Introduction

Pelvic floor disorders (PFD), particularly urinary incontinence (UI), affect a substantial proportion of women and can influence daily functioning, participation and sport performance [1, 2]. Although age and parity are established risk factors [3, 4], up to one in five nulliparous women of reproductive age report UI [5], indicating contributors beyond obstetric history. In sport, UI is frequently described among women athletes and may be normalised or under-reported despite potential impacts on women’s health and sports careers [6, 7]. Over time, unaddressed pelvic floor symptoms may impair quality of life, hinder athletic performance [8] and contribute to premature dropout from sport [9]. Ultimately, the culture of high-performance sport, characterised by physical intensity, performance pressure and limited pelvic health education, may perpetuate the normalisation and under-reporting of UI among women athletes, delaying timely support and intervention [10, 11]. Against this backdrop, interest has grown in sport-related determinants of pelvic floor health, including mechanical loading, neuromuscular fatigue and training demands [12, 13].

Previous systematic reviews and meta-analyses consistently report that UI is common among physically active women and athletes; however, translating these findings into athlete care remains limited by substantial methodological heterogeneity. Reviews frequently combine populations with differing competitive contexts, training exposures, parity status and sport-specific demands, complicating the interpretation of prevalence estimates. For instance, some syntheses restrict inclusion to nulliparous women [14, 15], whereas others pool nulliparous and parous athletes across wide age ranges, despite parity and age being strong confounders of UI risk [16, 17]. Similarly, elite competitors are frequently combined with recreational exercisers without explicit thresholds for training exposure, competitive level, or performance context [18–20]. Even within sport-specific analyses, samples often encompass wide variation in parity, training dose and experience, constraining inference regarding sport-level risk patterns [21–24]. As a result, reported prevalence estimates vary widely across syntheses, even within ostensibly similar sport categories, limiting comparability and leaving clinicians and performance staff without athlete-specific prevalence estimates to guide screening and prevention strategies.

Importantly, many syntheses have treated sport type or impact classification as key determinants of UI prevalence across athletic populations [6, 21], yet the extent to which these commonly used classifications actually explain between-cohort variability in prevalence has rarely been examined quantitatively. This reliance on nominal sport classifications assumes that sport type meaningfully reflects pelvic floor loading. Although specific high-impact tasks (e.g. jumping and landing activities) are recognised to provoke leakage in certain athletic contexts [6, 12], it remains uncertain whether broad sport labels independently explain between-cohort variability in UI prevalence when examined quantitatively. Such classifications may therefore incompletely capture how pelvic floor stress accumulates across training volume, intensity and recovery patterns in real-world sport participation [25, 26]. When sport impact is defined strictly according to primary studies’ classifications, comparisons between high-impact and low-impact activities are often based on a small number of cohorts [27]. Conversely, broader post hoc sport discipline classifications may attenuate apparent differences between impact categories because of substantial heterogeneity and potential exposure misclassification [20]. Consequently, the field lacks a robust quantitative approach to characterising pelvic floor loading during sport participation, and current interpretations may overemphasise nominal sport labels while overlooking cumulative training exposure and contextual sport factors.

Collectively, these limitations constrain the interpretability and generalisability of existing syntheses and highlight the difficulty of identifying modifiable sport-related factors relevant to pelvic floor health. Contemporary consensus statements and female athlete health frameworks emphasise the need for higher-quality context-specific evidence to inform screening, education and care pathways in high-performance environments [25]. In this context, an umbrella review alone would summarise inconsistent findings without resolving athlete-specific questions or empirically testing whether commonly used sport classifications meaningfully explain variability in UI prevalence across cohorts. Accordingly, this study was designed as an overview of reviews incorporating a de novo systematic review and meta-analysis of eligible primary cohorts, consistent with Cochrane guidance on overviews and outcome-centric synthesis [28–30]. This combined approach allows critical appraisal of the existing review literature while providing a more coherent synthesis of sport-related factors relevant to women athletes.

The aims of this umbrella review and meta-analysis were therefore to (1) synthesise and update evidence on the prevalence of UI among adult women athletes, and (2) examine how sport context (discipline, modality and competitive level), commonly used mechanical load classifications (sport impact) and cumulative exposure (training volume) contribute to between-cohort variability in pelvic floor symptom prevalence.