Risk factor analysis of insufficient fluid intake among urban adults in Wuxi, China: a classification and regression tree analysis

Abstract

Abstract Background Dehydration due to insufficient fluid intake (IFI) is detrimental to health. This cross-sectional study aimed to assess the fluid intake of urban adults in Wuxi, China, and to identify potential risk factors contributing to IFI. Methods Adults were selected from the urban area of Wuxi, China, using a multiple-stage random sampling method. The fluid intake information was obtained with a 24-h self-reported diary over seven consecutive days in both summer and winter of 2015. A classification and regression tree (CART) analysis was conducted to detect the potential risk factors associated with IFI. CART is a machine-learning algorithm that portions the data into subsets by threshold. Results A total of 584 adults aged 18–87 years were included. The results showed that the median (P25–P75) values of daily fluid intake of the participants were 1100 (800–1550) mL in summer and 1000 (750–1300) mL in winter. Women had a higher prevalence of IFI than men in both summer (odds ratio (OR) = 2.683, 95% confidence interval (CI): 1.830–3.934) and winter (OR = 2.636, 95% CI: 1.677–4.142). The results of CART analysis showed that, in summer, BMI < 25 kg/m2 (probability: 64.2%) and age < 64 years (probability: 67.4%) were main risk factors of IFI for men, and BMI < 29 kg/m2 (probability: 81.6%) and living in C Community (probability: 86.7%) were main risk factors for women. In winter, age < 40 years (probability: 81.8%) and BMI < 20 kg/m2 (probability: 94.5%) were identified as main risk factors of IFI for men and women, respectively. Conclusions Most of the participants living in the study site had IFI. The fluid consumption varied by gender, age, location, and BMI. The findings could be useful for the implementation and optimization of intervention programs by identifying the individuals who may at greater risk of dehydration.