Irene Sevilla-Arrabal

Background: Urinary incontinence is frequently reported in women athletes, yet prevalence estimates vary widely and are commonly interpreted using broad sport classifications that may not capture pelvic floor loading. Objective: We aimed to synthesise evidence on urinary incontinence prevalence in adult women athletes and examine whether commonly used sport classifications or reported training exposure explain between-cohort variability. Methods: Five databases were searched (January 2010–September 2025) for systematic reviews reporting urinary incontinence prevalence in women athletes to conduct an umbrella review. Eligible primary observational studies were identified from included reviews and forward citation tracking to conduct a de novo random-effects meta-analysis and meta-regression restricted to adult women athletes (aged 18–45 years). Results: Fourteen systematic reviews were included, most rated as low or critically low confidence. Thirty-two primary studies (120 cohorts; n = 4649) were meta-analysed, all rated as high or moderate quality. Urinary incontinence prevalence showed substantial between-cohort heterogeneity. In meta-regression analyses, prevalence did not differ significantly across sport discipline, impact category, or sport modality. Weekly training volume was the most consistently reported exposure metric and showed a positive association with urinary incontinence odds (6% per additional hour/week; p = 0.011), within the constraints of limited intensity and pelvic floor-specific exposure data, with no evidence of effect modification by impact category or competitive level. Competitive level showed a non-robust directional trend toward higher prevalence among professionals ( p = 0.08). Conclusions: Commonly used sport-label classifications showed limited explanatory value for urinary incontinence prevalence in women athletes when applied as stand-alone proxies for exposure, and may misdirect screening and prevention strategies. Within the current heterogeneous and cross-sectional evidence base, cumulative training exposure emerged as the most consistently reported correlate of urinary incontinence prevalence. These findings highlight the need for prospective studies using standardised training metrics with pelvic floor-specific measures to better characterise a multi-dimensional pelvic floor exposure.

Background: Velocity-based training (VBT) is used to estimate maximal strength and prescribe resistance-training loads, but evidence in untrained youth, especially early-adolescent females, is limited. In untrained early-adolescent females performing free-weight back squats, (1) the load–velocity relationship (LVR) is comparable to adult samples, albeit with greater between-subject variability, and (2) one-repetition maximum (1RM) estimates are affected by the minimum velocity threshold (MVT) anchor. Methods: Thirty-four untrained females (10–14 years) completed two progressive loading tests followed by actual 1RM attempts. Mean propulsive velocity (MPV) was recorded to model LVRs. Three MVTs were considered: (a) Actual (from Test 1), (b) General (0.30 m·s⁻¹), and (c) Optimal (individualized to minimize prediction error in Test 1). LVR-based 1RM estimates from Multi-point and Two-point approaches were generated in Test 2 using each MVT and compared with the actual 1RM. Results: MPV decreased near-linearly with load (median R² ≈ 0.996), from 1.00 ± 0.19 m·s⁻¹ at ~40%1RM to 0.30 ± 0.05 m·s⁻¹ at 100%1RM. Across MVTs, Two- and Multi-point models showed similar 1RM accuracy (≤~0.7% difference; p > 0.35). Actual and General MVTs overestimated 1RM (+5.1 kg; p < 0.001), whereas an individualized Optimal MVT (~0.38 m·s⁻¹) removed bias (+0.6 kg; p = 0.52) and reduced error (p ≈ 0.03). Conclusions: In untrained early-adolescent females, the back-squat LVR is highly linear, and 1RM estimation accuracy hinges on the MVT anchor. A streamlined Two-point LVR paired with an individualized Optimal MVT provides an efficient, accurate workflow for youth strength assessment.