Analysis of Key Driving Factors for Inpatient Service Experience in Obstetrics and Gynecology Specialized Hospitals Based on Random Forest Algorithm.

OBJECTIVE: To analyze the key drivers influencing the service experience of inpatients in obstetrics and gynecology specialty hospitals, thereby providing a reference for further enhancing the inpatient service experience. METHODS: From April to October 2025, a questionnaire survey on inpatient service experience was conducted among 405 inpatients at a tertiary Grade A obstetrics and gynecology specialty hospital. Using the Sklearn package in Python 3.13.3 software, the random forest algorithm was applied to calculate the importance of each questionnaire item. A data-driven importance matrix was constructed based on patients' evaluation scores to conduct key driver analysis. The random forest model exhibited favorable predictive performance (R2=0.783), allowing for the objective derivation of feature importance to replace traditional subjective importance ratings. RESULTS: A total of 365 valid questionnaires were collected, with an effective response rate of 90.12%. The overall score of the inpatient service experience was (3.84 ± 0.18). Through IPA, the 20 experience elements were categorized into four quadrants: 3 items in the urgent improvement zone (pain management, discharge health guidance, meal quality), 4 items in the key strength zone, 2 items in the secondary improvement zone, and 11 items in the maintenance zone. CONCLUSION: Admission/discharge procedures and information, pain management, and meal quality are key drivers of the service experience in obstetrics and gynecology specialty hospitals. The driving logic of patient experience can be summarized as a "three-tier pyramid" model: environmental cleanliness forms the foundation, doctor-patient communication constitutes the middle layer, and pain management, discharge guidance, and meal quality represent the apex.