Medical question classification is an important task in medical question data processing that can help medical Question answering (QA) systems to provide instant responses to users. However, medical questions frequently cover a wide of areas while medical QA systems may not be able to process such complex questions effectively, leading difficulties for users to find accurate and relevant answers. There may also be a data imbalance issue in the distribution of medical question categories, causing medical QA systems may not able to provide accurate answers when faced with questions trained in categories containing small numbers of samples. The Chinese Diabetes Question Classification task aims to automatically classify questions posed by patientusers as a way to help QA systems manage complex knowledge structures in the medical domain, as well as to improve the classification performance of lowfrequency categories of questions by finetuning or domainspecific training on fewersample categories to enhance the capabilities of automated medical QA systems. It also benefit users for easier access to actual and highquality medical knowledge. This shared task was organized with an international conference NCAA 2024, which attracted professional researchers from industry and academia. This paper presents the results of the Chinese Diabetes Question Classification Competition, providing an overview and summary of the technical approaches used by the teams that received the top 3 performance scores in this competition.