华南预防医学 ›› 2026, Vol. 52 ›› Issue (9): 1006-1011.doi: 10.12183/j.scjpm.2026.1006

• 论著 • 上一篇    下一篇

新疆喀什疏附县维吾尔族健康体检人群高血压现况及风险预测列线图模型构建

木亚斯尔·麦麦提1, 徐素琪1, 穆尼热克孜·买买提依明2, 吴雪霁3, 张周斌1,3   

  1. 1.南方医科大学公共卫生学院,广东 广州 510515;
    2.新疆喀什地区疏附县疾病预防控制中心;
    3.广州市疾病预防控制中心(广州市卫生监督所)
  • 收稿日期:2026-01-15 发布日期:2026-10-09
  • 通讯作者: 张周斌,E-mail:zhang_zhoubin@163.com
  • 作者简介:木亚斯尔·麦麦提(1997—),女,硕士研究生,研究方向为公共卫生;徐素琪(2000—),女,在读硕士研究生,研究方向为公共卫生;木亚斯尔·麦麦提与徐素琪同为第一作者
  • 基金资助:
    广州市科技计划项目(202103000073)

Prevalence of hypertension and development of a risk prediction nomogram among the Uyghur population undergoing health examinations in Shufu County, Xinjiang

Muyasier Maimaiti1, Xu Suqi1, Munirekezi Maimaitiyiming2, Wu Xueji3, Zhang Zhoubin1,3   

  1. 1. School of Public Health, Southern Medical University, Guangzhou, Guangdong 510515, China;
    2. Shufu Center for Disease Control and Prevention, Kashgar, Xinjiang;
    3. Guangzhou Center for Disease Control and Prevention (Guangzhou Health Supervision Institute)
  • Received:2026-01-15 Published:2026-10-09

摘要: 目的 分析新疆喀什疏附县维吾尔族健康体检人群高血压患病情况及影响因素,构建发生高血压的风险预测列线图模型,为当地高危人群早期识别及高血压精准防控提供科学依据。方法 选取2021年在疏附县进行健康体检的维吾尔族成年人群作为研究对象,采用新疆全民健康体检问卷进行问卷调查并收集体格检查数据,描述性分析高血压患病情况,χ2检验、多因素logistic回归方法分析其影响因素。采用R 4.3.1构建发生高血压的风险预测列线图模型,Hosmer-Lemeshow拟合优度检验评估该列线图模型的拟合情况,ROC曲线评估该列线图模型的区分度,校准曲线评估该列线图模型的校准度。结果 本研究共纳入119 869名健康体检者,共检查出13 183例高血压患者,检出率11.0%。多因素logistic回归分析结果显示年龄增加(30~<40岁OR=2.780;40~<50岁OR=10.879;50~<60岁OR=32.666;60~<70岁OR=69.929;≥70岁OR=111.665)、文盲(OR=1.306)、农牧民(OR=1.202)、有高血压家族史(OR=1.524)、腹部肥胖(OR=1.267)及高BMI(24.0~<28.0 kg/m2 OR=1.325;≥28.0 kg/m2 OR=2.294)是高血压的危险因素(均P<0.01)。构建发生高血压的风险预测列线图模型,该模型拟合情况较好(P>0.05),ROC曲线下面积为0.859,模型预测高血压发生率与实际发生率基本一致。结论 新疆喀什地区疏附县维吾尔族健康体检人群的高血压患病率处于较低水平。患病风险与高年龄、低教育水平、农牧民、高BMI、腹部肥胖及高血压家族史等因素密切相关,基于此构建的列线图风险预测模型有助于当地医疗机构及医护人员区分高血压的高危人群并指导制定干预措施。

关键词: 健康体检, 高血压, 影响因素, 预测模型

Abstract: Objective To analyze the prevalence and influencing factors of hypertension among the Uyghur population undergoing routine health examinations in Shufu County, Kashgar Prefecture, Xinjiang, and to construct a nomogram risk prediction model for incident hypertension, thereby providing an empirical foundation for the early identification of high-risk cohorts and the precision prevention and control of hypertension in the region. Methods Uyghur adults who underwent health examinations in Shufu County in 2021 were selected as the study cohort. Data were acquired utilizing the Xinjiang Universal Health Examination Questionnaire alongside standardized physical examination metrics. The prevalence of hypertension was evaluated using descriptive statistics, while its influencing factors were analyzed via chi-square tests and multivariate logistic regression. A nomogram risk prediction model for hypertension was developed utilizing R software (version 4.3.1). The goodness-of-fit of the model was assessed employing the Hosmer-Lemeshow test, discriminatory capacity was evaluated via Receiver Operating Characteristic (ROC) curves, and calibration was examined utilizing calibration plots. Results The present study enrolled a total of 119 869 participants, among whom 13 183 cases of hypertension were identified, yielding a prevalence rate of 11.0%. Multivariate logistic regression analysis indicated that advancing age (30-<40 years: OR=2.780; 40-<50 years: OR=10.879; 50-<60 years: OR= 32.666; 60-<70 years: OR=69.929; ≥70 years: OR=111.665), illiteracy (OR=1.306), agricultural or pastoral occupation (OR=1.202), a family history of hypertension (OR=1.524), abdominal obesity (OR=1.267), and an elevated Body Mass Index (BMI) (24.0-<28.0 kg/m²: OR=1.325; ≥28.0 kg/m²: OR=2.294) constituted significant risk factors for hypertension (all P<0.01). The constructed nomogram risk prediction model demonstrated satisfactory goodness-of-fit (P>0.05), with an Area Under the ROC Curve (AUC) of 0.859. The model's predicted incidence of hypertension demonstrated robust concordance with the observed actual incidence. Conclusion The prevalence of hypertension among the Uyghur population in Shufu County, Kashgar Prefecture, Xinjiang, remains comparatively low. The risk of developing hypertension is significantly associated with advanced age, low educational attainment, agricultural or pastoral occupations, elevated BMI, abdominal obesity, and a familial predisposition to the condition. The nomogram risk prediction model derived from these variables provides a reliable tool for local healthcare institutions and practitioners to stratify high-risk populations, thereby informing the formulation of targeted clinical interventions.

Key words: Health examination, Hypertension, Influencing factors, Predictive model

中图分类号: 

  • R195.4