Reliability of the PREFIT fitness-test battery in Chilean preschoolers

Introduction This study explores the test–retest reliability of the FITness in PREschoolers (PREFIT) battery, including cardiorespiratory fitness 20 m shuttle run test (SRT), upper-body muscle strength (handgrip strength), lower-body explosive strength (standing long jump), speed (4 × 10 m shuttle run), and static balance (one-leg stance), in Chilean preschoolers considering weight status, sex, and age. Methods The study included 171 Chilean preschoolers (5.4 ± 0.7 years). The PREFIT battery was applied using a test–retest design within 7–10 days between evaluations, and weight status was determined using BMI z -scores ( ClinicalTrials.gov : NCT04269135). Statistical analyses included test–retest differences, Lin's concordance correlation coefficient, Cohen's effect size ( d ), standard error of measurement, intraclass correlation coefficient (ICC), and Bland–Altman plots. Analyses were performed using STATA. Results Among the 171 preschoolers (5.4 ± 0.7 years), test–retest reliability ranged from excellent to moderate for handgrip strength (ICC = 0.8502; 95% CI: 0.81–0.90), 4 × 10 m shuttle run test (ICC = 0.7696, 95% CI: 0.71–0.84), and PREFIT 20 m SRT (ICC = 0.6656; 95% CI: 0.58–0.74), but was poor for the one-leg stance test (ICC = 0.46; 95% CI: 0.36–0.63). SEM and MDC95 values indicated that changes of 0.85 kg, MDC = 9,502.35 kg (handgrip strength), ≥0.89 s (4 × 10 m shuttle run test), ≥5.33 laps (20 m shuttle run test), and ≥12.43 cm (standing long jump) exceeded measurement error. Reliability was similar across sex, age, and weight-status groups. Bland–Altman analyses showed proportional bias and heteroscedasticity for the 20 m shuttle run test and one-leg stance. At the same time, the remaining tests demonstrated consistent measurement error throughout the performance spectrum. Conclusion The PREFIT battery is appropriate for field application among Chilean preschool children, except for the one-leg stance test, which demonstrated inadequate reliability and proportional bias. Reporting MDC95 enables the interpretation of significant change within educational and intervention contexts.