中国全科医学 ›› 2026, Vol. 29 ›› Issue (14): 1827-1839.DOI: 10.12114/j.issn.1007-9572.2025.0256

所属专题: 内分泌代谢性疾病最新文章合辑

• 论著 • 上一篇    下一篇

糖尿病患者多维就医行为模式及影响因素研究

于杰*(), 吴春香, 蔡宁, 朱小霞, 沈莉   

  1. 200333 上海市普陀区疾病预防控制中心(上海市普陀区卫生健康监督所)慢性病防制科
  • 收稿日期:2025-05-12 修回日期:2025-09-02 出版日期:2026-05-15 发布日期:2026-04-14
  • 通讯作者: 于杰

  • 作者贡献:

    于杰提出主要研究目标,研究的构思与设计,负责研究的实施和撰写论文,最终版本修订,并对论文负责;吴春香、沈莉负责研究实施过程质量控制和指导;蔡宁和朱小霞负责数据采集和整理。

  • 基金资助:
    上海市普陀区卫生健康系统公共卫生专项项目(ptgw202304)

Study on Multidimensional Healthcare-seeking Behavior Patterns and Influencing Factors in Diabetic Patients

YU Jie*(), WU Chunxiang, CAI Ning, ZHU Xiaoxia, SHEN Li   

  1. Department of Chronic Disease Prevention and Control, Putuo District Center of Disease Control and Prevention (Putuo District Health Inspection and Supervision Office), Shanghai 200333, China
  • Received:2025-05-12 Revised:2025-09-02 Published:2026-05-15 Online:2026-04-14
  • Contact: YU Jie

摘要: 背景 我国成年糖尿病患者达1.48亿,疾病负担沉重,患者就医行为作为疾病管理的核心环节,对糖尿病的防控至关重要。目的 探索糖尿病患者多维就医行为模式,分析其与血糖控制及医疗资源利用的关系,为精准管理提供依据。方法 基于2023年上海市普陀区慢性病管理系统中2型糖尿病患者随访信息、普陀区医院信息系统及实验室信息系统中30 509例2型糖尿病患者的随访和诊疗数据,采用潜在类别分析(LCA)识别就医行为分型,通过多分类Logistic回归分析探讨人口学特征、行为特征、临床特征对就医类别的影响;通过多因素Logistic回归分析探讨各就医类别和其他因素对年度血糖达标的影响。结果 LCA将患者就医行为分为4类:专科主导型(n=4 480,14.68%)、社区基础型(n=7 161,23.47%)、社区强化型(n=11 812,38.72%)和综合复杂型(n=7 056,23.13%)。多分类Logistic回归分析结果显示,男性患者选择社区强化型就医行为的倾向性升高(OR=1.133,P<0.001);≥60岁患者更倾向选择社区基础型或社区强化型就医行为(OR=2.117~2.667,P<0.001);病程延长降低社区基础型倾向性(OR=0.983,P<0.001)、增加综合复杂型倾向性(OR=1.041,P<0.001);有糖尿病并发症、糖尿病合并症患者倾向选择社区强化型(OR=1.498、2.506,P<0.001)和综合复杂型(OR=3.865、3.003,P<0.001);曾经吸烟或正在吸烟患者选择综合复杂型的倾向性降低(OR=0.772,P=0.011);规律活动患者选择社区基础型(OR=0.835,P<0.001)、强化型(OR=0.674,P<0.001)及综合复杂型(OR=0.672,P<0.001)的倾向性均降低;糖化血红蛋白检测次数≥5次/年的患者选择社区基础型和强化型的倾向性降低(OR=0.244、0.356,P<0.01)、综合复杂型倾向性升高(OR=1.464,P<0.01);年度血糖控制达标患者选择社区强化型和综合复杂型的倾向性降低(OR=0.926,P=0.048;OR=0.776,P<0.001)。多因素Logistic回归分析结果显示,以专科主导型为参照,社区强化型(OR=0.923,P=0.041)和综合复杂型(OR=0.791,P<0.001)与年度血糖控制达标可能性降低相关,规律活动(OR=1.107,P=0.002)、糖化血红蛋白检测次数≥2次/年(OR=2.891~4.126,P<0.001)、合并高血压可促进年度血糖控制达标,吸烟(OR=0.851,P=0.008)、男性(OR=0.906,P<0.001)、糖尿病并发症(OR=0.790,P<0.001)、糖尿病合并症(OR=0.620,P<0.001)、甲状腺疾病(OR=0.760,P<0.001)及糖尿病病程延长(OR=0.977,P<0.001)与年度血糖控制达标可能性降低相关。结论 2型糖尿病患者就医行为存在显著异质性,老年、病程及并发症是行为模式分化的核心驱动因素。不同就医行为类别血糖控制效果差异显著,需针对专科主导型强化多学科协作、社区型提升综合管理能力、综合复杂型优化资源配置,以实现分层精准干预。

关键词: 2型糖尿病, 就医行为, 潜在类别分析, 医疗资源利用, 上海市

Abstract:

Background

China has 148 million adult patients with diabetes, imposing a substantial disease burden, and patients' healthcare-seeking behaviors, as a core aspect of disease management, play a critical role in diabetes prevention and control.

Objective

Exploring multidimensional healthcare-seeking behavior patterns in diabetic patients and analyzing their associations with glycemic control and healthcare resource utilization to inform precision management strategies.

Methods

Based on the follow-up and diagnosis data of 30 509 patients with type 2 diabetes in Putuo District, Shanghai in 2023, latent class analysis (LCA) was used to identify the classification of medical-seeking behaviors. Multinomial Logistic regression was used to analyze the influence of demographic, behavioral and clinical characteristics on medical-seeking categories. Multivariate Logistic regression was used to analyze the influence of each medical-seeking category and other influencing factors on the annual blood glucose compliance.

Results

LCA classified patients' medical-seeking behaviors into four types: specialist-dominated type (n=4 480, 14.68%), community-based type (n=7 161, 23.47%), community-intensive type (n=11 812, 38.72%), and comprehensive complex type (n=7 056, 23.13%). Multinomial Logistic regression showed that male patients had a significantly increased likelihood of choosing the community-intensive healthcare-seeking behavior (OR=1.133, P<0.001). Patients aged 60 years and above were more likely to select the community-based or community-intensive type (OR=2.117-2.667, P<0.001). Prolonged disease duration was associated with a decreased tendency for the community-based type (OR=0.983, P<0.001) and an increased tendency for the comprehensive complex type (OR=1.041, P<0.001). Patients with complications/comorbidities significantly tended toward the community-intensive type (OR=1.498, 2.506, P<0.001) and comprehensive complex type (OR=3.865, 3.003, P<0.001). Ever or current smokers had a decreased tendency for the comprehensive complex type (OR=0.772, P=0.011). Patients with regular physical activity had decreased tendencies for the community-based (OR=0.835, P<0.001), community-intensive (OR=0.674, P<0.001), and comprehensive complex (OR=0.672, P<0.001) types. Patients with glycated hemoglobin testing frequency ≥5 times/year showed reduced tendencies for the community-based and community-intensive types (OR=0.244, 0.356, P<0.01), but an increased tendency for the comprehensive complex type (OR=1.464, P<0.01). Patients with achieved glycemic control had a decreased tendency for the community-intensive type (OR=0.926, P=0.048) and comprehensive complex type (OR=0.776, P<0.001). Multivariate Logistic regression analysis revealed that, with the specialist-dominated type as the reference, the community-intensive type (OR=0.923, P=0.041) and comprehensive complex type (OR=0.791, P<0.001) were associated with decreased annual glycemic control achievement. Regular physical activity (OR=1.107, P=0.002) and glycated hemoglobin testing ≥2 times/year were protective factors for annual glycemic control achievement (OR=2.891-4.126, P<0.001). Smoking (OR=0.851, P=0.008), male gender (OR=0.906, P<0.001), complications (OR=0.790, P<0.001), comorbidities (OR=0.620, P<0.001), thyroid disease (OR=0.760, P<0.001), and diabetes with prolonged disease duration (OR=0.977, P<0.001) were identified as risk factors for annual glycemic control achievement.

Conclusion

There is significant heterogeneity in the medical-seeking behaviors of patients with type 2 diabetes. Advanced age, disease duration, and complications are the core driving factors for the differentiation of behavioral patterns. The blood glucose control effects vary significantly among different medical-seeking behavior categories. It is necessary to strengthen multidisciplinary collaboration for the specialist-dominated type, enhance comprehensive management capabilities for the community-based types, and optimize resource allocation for the comprehensive complex type, so as to achieve stratified and precise intervention.

Key words: Type 2 diabetes mellitus, Healthcare-seeking behavior, Latent class analysis, Healthcare resource utilization, Shanghai