Analysis of the interaction effects among the risk factors for hyperuricaemia in adults: a cross-sectional survey in Guilin, China

作者：Hu, Mengyan; Liu, Jinbo; Zhou, Chunhua; Li, Xinli^*

来源：International Journal of Clinical and Experimental Medicine, 2017, 10(6): 9376-9386.

• Hyperuricaemia
• risk factor
• interaction
• classification and regression tree model

In recent years, the prevalence of hyperuricaemia has risen, especially in younger populations. This study was designed to explore the risk factors and their interactions of hyperuricaemia for establishing effective preventive measures to reduce the prevalence of hyperuricaemia. We conducted a cross-sectional survey of 6241 adults selected with a cluster sampling method. Survey items included a questionnaire survey, physical measurements, chemistry parameters, and liver Doppler ultrasonography. We analysed the metabolic characteristics of subjects with hyperuricaemia, and developed the CRT model to screen the risk factors for hyperuricaemia and the interactions among the risk factors. The non-conditional logistic regression model was used in combination with the additive model to quantitatively analyse the interactions among the risk factors. A total of 1035 individuals were found to have hyperuricaemia, and were included in the high uric acid (HUA) group (men: 755, women: 280), the overall prevalence of hyperuricaemia was 16.6%. The levels of systolic blood pressure, diastolic blood pressure, body mass index (BMI), uric acid, fasting plasma glucose (FPG), triglycerides (TG), total cholesterol, low-density lipoprotein cholesterol (LDL-C) and the prevalence of non-alcoholic fatty liver disease (NAFLD) were significantly higher in the HUA group than in the control group (P < 0.01). The CRT model selected 7 explanatory variables: age, gender, BMI, TG, LDL-C, FPG, and NAFLD; among these, TG, BMI, and NAFLD were most closely related to hyperuricaemia. The multivariate logistic regression analysis showed that the prevalence of hyperuricaemia was 3.957-fold higher in individuals with high TG and high BMI than in those with low TG and low BMI (OR = 3.957, CI: 3.129 similar to 5.004), and the attributable percent was 9.17%. The results suggested that individuals with hyperuricaemia had multiple metabolic abnormalities; age, gender, BMI, TG, LDL-C, FPG, and NAFLD were closely related to hyperuricaemia. Moreover, the positive interaction between high TG and high BMI was one of the most important risk factors for hyperuricaemia, and the co-presence of high TG and high BMI significantly increased the risk of hyperuricaemia.

• 出版日期2017
• 单位苏州大学