Individual Factors in Young Female Athletes' Bladder Health (the Y-FAB Study).

One-hundred female athletes enrolled and completed the study. One-hundred female non-athletes enrolled, and 98 completed the study. Participants were young (mean age 20±2), predominantly Caucasian (74%), with healthy BMI (mean 23±4). Non-athletes were slightly older (21 vs 20, p < 0.001). Athletes’ sports were predominantly high impact: cross country/track (39%), softball (17%), volleyball (17%), lacrosse (7%), basketball (5%), and others (6% - soccer, dance, hockey, and tennis). The remaining athletes participated in low-impact sports: swimming (9%) and other (2% - golf, water polo). Non-athletes more frequently reported eczema (12% vs 2%), migraine headaches (15% vs 4%), anemia (16% vs 4%), dysmenorrhea (46% vs 19%), and anxiety (50% vs 20%) (p<0.01 for all comparisons). Athletes reported more stress fractures (27% vs 7%, p < 0.001). Non-athletes reported more pain medications and substance use. Athletes consumed lower total fluids (81 vs 101 oz., p = 0.03) but similar water intake (p = 0.16). Bladder health education rates in school were similar, but non-athletes were more likely to have heard of pelvic floor exercises (74% vs 43%, p<0.001). Over half of participants had high social desirability scores with no observed group differences. Table 1 . Prevalent toileting behaviors among athletes and non-athletes included: emptying the bladder before leaving home (95%), delaying emptying the bladder when busy (85%), hovering over the toilet away from home (74%), pushing to empty the bladder faster (64%), avoiding public toilets (60%). Non-athletes were more likely to hold urine until they got home (62% vs 47%, p = 0.04) and empty away from home without feeling the need (19% vs 9%, p = 0.04). Table 2 . TB-WEB summary scales did not differ between groups (p>0.05 for all measures). Table 4 . Overall prevalence of LUTS in athletes and non-athletes included: 59% with any filling symptom, 23% with any incontinence symptom, and 41% with any voiding symptom. Both groups had high LUTS prevalence: frequency (34%), hesitancy (33%), urgency (33%), SUI (15%), and UUI (12%), among others. Athletes were more likely to have urgency (39% vs 27%, p = 0.06) and SUI (20% vs 10%, p = 0.047), and their incontinence sub-score was higher on univariable analysis (2.2±2.2 vs 1.6±2.2, p = 0.03) but this difference attenuated in multivariable analysis. Table 3 . High impact athletes in this study had a lower SUI prevalence of 19.5% (17/87) compared to low impact athletes (swimming, golf), 27.3% (3/11). Compared to athletes, non-athletes reported significantly more impact on quality of life (QOL) in two areas including: cutting down on fluid intake (p=0.02) and overall interference with life (p=0.01). Controlling for age and daily beverage intake, BFLUTS summary scores were 0.6 points higher on the incontinence subscale (95% CI: −0.04, 1.3; p=0.07) and 0.8 points higher on the total score (95% CI: −0.5, 2.1; p=0.25) for athletes, whereas non-athletes had 0.6 point lower adjusted QOL summary scores (95% CI: −1.1, −0.04; p = 0.04). Table 4 .

This cross-sectional survey study was part of a broader study (Social & Environmental Risk Factors for Young Female Athlete Bladder Health: The YFAB Study) and was funded by a supplemental grant through NIH PLUS. High school and collegiate females aged 13 to 23 were included; exclusions were made for parity, non-English speakers, and history of neurogenic bladder. Potential participants were consented (age >18) or assented with parental consent (age <18). Recruitment occurred between 10/29/2019 -6/20/2022. Female athletes from 3 local universities and 3 local high schools were approached for participation. Non-athlete participants were initially recruited from the same schools, but recruitment was subsequently opened to universities across the country. The COVID-19 pandemic and remote learning prompted the change in recruitment for non-athlete participants where the wider recruitment net made timely recruitment feasible. Recruitment, consent, data collection, and objective measurements were collected during a single study visit at the participants’ school for the first 77 athlete participants (prior to the COVID pandemic). During the COVID-19 pandemic, infection control measures necessitated that subsequent data collection be completed virtually through Research Electronic Data Capture (REDCap). The participants who were recruited following onset of the COVID pandemic received an individualized link to complete all survey measures. No adolescent participants were enrolled following onset of the COVID-19 pandemic. Participants received a $25 gift card. Demographics included: age, body mass index (BMI), age at menarche, sexual activity, medical problems, history of dysmenorrhea, history of urinary tract infection (UTI), history of recurrent UTI, history of pelvic floor muscle exercises or physical therapy (PT), and knowledge about bladder health. Participants were asked to self-report their medical history including free text and checkbox items for items prevalent in this population or pertinent to the pelvic floor. Independent, validated outcomes measures were used. Toileting behaviors were assessed via the Toileting Behaviors: Web-Based Toileting Behavior (TB-WEB, 18-items, each item scored 1: never to 5: always), which assesses five categories of problematic toileting behaviors (premature voiding, delayed voiding, voiding in atypical positions, straining to void, and place preference when voiding). 12 TB-WEB subscales are calculated as the sum of the Likert-scale items, with total scores ranging from 18-90. LUTS were identified via the Bristol Female Lower Urinary Tract Symptoms (BFLUTS, 19-items, each item scored 0 to 4), which assesses five domains: filling, voiding, incontinence, sexual dysfunction, and quality of life. 13 BFLUTS subscales are calculated as the sum of the Likert-scale items, with total scores ranging from 0-48. Note that quality of life (QOL) is not included in the BFLUTS total score. For this study, “bothersome symptoms” were defined as >/= 2 points per symptom. Fluid intake (total volume of fluid across 19 different beverages and overall in 24 hours) was measured via the Beverage Questionnaire (Bev-Q, 19-items). 14 , 15 The Marlow-Crowne Social Desirability Scale (33-items) assessed whether participants were prone to providing socially desirable answers on questionnaires. 16 – 18 A prior study 19 informed effect sizes for sample size estimations. Assuming a sample size of 200 (50% athletes) and r-squared = 0.05 for covariate adjustment, the study has 80% power to detect differences of at least 1.2 points on BF-LUTS (SD=3) and 3.6 points on TB-WEB (SD=9). Lacking data on minimally important clinical differences at the time of our study design, effect sizes proposed here (Cohen’s d = 0.4) are small to moderate by conventional interpretations. Descriptive statistics were presented for participant characteristics. Differences between athletes and non-athletes were assessed using two-sample t-tests for age and BMI, Wilcoxon rank sum tests for beverage intake, and chi-square or Fisher’s exact tests for all other comparisons. Statistically significant prevalence differences of LUTS and toileting behaviors were assessed using chi-square or Fisher’s exact tests. Summary scales were compared using two-sample t-tests for univariable comparisons and linear regression models (adjusted adjustment for age and daily total beverage intake) in multivariable comparisons. Analyses were performed using SAS 9.4 (SAS Institute, Cary, NC).

This study found evidence of impaired bladder health (toileting behaviors and LUTS) in young females, regardless of athletic status. Some may argue the prevalence of toileting behaviors and LUTS is representative of “normal” behaviors or symptoms in this population. However, premature voiding, straining, and delayed voiding, and hovering over the toilet when voiding away from home align with toileting behavior domains labeled problematic due to associations with “bothersome” LUTS in prior research. 21 Toileting behaviors may serve as an individual risk factor for LUTS and impaired bladder health via weakened pelvic floor muscles, increased bladder neck descent, impaired bladder sensation, and bladder over-distension. 24 , 25 Toileting behaviors and LUTS were prevalent regardless of athletic participation, consistent with prior research showing similar findings in female medical trainees 19 , 22 and female nurses. 23 This affirms that female gender is an individual risk factor for impaired bladder health. Toileting behaviors were broadly similar, contrary to the study’s hypothesis, although athletes were less likely to hold urine until they got home, perhaps due to being away from home for longer periods. LUTS were prevalent among all participants, but female athletes had higher rates of SUI and urgency. Participation in high impact sports is an individual risk factor for SUI. 5 , 26 – 28 In contrast, participation in low-impact moderate intensity exercise is associated with lower UI overall 29 and SUI. 26 The SUI prevalence in our study was 20% in the athlete group (high and low impact sports), slightly lower than prior published prevalence (30%). 5 , 6 , 8 Additionally, our study found a higher rate of SUI in low impact athletes (27.3%) compared to high impact athletes (19.5%), contrary to prior studies. However, the percentage of low impact athletes in this study was small (11%) and a larger sample of low impact athletes may have given a better estimate of LUTS. The athlete population in this study did not include gymnasts, a population known to have the highest SUI rates 5 , 26 – 28 , which may explain the lower overall SUI prevalence compared to prior published estimates. Other LUTS have not been well described in young females, with data suggestive that0.5% of older adolescents (16-18 years old) have overactive bladder (OAB) 7 and 26% of high school and collegiate female athletes have UUI. 6 Our study found that LUTS overall were more prevalent than previously reported including one-third of all young females reporting urgency-frequency and hesitancy symptoms, and one-fifth reporting UUI. Sports participation is one individual risk factor that contributes to LUTS (SUI specifically), but female gender and problematic toileting behaviors are individual risk factors that may bridge the LUTS gap in young females regardless of athletic status. Interestingly, the non-athletes reported higher impact to their QOL compared to athletes including with cutting down fluid intake and overall interference with life. Athletes had overall lower reported daily fluid intake volume compared to non-athletes. It is possible fluid intake restriction was underestimated on the Bev-Q in the athlete group. It may be that athletes have less time to worry about these symptoms due to sporting commitments or may have other physical symptoms related to sport participation that outweigh their LUTS. Other QOL symptom domains did not differ between groups. Further research on the impact of LUTS on QOL in young women would help further understand the overall QOL impact in athletes and non-athletes. The racial/ethnic demographics of these healthy, young female athletes and non-athletes were representative of the general US population as of 2023. 30 The athletes were generally healthier and used fewer substances but had higher rate of stress fractures, consistent with high-level participation in competitive sport. The Female Athlete Triad 31 (an interrelationship of menstrual dysfunction, low energy availability through diet or exercise, and decreased bone mineral density in female athletes) has been associated with pelvic floor disorders, including UI, in young female athletes and represents an individual risk factor for impaired bladder health outside of impact sport participation. 26 , 32 , 33 Further investigation of this was outside of the scope of this paper but will be presented in a future analysis. Athletes in this study were more likely to restrict overall fluids (though water intake was similar), counterintuitive to the demands of competitive sport. This has been demonstrated in older female athletes as a way to mitigate LUTS and continue sport participation. 34 – 36 All females in this study had low rates of bladder health education, suggesting there is substantial opportunity to improve education in this population (in line with the PLUS Consortium’s call to action). 37 A prior study demonstrated improvement in pelvic health knowledge with a 6-day pelvic health education program. 38 Bladder health education and reduction of LUTS at a young age may have far-reaching social and financial implications. Financial burden for SUI, UUI, and OAB exceeded $100 billion annually in the US as of 2020. 39 , 40 LUTS also have high social/emotional 41 and physical tolls including poor health-related quality of life 42 , impaired mobility 43 , and institutionalization 44 in older adults. These burdens and stigma 6 , 10 , 11 might be negated through earlier school- and clinic-based screening and education. This study had many strengths. To our knowledge, it is the first to compare fluid intake, toileting behaviors, and LUTS in young female athletes and non-athletes. It adds to a growing body of evidence that LUTS start at a young age and supports enhanced early education and screening. However, it does have several limitations. While the demographics were representative of the US population, the study was performed in a small geographic region and may not be generalizable. The participants in this study were community dwelling high school and collegiate young women who did not have differences in their baseline demographic data. However, not all covariates could be considered, and future longitudinal prospective studies should further assess causality while addressing potential confounding. Baseline social determinants of health (including parental education, family socioeconomic status), rigor of academic studies, and the overall activity level of the non-athlete group may be important factors to consider as well. Collection of demographic information and use of validated survey-based outcomes measures are also susceptible to bias. With the exception of the Bev-Q, the other outcomes measures utilized in this study (TB-WEB, BFLUTS, and Marlowe-Crowne Social Desirability Scale) have not been validated in adolescent participants. While there was no difference between athletes and non-athletes regarding their social desirability scores, both groups scored in the “average” to “high” domains of social desirability, indicating they may be prone to influence in their survey responses. Given bladder health stigma, it is possible that socially desirable responses to the BFLUTS and TB-WEB could be falsely low in this study. The effect of the pandemic amid study enrollment cannot be understated, though its effect cannot be directly measured. Finally, while this study was not formally powered for an equivalency design, only small differences were observed across TB-WEB items and subscales. Conversely, the BFLUTS total demonstrated a signal towards more incontinence for athletes, but more power and a larger sample size may be needed to precisely measure this difference while controlling for confounders. In conclusion, high school and collegiate females are an at-risk population for impaired bladder health with high rates of problematic toileting behaviors and LUTS, and low rates of bladder health education. Female athletes were more likely to restrict fluids and had higher rates of SUI and urgency compared to non-athletes. Future analyses will delve deeper into the individual risk factors in female athletes and will also report on interpersonal and institutional risk factors in the general population of young females.

The National Institutes of Health Prevention of Lower Urinary Tract Symptoms (NIH PLUS) Consortium’s mission is “to identify promising strategies for promoting bladder health and reducing lower urinary tract symptoms (LUTS)… in women throughout the lifespan.” 1 PLUS defines bladder health as a “complete state of physical, mental, and social well-being related to bladder function” 2 , 3 and defines LUTS as 1) Storage/Filling Symptoms : frequency, urgency, nocturia, and urinary incontinence (UI) subtypes including: stress UI (SUI), urgency UI (UUI), mixed UI (MUI), postural UI, continuous UI, coital UI, as well as post-void dribbling, nocturnal enuresis, and bladder pain; 2) Emptying Symptoms: hesitancy, slow stream, spraying/splitting, intermittency, retention, persistent urgency, need to re-void, incomplete bladder emptying, dysuria, and post-void dribbling; and 3) Bioregulatory Symptoms : disruption in the protective bladder barrier, dysbiosis, and/or infection. 2 Impaired bladder health and LUTS have been attributed to aging and medical factors. 4 – 7 SUI is the most common LUTS in healthy high school and collegiate female athletes (overall prevalence estimated around 30%). 5 , 6 , 8 Yet, bladder health in young females is not routinely screened in primary care settings, nor on pre-participation sports clearance examinations (though one tool has been proposed 9 ). Bladder health stigma in women of all ages 6 , 10 , 11 may contribute to lack of screening, education, and research. High SUI prevalence, lack of bladder health screening, and social stigma are somewhat established in young female athletes, but the individual, social, and environmental factors that impact bladder health in young female athletes are currently unknown. PLUS suggests these factors are an important part of the bladder health conceptual framework, 1 and guided this study, which aimed to assess individual, interpersonal, and institutional risk factors for impaired bladder health in female athletes compared to female non-athlete peers ( Figure 1 ). This study aimed to investigate individual factors (toileting behaviors and fluid intake) and bladder health (LUTS) in this population and hypothesized that female athletes would have significantly more problematic toileting behaviors, LUTS, and fluid intake restriction due to the unique demands of sports on their bodies.