tension, and a tension-side lesion can complete and displace), the navicular (poor blood supply, delayed union, and second-busiest site in Bennell's track census at 15% [4]), the base of the fifth metatarsal, the medial malleolus, the pars interarticularis of the spine [2][3][9]. The programming difference is absolute, not a matter of degree: a low-risk tibial stress reaction on grade-appropriate rest can often begin a walk-jog ladder in weeks; a femoral-neck stress reaction is a clinician-now file where some presentations are surgical decisions before they are rehabilitation decisions [2][3]. Knobloch's review names the consequence of respecting the split: identification of low-risk versus high-risk sites, in conjunction with addressing modifiable risk factors and the two-phased approach, will optimize recovery [9]. The beginner's version: the site names the rulebook before the grade does — the tibia earns a ladder, the femoral neck earns a specialist.

The under-fuel door: the risk file the whole family funnels into

Ask what actually predicts who breaks, and the answer runs through food before it runs through shoes. Bennell's risk-factor census — same cohort, twelve months, DXA-measured bone and soft tissue, menstrual and dietary questionnaires, a physical therapist's biomechanical assessment — found no risk factor that predicted stress fractures in men at all, and in women a specific ledger: lower bone density, a history of menstrual disturbance, less lean mass in the lower limb, a discrepancy in leg length, and a lower fat diet; the best independent predictors were age of menarche and calf girth, and the bivariate model correctly assigned 80% of the female athletes into fracture or non-fracture groups [5]. Tenforde's adolescent files sharpen it for the youngest lifters and runners: of 748 competitive high-school runners (442 girls, 306 boys) followed prospectively, stress fractures developed in 5.4% of girls (23) and 4.0% of boys (11) — tibia most common in girls, metatarsus in boys — and the multivariate model kept prior fracture as the one robust predictor in both sexes, plus in girls BMI under 19, menarche at 15 or later, and prior gymnastics or dance, while in boys prior basketball participation was protective [11]. The DXA companion file found low bone mass (Z-score ≤ −1) in 14% of the girls and 21% of the boys, flagged BMI ≤ 17.5 kg/m² in both sexes, and noted that girls with menstrual irregularity plus a fracture history were significantly more likely to carry low bone mass [12]. Above all of it sits the ACSM position stand that named the mechanism: the female athlete triad is the interrelationship of energy availability, menstrual function and bone mineral density, where energy availability — dietary intake minus exercise expenditure — is the factor that impairs reproductive and skeletal health, with most effects appearing below 30 kcal per kilogram of fat-free mass per day, and where the first aim of treatment is to increase energy availability, because no pharmacological agent adequately restores the bone loss [13]. De Souza's Triad Coalition consensus turned that physiology into administration — a risk-stratification point system to guide sport participation, clearance and return to play [14] — and the IOC's RED-S consensus extended the whole file beyond women, because the under-fueled male athlete breaks too [15]. Translated to the training question: a second stress fracture is not bad luck until the fuel line has been audited — the recurrence rate that Warden's rulebook flags as this injury's defining feature is, more often than the shoe drawer admits, an energy-availability finding wearing a running injury's uniform [3][13].

The vitamin-D door: a real association with honest nulls

The supplement aisle's proudest claim in this file finally has a proper meta-analysis, and it deserves to be quoted with its own fine print. The Brazilian-led synthesis — 15 studies, 4,183 participants, military and athlete populations — pooled the serum 25-hydroxyvitamin D data: stress-fracture groups ran vitamin D levels 5.82 nmol/L lower than controls (95% CI −10.35 to −1.29), with the association stronger in military personnel (−7.33 nmol/L) and in males (−9.14 nmol/L) — and then the two sentences that make this file honest: the subgroups of female participants and of athletes did not detect a statistically significant difference, and the association was present only where the control groups were already vitamin-D sufficient (−13.53 nmol/L, 95% CI −20.57 to −6.49), with baseline vitamin-D status explaining over half of the between-study heterogeneity (R² = 53.8%) [16]. The translation is narrower than the shelf talk: vitamin D status should be assessed in high-risk populations to guide preventive strategies focused on maintaining optimal levels [16] — a blood test and a correction where deficient, not a talisman against training errors. The Brazilian review's version of the same modesty: studies have suggested calcium and vitamin D supplementation might be helpful, and beyond it, other treatment regimens are not well established [10]. Deficiency is a real, modifiable risk factor; sufficiency is not a license for the mileage jump that actually broke the bone [16][10].

The prevention door: what the trials actually proved (and the 10% rule they retired)

The most cited prevention advice in running is the 10% rule — never raise weekly mileage more than a tenth — and it carries an uncomfortable distinction: one of the only prevention ideas ever tested in a proper randomized trial. GRONORUN randomized 532 novice runners preparing for a recreational 4-mile event into a standard 8-week program or a graded 13-week program based on the 10% rule, and found running-related injury incidence of 20.8% in the graded group versus 20.3% in the standard group (χ² = 0.016, P = .90) — no effect [17]. The honest reading is not "progression doesn't matter" — a graded ramp is still the logical application of everything the continuum door says — but that the 10% rule is not an immunity charm, and injury causation lives elsewhere too [17]. Warden's rulebook organizes where it does live: microdamage accumulates as a function of the load applied to the bone and the bone's ability to resist load, and the first variable is the modifiable one — training-program design, reducing impact-related forces by instructing an athlete to run "softer" or with a higher stride rate, and increasing the strength and endurance of the local musculature: strengthening the calf for tibial bone stress injuries, the foot intrinsics for metatarsal ones [3]. Knobloch's review closes the loop with the part lifters already know: prevention is careful management of training loads, appropriate periodization and rest, gradual progression, resistance training, and nutrition optimization [9] — which is to say, the boring stack. The strength connection is not decoration: muscle is the first shock absorber upstream of bone, calf girth predicted fractures in Bennell's female athletes [5], and the long-game file on lifting strengthening bones is the same physiology running forward instead of broken.

The programming door: what trains, what modifies, what stops

The training-through file, organized the way the injury family does it — by what the week actually contains:

  • Everything that does not load the affected bone never stopped. Upper body, trunk, the healthy leg: day one, unmodified — the continued-training identity that Warden's "most heal with modified loading" quietly relies on [3]. A tibial stress reaction does not cost you a single bench press.
  • The other leg does double duty and does not apologize for it. Single-leg work on the healthy side carries the training frequency, holds the neuromuscular gains, and shrinks the comeback — the left-right tracking logic the family files already use for exactly this window.
  • The engine moves, it does not stop. Zone-2 cycling, swimming, aqua-jogging, the machine variants that route around the site — cardiovascular fitness outlives the impact moratorium, and the athlete who kept the engine returns to a ladder, not a rebuild.
  • Phase one is load control, phase two is pain-free progression. The two-phase program the modern reviews converge on: off impact until walking is quiet, then a graded return-to-impact ladder — walk, walk-jog intervals, continuous easy running, then intensity — with the rungs set by the MRI grade and re-set by morning-after pain, the same traffic-light gauge as training when sore except here the amber means step DOWN a rung [7][9].
  • The grade and the site write the calendar, not the deadline. Arendt's grade-to-return correlation is the reason the ladder is honest [7]; Diehl's high-risk/low-risk split is the reason some files never see a self-run ladder at all — the femoral neck and the navicular are specialist files from the first paragraph [2][4].
  • The recurrence door is the real deadline. The rulebook's own warning — the high recurrence rate of bone stress injuries signals a need to address the underlying causative factors — means the comeback program includes the fuel audit, the deload habits, the calf and foot-intrinsic strength work, and the vitamin-D blood test, not just the jog-walk intervals [3][5][16].
  • What stops: the fences. Femoral-neck and high-risk-site presentations, night pain unexplained by load, a second fracture without a fuel audit, and any plan that puts a grade on a calendar before the MRI gave it one — clinician doors, and nothing in this paragraph outranks them [2][3][13].

The training-log door: what RightRep sees

This file has a metric structure the camera-and-log app can actually hold:

  • The load spike is visible before the fingertip is. Weekly volume, intensity and exercise selection around the onset — the exact "recent change in training" the reviews name as this injury's signature [9] — already dated and rep-counted in the log, reconstructing for the clinician what memory routinely edits.
  • The pain-milestone diary as data, not vibes. Pain with running → pain with walking → pain at night is the continuum running the wrong way, and the reverse is the ladder working [1][9]; session-tagged notes make each rung's promotion an entry, not a feeling.
  • The comeback as a normal block. Walk-jog intervals, graded mileage, the calf and foot-intrinsic accessory work — logged with workout tracking like any other block, with the progressive overload visible as the slowest, most honest line in the file.
  • The left-right ledger while the bone rests. Single-leg work on the healthy side tracked asymmetrically, so the day the injured side rejoins the program, the ledger — not the ego — sets its opening dose.
  • The recurrence audit on the record. The fuel, the sleep, the mileage ramp, the strength accessories: the causative-factor work the rulebook demands [3], with the training history to prove it happened.

The bottom line

  • The continuum: stress reaction → stress fracture → complete fracture — periosteal edema through progressive marrow involvement to frank cortical fracture on MRI; the position on the ladder is measurable and it sets the calendar [1][3][7].
  • The census: 21.1% of competitive track athletes over twelve months, 0.70 fractures per 1,000 training hours, 20% of all injuries; tibia 46%, navicular 15%, fibula 12%; sprinters break feet, distance runners break long bones and pelvises; in college athletes the foot region leads at 44% [4][7].
  • The risk door: in women — lower bone density, menstrual disturbance history, less lower-limb lean mass, leg-length discrepancy, lower-fat diet, with menarche age and calf girth predicting at 80%; in adolescents — prior fracture the one robust predictor in both sexes, BMI under 19 and menarche at 15-plus in girls; in the clinic census — BMI under 21 in women behind tibial stress fractures and spinal injuries [5][6][11].
  • The under-fuel mechanism: energy availability under 30 kcal/kg fat-free mass per day is where reproductive and skeletal health break; treat energy first — no drug restores the bone [13][14][15].
  • The imaging sequence: X-ray first, MRI when the finger still finds a spot — MRI is the gold standard for diagnosis, grading and return-to-sport prognosis; the grade correlates with time to full return [1][7][8][9].
  • The site rules: low-risk sites earn the two-phase ladder — load control, then pain-free progression; high-risk sites (femoral neck above all, then navicular) are specialist files before they are rehabilitation files [2][4][9].
  • The vitamin-D numbers: −5.82 nmol/L pooled (military −7.33, males −9.14), but the athlete and female subgroups found nothing, and the association exists mainly where controls were already sufficient — test and correct, don't charm [16].
  • The prevention honesty: the 10% rule failed its RCT (20.8% vs 20.3%, P = .90 in 532 novices); what the evidence supports is load management, periodization, softer landings, higher stride rate, calf and foot-intrinsic strength, resistance training and nutrition [3][9][17].

Take the bone off impact, keep everything else training, let the grade and the site write the ladder, audit the fuel line before the second fracture — and let the log keep the comeback honest at rightrep.app.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. Fredericson M, Bergman AG, Hoffman KL, Dillingham MS. Tibial stress reaction in runners. Correlation of clinical symptoms and scintigraphy with a new magnetic resonance imaging grading system. American Journal of Sports Medicine, 1995;23(4):472–481. pubmed.ncbi.nlm.nih.gov
  2. Diehl JJ, Best TM, Kaeding CC. Classification and return-to-play considerations for stress fractures. Clinics in Sports Medicine, 2006;25(1):17–28. pubmed.ncbi.nlm.nih.gov
  3. Warden SJ, Davis IS, Fredericson M. Management and prevention of bone stress injuries in long-distance runners. Journal of Orthopaedic & Sports Physical Therapy, 2014;44(10):749–765. pubmed.ncbi.nlm.nih.gov
  4. Bennell KL, Malcolm SA, Thomas SA, Wark JD, Brukner PD. The incidence and distribution of stress fractures in competitive track and field athletes. A twelve-month prospective study. American Journal of Sports Medicine, 1996;24(2):211–217. pubmed.ncbi.nlm.nih.gov
  5. Bennell KL, Malcolm SA, Thomas SA, Reid SJ, Brukner PD, Ebeling PR, Wark JD. Risk factors for stress fractures in track and field athletes. A twelve-month prospective study. American Journal of Sports Medicine, 1996;24(6):810–818. pubmed.ncbi.nlm.nih.gov
  6. Taunton JE, Ryan MB, Clement DB, McKenzie DC, Lloyd-Smith DR, Zumbo BD. A retrospective case-control analysis of 2002 running injuries. British Journal of Sports Medicine, 2002;36(2):95–101. pubmed.ncbi.nlm.nih.gov
  7. Arendt E, Agel J, Heikes C, Griffiths H. Stress injuries to bone in college athletes: a retrospective review of experience at a single institution. American Journal of Sports Medicine, 2003;31(6):959–968. pubmed.ncbi.nlm.nih.gov
  8. Matcuk GR Jr, Mahanty SR, Skalski MR, Patel DB, White EA, Gottsegen CJ. Stress fractures: pathophysiology, clinical presentation, imaging features, and treatment options. Emergency Radiology, 2016;23(4):365–375. pubmed.ncbi.nlm.nih.gov
  9. Knobloch AC, Caulkins BK, Rennicke JC, Do KH, Covey CJ. Bone Stress Injuries in Endurance Athletes: A Review of Risk Factors, Screening and Evaluation Pearls, Preventive Strategies, and Evidence-Based Management Approaches. Current Sports Medicine Reports, 2025;24(9):281–291. pubmed.ncbi.nlm.nih.gov
  10. da Rocha Lemos Costa TM, Borba VZC, Correa RGP, Moreira CA. Stress fractures. Archives of Endocrinology and Metabolism, 2022;66(5):765–773. pubmed.ncbi.nlm.nih.gov
  11. Tenforde AS, Sayres LC, McCurdy ML, Sainani KL, Fredericson M. Identifying sex-specific risk factors for stress fractures in adolescent runners. Medicine & Science in Sports & Exercise, 2013;45(10):1843–1851. pubmed.ncbi.nlm.nih.gov
  12. Tenforde AS, Fredericson M, Sayres LC, Cutti P, Sainani KL. Identifying sex-specific risk factors for low bone mineral density in adolescent runners. American Journal of Sports Medicine, 2015;43(6):1494–1504. pubmed.ncbi.nlm.nih.gov
  13. Nattiv A, Loucks AB, Manore MM, Sanborn CF, Sundgot-Borgen J, Warren MP; American College of Sports Medicine. American College of Sports Medicine position stand. The female athlete triad. Medicine & Science in Sports & Exercise, 2007;39(10):1867–1882. pubmed.ncbi.nlm.nih.gov
  14. De Souza MJ, Nattiv A, Joy E, Misra M, Williams NI, Mallinson RJ, Gibbs JC, Olmsted M, Goolsby M, Matheson G; Expert Panel. 2014 Female Athlete Triad Coalition Consensus Statement on Treatment and Return to Play of the Female Athlete Triad. British Journal of Sports Medicine, 2014;48(4):289. pubmed.ncbi.nlm.nih.gov
  15. Mountjoy M, Sundgot-Borgen JK, Burke LM, Ackerman KE, Blauwet C, Constantini N, Lebrun C, Lundy B, Melin AK, Meyer NL, Sherman RT, Tenforde AS, Klungland Torstveit M, Budgett R. IOC consensus statement on relative energy deficiency in sport (RED-S): 2018 update. British Journal of Sports Medicine, 2018;52(11):687–697. pubmed.ncbi.nlm.nih.gov
  16. Moraes BF, Leitão CVFS, Rodrigues VMS, Dressler HB, Paixão JC, da Silva ABN, de Oliveira GM, Medina G, Mansur NSB, Lintz F, Zambelli R, Souza MLAT. Vitamin D and the Risk of Stress Fractures in Athletes and Military Personnel: A Systematic Review and Meta-analysis. American Journal of Sports Medicine, 2026;54(11):2854–2863. pubmed.ncbi.nlm.nih.gov
  17. Buist I, Bredeweg SW, van Mechelen W, Lemmink KA, Pepping GJ, Diercks RL. No effect of a graded training program on the number of running-related injuries in novice runners: a randomized controlled trial. American Journal of Sports Medicine, 2008;36(1):33–39. pubmed.ncbi.nlm.nih.gov