Abstract: Background　Currently, the health management of diabetic patients in the community is still facing many challenges, and it is difficult to effectively improve the glycemic control rate, so exploring the factors affecting glycemic control to delay or reverse the development of complications in diabetic patients is important. Objective　Exploring the characteristics of specific categories of glycemic control in community diabetic patients, the influencing factors, and the association with common complications to provide a scientific basis for the management of community diabetic patients. Methods　A questionnaire survey conducted from March 2023 to April 2024 was used to collect demographic characteristics, health education, diseases, and health management of 417 community diabetes patients in three county-level regions of Taizhou City: Qintong Town in Jiangyan District, Jishi Town in Jingjiang City, and Chenbao Town in Xinghua City. The patients' glycemic control was categorized by potential category analysis, and the prevalence of diabetic complications and their influencing factors under different glycemic control categories were investigated by the χ2 test and multivariate logistic regression analysis. Results　Based on the latent class analysis results, the 417 diabetic patients were categorized into three groups: Class 1: the older-age/low-health-literacy group (n=181, 43.41%); Class 2: the middle-age/moderate-health-literacy group (n=158, 37.89%); and Class 3: the younger-age/high-health-literacy group (n=78, 18.71%). Significant differences were observed among the three groups in the prevalence of hypertension, stroke, and neuropathic complications (P<0.05). Specifically, the older-age/low-health-literacy group and the middle-age/moderate-health-literacy group exhibited higher rates of comorbid hypertension, stroke, and neuropathy compared to the younger-age/high-health-literacy group (all P<0.01). Multivariate logistic regression analysis revealed that, compared to the younger-age/high-health-literacy group, having a junior high school education or above (OR=0.256, 95%CI=0.129-0.510, P<0.001; OR=0.355, 95%CI=0.181-0.696, P=0.003) and implementing more than three blood glucose control measures (OR=0.272, 95%CI=0.148-0.499, P<0.001; OR=0.542, 95%CI=0.298-0.986, P=0.045) were protective factors for blood glucose control in the older-age/low-health-literacy group and the middle-age/moderate-health-literacy group, respectively. Conversely, being unmarried/divorced/widowed (OR=3.303, 95%CI=1.208-9.035, P=0.020) was identified as a risk factor for blood glucose control in the middle-age/moderate-health-literacy group.. Conclusion　Glycemic control of diabetic patients in the Taizhou community has obvious categorical characteristics. The distribution of hypertension, stroke, and neurological complications is different among different potential categories, and the influencing factors are glycated hemoglobin, age, BMI, duration of disease, health literacy, education level, marital status, glycemic control measures, etc. Categorical interventions should be carried out according to the characteristics of the different categories of the population.

Key words: Diabetes mellitus, Glycemic control, Latent class analysis

摘要： 背景　当前社区糖尿病患者健康管理仍面临诸多挑战，血糖控制率难以有效提升，探究影响血糖控制因素以延缓或逆转糖尿病患者并发症的发生、发展具有重要意义。目的　探索社区糖尿病患者血糖控制的具体类别特征、影响因素及其与常见并发症之间的关联，为社区糖尿病患者管理提供科学依据。方法　于2023年3月—2024年4月采用问卷调查收集泰州市姜堰区溱潼镇、靖江市季市镇、兴化市陈堡镇共417例糖尿病患者人口学特征、健康教育、疾病以及健康管理情况。采用潜在类别分析方法对患者血糖控制情况进行分类，采用多元Logistic回归分析方法探究不同血糖控制类别下糖尿病患者并发症患病情况及其影响因素。结果　根据潜在类别分析结果，417例糖尿病患者分为3类，类别1：高龄-低健康素养组181例（43.41%）、类别2：中龄-中等健康素养组158例（37.89%）、类别3：低龄-高健康素养组78例（18.71%）。3组高血压、脑卒中以及神经病变并发症患病情况比较，差异有统计学意义（P<0.05）；且高龄-低健康素养组和中龄-中等健康素养组并发高血压、脑卒中以及神经病变比率均高于低龄-高健康素养组（P均<0.01）。多元Logistic回归分析结果显示，与低龄-高健康素养组相比，高龄-低健康素养组和中龄-中等健康素养组受教育程度为初中及以上（OR=0.256，95%CI=0.129~0.510，P<0.001；OR=0.355，95%CI=0.181~0.696，P=0.003）、控制血糖措施数>3项（OR=0.272，95%CI=0.148~0.499，P<0.001；OR=0.542，95%CI=0.298~0.986，P=0.045）是影响糖尿病患者血糖控制情况的保护因素；中龄-中等健康素养组未婚/离婚/丧偶（OR=3.303，95%CI=1.208~9.035，P=0.020）是影响糖尿病患者血糖控制情况的危险因素。结论　泰州市社区糖尿病患者血糖控制情况具有明显的分类特征，不同潜在类别间高血压、脑卒中以及神经病变并发症的分布不同，影响因素有糖化血红蛋白、年龄、BMI、病程、健康素养、受教育程度、婚姻状况、控制血糖措施等，应针对不同类别人群特征情况进行分类干预。

关键词: 糖尿病, 血糖控制, 潜在类别分析