Statistical analysis
The associations of eating behaviour with symptoms of pelvic floor disorders were analyzed using simple (Model 1) and multiple logistic regression models (Model 2). Model 2 was adjusted with age, BMI, education, physical workload, previous physical activity (age 17–29), current physical activity (MET-h/d), menopausal status, parity, and hysterectomy. Correlation analysis, residual plots and scatter plots between each continuous predictor and the logits values were used for testing the model assumptions. Statistical analyses were performed using R and IBM SPSS Statistics 22.0 (SPSS Inc., Chicago, IL). The level of significance was set at p≤0.05.
RESULTS
Participants’ demographical, gynaecological, and physical activity status in total sample and in different pelvic floor disorder subsamples have been reported previously11. The frequencies of various eating styles among women with different symptoms of pelvic floor disorders are presented in Table S1.
In comparison to women reporting normal eating, women with overeating (OR 1.49, CI 1.14–1.96, p=0.004) and restrictive eating (OR 1.63, CI 1.09–2.44, p=0.017) behaviour were more likely to experience symptoms of stress urinary incontinence (Table 1: Model 1), but these associations attenuated after controlling for confounding factors (Table 1: Model 2). Restrictive eating was also associated with constipation and defecation difficulties in Model 1 (OR 1.90, CI 1.18–3.07, p=0.008) and Model 2 (OR 1.73, CI 1.03–2.90, p=0.039).
No snacking between meals (OR 0.69, CI 0.50–0.95, p=0.022) and grazing throughout the evening (OR 1.59, CI 1.09–2.31, p=0.016) were associated with symptoms of stress urinary incontinence in Model 1 (Table 2). In addition, no snacking between meals (OR 0.43, CI 0.20–0.90, p=0.025) was associated with symptoms of faecal incontinence in Model 1. Women who had reported to have highest food consumption in the evening were more likely to experience symptoms of urge urinary incontinence according to both Model 1 (OR 1.84, CI 1.23–2.76, p=0.003) and Model 2 (OR 2.01, CI 1.32–3.07, p=0.001).
Attempting to maintain healthy eating patterns was associated with symptoms of stress urinary incontinence (OR 0.60, CI 0.36–0.99, p=0.047) and urge urinary incontinence (OR 0.48, CI 0.26–0.88, p=0.018) in Model 1 (Table 3). The association remained statistically significant for symptoms of urge urinary incontinence (OR 0.45, CI 0.24–0.85, p=0.014) when adding confounding factors in Model 2.