REF-1618 · Kwon et al., 2025 · International Journal of Environmental Research and Public Health · Paper · PubMed · Free full text

Youth sport went with more bone and lean mass in young men and less fat in young women: in the Iowa Bone Development Study (328 people measured by DXA and CT at 23), men who played organised sport throughout childhood and adolescence had 377 g more bone mineral, 1.6 kg/m² more lean mass index and stiffer shin bones than men who never did, and those who dropped out still had much of the difference, while women with consistent sport had 1.4 kg/m² less fat mass index (p = 0.04, above the authors' corrected threshold of 0.01). Read in full

Main caveat: Observational: families and children who choose sport may differ in ways not measured (calcium and protein intake, stress and others named by the authors).
moderate evidenceMeasured on non-athletes and recreational peopleRead in full

Kwon and colleagues asked whether organised sport through childhood and adolescence goes with bone, muscle and fat in early adulthood. Sport participation was reported every six months or so between 6 and 17 years, by parents up to 11 and by the participants after, and earlier trajectory analysis had sorted people into consistent participation, drop-out and no participation.

At 23, whole-body DXA gave bone mineral content, lean mass index and fat mass index, and CT of the distal tibia with finite element analysis gave bone stiffness.

After adjustment for age, education, current accelerometer-measured activity and, by outcome, baseline bone, height, childhood BMI percentile or energy intake, men with consistent sport and men who dropped out had more bone mineral, more lean mass and stiffer tibias than men without youth sport.

Women with consistent sport had lower fat mass index and a higher bone-to-lean ratio, but no significant bone or lean difference. High-school "high-power" sport (at least two years of basketball, cheerleading, football, soccer or volleyball) went with more bone mineral in both sexes.

The authors conclude that youth sport brings healthier bone and muscle in men and healthier body fat in women.

Participants

328 Iowa Bone Development Study participants measured at 23 (144 men, 184 women), of 630 in the cohort; included and excluded did not differ significantly by race, sex or maternal education. Men: consistent 56, drop-out 53, none 35. Women: consistent 73, drop-out 66, none 45. About 96% non-Hispanic White.

Protocol

Exposure: organised sport (coached sport or lessons, competitive or recreational) on a questionnaire about 14 times from 6 to 17 years; trajectory groups consistent, drop-out, none. High-school interscholastic sport by type at 17.
Outcomes at 23: whole-body-less-head bone mineral content (BMC), lean mass index (LMI) and fat mass index (FMI) by DXA; BMC-to-lean and lean-to-fat ratios; distal-tibia bone stiffness by CT and finite element analysis.
Adjustment: all models for age, education and accelerometer MVPA at 23; plus baseline BMC and height (BMC), baseline BMC (stiffness), childhood BMI percentile (LMI and ratios), childhood BMI percentile and energy intake (FMI).
Threshold: Bonferroni-corrected α = 0.01 across the five main outcomes.
Power: 62-74% to detect an effect size of 0.5 between groups.

Key Results

Adjusted difference vs no youth sport (β ± SE)Men, consistentMen, drop-outWomen, consistentWomen, drop-out
BMC (g)+377 ± 67, p < 0.01+192 ± 67, p < 0.01+58 ± 41, p = 0.16−46 ± 41, p = 0.26
LMI (kg/m²)+1.6 ± 0.5, p < 0.01+1.5 ± 0.5, p < 0.01——
Tibial stiffness (kN/mm)+112 ± 30, p < 0.01+76 ± 30, p = 0.01——
FMI (kg/m²)——−1.4 ± 0.7, p = 0.04−1.3 ± 0.7, p = 0.08
BMC-to-lean ratio (g/kg)+2.0 ± 0.9, p = 0.02+0.06 ± 0.9, p = 0.95+2.3 ± 0.8, p < 0.01+0.4 ± 0.8, p = 0.60
Lean-to-fat ratio (kg/kg)+0.36 ± 0.19, p = 0.06+0.18 ± 0.19, p = 0.36+0.14 ± 0.09, p = 0.11+0.12 ± 0.09, p = 0.18
Dashes: outcome not modelled for that sex, because it was not significant in the unadjusted comparisons.
Unadjusted means, men (consistent / drop-out / none): BMC 2689 / 2513 / 2280 g; tibial stiffness 715 / 680 / 601 kN/mm.
High-school high-power sport: higher BMC than other or no sport in both sexes unadjusted (men 2703 vs 2325 g, women 1919 vs 1743 g against no sport); after adjustment, men with high-power sport had higher BMC, bone-to-lean and lean-to-fat ratios and stiffness than men with no high-school sport, while in women the BMC difference fell just short (p = 0.06).
Authors' comparison: the effect of consistent youth sport on men's BMC equalled about 170 extra min/day of MVPA at 23, and on women's fat mass about 30 min/day.
Share kept by drop-outs (our arithmetic): 192 of 377 g of BMC (about half), 1.5 of 1.6 kg/m² of LMI and 76 of 112 kN/mm of stiffness (about two thirds).

Practical Application

In a US cohort measured at 23, men who had played organised sport as children and teenagers carried about 377 g more bone mineral, more lean mass and stiffer shin bones than men who never had, and men who dropped out still had much of that difference; in women the clearest differences were less fat and more bone relative to lean mass. For an adult who played football and ran as a child, this is the most concrete lasting trace measured in the base, more bone and muscle in early adulthood, but only to 23 and as an association, since sport was chosen, not assigned.
Soccer counted among the "high-power" sports, and two or more high-school years of such sport went with more bone mineral in both sexes (in women not quite significant after adjustment).
The women's fat result (p = 0.04) and the men's bone-to-lean ratio (p = 0.02) do not pass the authors' own corrected threshold of 0.01, so they are weaker than the abstract makes them sound.
Inference, not tested here: more bone in early adulthood may help a returning runner tolerate impact, but no injury was measured. Bone keeps responding to impact and heavy lifting, and cycling and swimming do not load it (PHYS-37), so what an adult does now still matters.

Limitations

Observational: families and children who choose sport may differ in ways not measured (calcium and protein intake, stress and others named by the authors).
55% of the cohort measured at 23; mostly non-Hispanic White Midwesterners.
Small subgroups, 62-74% power for a medium effect; two headline results miss the corrected threshold.
Timing of drop-out and non-organised activity were not considered.
Outcomes at 23 only; no performance or injury outcomes.
The supplementary table with adjusted high-school sport models was not read.
How this entry was checked and changed
Why this level

the rubric's cohort line, ≥200 (n = 328), with bone and body-composition outcomes; not higher because sport was chosen, not assigned, only 55% of the cohort was measured at 23, and groups shrink to 35-73 per sex; the authors set a Bonferroni-corrected threshold of 0.01, which the women's fat result (p = 0.04) and the men's bone-to-lean ratio (p = 0.02) do not meet