Prediction of postpartum pelvic floor dysfunction with a nomogram model based on big data collected during pregnancy

BACKGROUND Pregnancy and childbirth are the main causes of pelvic floor dysfunction (PFD). Although pelvic floor muscle tension is typically measured at 42 days postpartum to assess the severity of PFD and provide timely rehabilitation, it is still impossible to predict PFD and take targeted preventive measures in clinical practice. A PFD prediction model based on big data obtained in prenatal check-ups was established in this study to allow the formulation of personalized preventive strategies to reduce the incidence of PFD. METHODS A total of 1,500 women who underwent regular prenatal checkups and examinations for PFD at 42 days postpartum at the Zhuji Maternal and Child Health Hospital between May 2015 and May 2020 were selected. The data from 1,000 of them were selected as the training cohort, and the data from 500 of them were used as the validation cohort. The women were divided into a PFD group and a non-PFD group according to whether PFD was diagnosed at 42 days postpartum. A nomogram prediction model was created using the influencing factors that lead to PFD, and the discrimination and calibration of the nomogram were evaluated through internal and external validation. RESULTS A total of 389 cases (38.9%) of PFD were included in the training cohort. Multivariate analysis showed that age (odds ratio (OR) =1.896, P