Large language model-assisted machine learning for postpartum urinary incontinence prediction: Model development and validation study - Scorecard - MDSpire
Clinical Scorecard: Development and Validation of a Machine Learning Model Using Large Language Model Assistance to Predict Postpartum Urinary Incontinence
At a Glance
Category
Detail
Condition
Clinical topic addressed by the source article: Development and Validation of a Machine Learning Model Using Large Language Model Assistance to Predict Postpartum Urinary Incontinence
Key Mechanisms
The study aims to integrate LLM-assisted extraction of information from unstructured EMR text with machine-learning algorithms to develop a postpartum UI prediction model.
Target Population
Population described in the source article.
Care Setting
Development and external-validation participants were recruited at two tertiary maternity hospital campuses in Hangzhou, China. The outcome was assessed 6–8 weeks after delivery at routine postpartum follow-up.
Key Highlights
The development cohort included 1,243 consecutively recruited pregnant women enrolled from January to June 2020.
External validation included 452 consecutively recruited pregnant women enrolled from January to May 2025 at a second campus serving a relatively independent patient population.
The study used the validated Chinese ICIQ-UI SF to assess urinary leakage 6–8 weeks after delivery; any reported leakage was classified as postpartum UI.
The supplied excerpt describes the study design and methods but does not report model performance or predictive factors.
Guideline-Based Recommendations
Diagnosis
The study measured UI at 6–8 weeks postpartum using the validated Chinese ICIQ-UI SF.
Participants reporting any urinary leakage on the ICIQ-UI SF were categorized as having postpartum UI.
Management
The introduction states that pelvic floor muscle training is recommended as a first-line measure for preventing and treating UI.
Monitoring & Follow-up
The study outcome was assessed at routine postpartum follow-up 6–8 weeks after delivery.
Risks
The introduction reports that postpartum UI is strongly associated with UI later in life.
Reported consequences include somatic discomfort, social embarrassment, diminished work ability, and lower social participation.
The introduction reports an average 45% increase in women’s risk of postpartum depression associated with UI.
Patient & Prescribing Data
Study participants were adults with singleton, term pregnancies recruited at two tertiary maternity hospital campuses in Hangzhou, China.
This is a prediction-model study; the supplied excerpt does not report a prescribed treatment or model-based treatment outcomes.
Clinical Best Practices
The study used the validated Chinese ICIQ-UI SF to assess postpartum urinary leakage.
The study protocol was reported using the TRIPOD+AI statement.
