The home physical activity environment as a potential mediator of the association between parental physical activity and preschoolers’ physical fitness: a cross-sectional study in urban China

Introduction Family-level associations with preschool children’s physical fitness remain incompletely understood. Objective This study examined cross-sectional associations among parental physical activity, the home physical activity environment (HPAE), and preschool children’s fitness, including a potential indirect association through the HPAE. Methods We recruited 568 parent–child dyads from 23 kindergartens in Nanchang, China, using multistage stratified random sampling. Parental activity was assessed using the International Physical Activity Questionnaire-Short Form, the HPAE using a purpose-developed 18-item scale, and children’s fitness using an adapted Chinese National Physical Fitness Measurement framework. Mediation analysis used 5,000 bootstrap resamples. The primary model controlled for parental education and family income. A sensitivity model additionally controlled for children’s outdoor activity, organized sport participation, and screen time. Results In the primary model, parental activity was positively associated with children’s fitness (path c, β = 0.167) and the HPAE (path a, β = 0.349), which was also associated with fitness (path b, β = 0.234; all p < 0.001). The indirect association was statistically significant (a × b = 0.081, 95% CI [0.055, 0.109]), and the direct association remained significant (path c′, β = 0.085, p = 0.006). The 48.9% mediated proportion applies only to the primary model. After additional adjustment for children’s outdoor activity, organized sport participation, and screen time, the estimated indirect association was close to zero and was not statistically significant (a × b = −0.002, 95% CI [−0.013, 0.007]). Conclusion The primary model showed a pattern of associations consistent with a potential pathway through the HPAE, but the indirect association was not robust to adjustment for child activity variables. These cross-sectional findings do not establish a mediation mechanism. Longitudinal and experimental studies are needed to clarify temporal ordering and evaluate potential intervention targets.