The prevalence rate of hyperuricemia (HUA) is increasing in China. To control and reduce the incidence has become the main direction of population public health. High-risk groups of endemic HUA can be identified, the effective risk prediction model can be used to carry out health intervention for high-risk groups, which is expected to become an important means of health intervention program effect evaluation, disease prevention and government decision-making.

To investigation the non-genetic independent risk of HUA in the Guiyang Cohort, constructing a risk predictive model of HUA within a 5 year, and provide a tool for the prevention and screening of HUA in the Guiyang Cohort.

A retrospective study was conducted to collect a total of 2 926 cohort of HUA from 2019 to 2023 in affiliated hospital of Guizhou Medical University in Guiyang. The information covered basic information and testing index of blood. The participants investigate the prevalence of HUA by gender and age; and the basic situation of HUA patients were compared; the independent risk factors for HUA was analyzed by Cox regression, and construced a predictive model risk of HUA for non-genetic factors by the non-genetic risk scoring (NGRS) , The diagnostic and predictive efficacy was assessed by using receiver operating characteristic (ROC) curves.

The independent risk factors for HUA were sex, obesity, hypertension, hypertriglyceridemia and hypercholesterolemia by Cox regression (P<0.05); HUA risk prediction equation was established. Logist P (non-elderly population) =-1.206+2.132×(0.959X₁+1.441X₃+2.383X₄+2.892X₅+1.521X₆+1.384X₇), Logist P (elderly population) =-3.102+2.114×(0.959X₁+1.441X₃+2.383X₄+2.892X₅+1.384X₇). Remark: X₁, X₃, X₄, X₅, X₆,X₇ were gender, obesity, hypertension, hypertriglyceridemia, hypercholesteremia and uric acid reference range, the AUC (95%CI) of the two models were 0.88 (0.77-1.18) , P=0.001, and 0.89 (0.77-1.24), P=0.001; sensitivity and sepcificity were 92.1% and 89.5%, 77.8% and 78.6%. Comparing the predictive effectiveness of risk prediction model for HUA, the AUC (95%CI) of the validation and model were 0.87 (0.76-1.16) and 0.88 (0.77-1.18) in elderly, the AUC (95%CI) of the validation and model were 0.88 (0.77-1.22) and 0.89 (0.77-1.24) in non-elderly, the results showed little difference in fitting AUC with the model, and the internal validation significantly effect.

The results of cohort study showed that men, obesity, hypertension, hypertriglyceridemia, hypercholesteremia, high-normal glucose uric acid were independent risk factors for HUA. The risk prediction model of nongenetic factors for HUA established amang the non-elderly and the elderly based on Cohort.

我国高尿酸血症（HUA）患病率日益增加，控制并降低发病率已成为我国公共卫生事业的主要方向。通过及时识别地方性HUA高危人群，运用有效的风险预测模型能够针对高危人群开展健康干预，有望成为卫生干预方案效果评价、疾病预防以及政府决策的重要手段。

探讨人群HUA的非遗传独立危险因素，并建立5年内HUA风险预测模型，为贵阳市人群HUA的预防和筛查提供工具。

选取2019—2023年贵阳市贵州医科大学第一附属医院体检人群队列为研究对象，共纳入2 926人。收集人群基本信息与血液检测指标，分别按性别和年龄调查HUA患病率，比较分析HUA患病人群基本情况；采用Cox比例风险回归分析HUA的独立危险因素；运用不同组合遗传风险评分（NGRS）方式建立HUA风险预测模型，并采用受试者工作特征曲线（ROC）评价模型预测能力。

通过Cox比例风险回归分析得到性别、肥胖、高血压、高甘油三酯血症、高胆固醇血症为HUA的独立危险因素；建立了HUA非遗传风险预测模型，分别为：Logist P（非老年人群）=-1.206+2.13×（0.959X₁+1.441X₃+2.383X₄+2.892X₅+1.521X₆+1.384X₇），Logist P（老年人群）=-3.102+2.114×（0.959X₁+1.441X₃+2.383X₄+2.892X₅+1.384X₇）。其中，X₁、X₃、X₄、X₅、X₆、X₇分别为性别、肥胖、高血压、高甘油三酯血症、高胆固醇血症、尿酸参考值；两种模型ROC曲线下面积（AUC）（95%CI）分别为0.88（0.77~1.18），P=0.001及0.89（0.77~1.24），P=0.001；灵敏度和特异度分别为92.1%和89.5%，77.8%和78.6%；HUA风险预测模型预测效力，老年人群体验证与模型的AUC（95%CI）分别为0.87（0.76~1.16）和0.88（0.77~1.18），非老年人群体验证与模型的AUC（95%CI）分别为0.88（0.77~1.22）和0.89（0.77~1.24），结果均与原模型拟合AUC差异性较小，模型内部验证效果较好。

研究结果提示性别、肥胖、高血压、高甘油三酯血症以及高胆固醇血症是HUA的独立危险因素；基于队列研究，在非老年人群及老年人群中成功建立了HUA风险预测模型。