Top health service concerns: a data mining study of the Shanghai health hotline.

Objective: Our study aims to explore the health service issues of public concern through analyzing the basic characteristics of callers and information from the health hotline in Shanghai. The findings of this study will provide a reference to relevant government departments and assist the government in optimizing the allocation of health resources. Methods: Our research utilized 16,962 original work orders from the 12,320 health hotline, collected since 2015. We applied natural language processing (NLP) to analyze the content of these work orders, facilitating effective text mining and information extraction. Initially, we performed data cleaning to remove irrelevant information and protect caller privacy by anonymizing personal details. This cleaned data was then organized into a structured database for further analysis. Using text mining, we examined various aspects of the calls, including duration, purpose, and topics discussed, to identify patterns and themes that emerged. Results: The calls were categorized into four main groups: complaints, suggestions, inquiries, and requests for assistance. Complaints were the most frequent category, totaling 8,669 (51.11%), followed by help-seeking at 3,335 (19.66%), consultations at 2,727 (16.08%), and comments and suggestions at 1,484 (8.75%). The analysis revealed that men made 6,689 (56.88%), surpassing the 5,071 (43.12%) from women. Additionally, calls from parents numbered 2,126 (56.84%), slightly exceeding the 1,614 (43.16%) from children. The top 10 health service concerns identified in Shanghai included medical staff attitudes, medications, fees, registration, family planning, medical disputes, ambulance services, environmental health, illegal medical practices, and immunization. Conclusions: This study not only identifies critical issues within the Shanghai health service system but also offers actionable insights to inform targeted policy interventions. The high volume of complaints regarding service attitudes and medical expenses underscores the need for stronger policies to improve patient-provider communication and ensure transparency and fairness in healthcare costs. Additionally, the data reveals considerable public concern about the availability and quality of medical services, suggesting that existing policies on resource allocation and service delivery may not adequately meet population needs. The methodologies employed here can be applied to other urban health contexts, providing a valuable framework for improving public health strategies globally.