2.1 Study population
The analyses are based on data from the Henan Rural Cohort Study, which was a prospective study of chronic non-communicable diseases, the cohort profile and study population has been described in elsewhere [18, 19]. In briefly, the study was conducted in Yuzhou, Xinxiang, Tongxu, Yima and Suiping county of Henan Province, China from July 2015 to September 2017. Overall, 39,259 eligible individuals aged 18–79 years were asked to completed baseline survey. In the current study, a total of 404 individuals were excluded from the present analysis due to missing data on serum uric acid level (n = 54), and diagnosed with cancer and serious renal disease (n = 350). Finally, 38,855 participants (15,371 men and 23,484 women) were included in current analysis. The study complied with the 1975 Declaration of Helsinki and was approved by the ethics committee of Zhengzhou University. Written informed consent was obtained from each participant before starting this study.
2.2 Definitions of exposure
PA and ST levels were assessed by using the International Physical Activity Questionnaire (IPAQ) [20]. As the details have been described previously [16, 19], PA-metabolic equivalent (MET) was calculated by using MET coefficient of activity × duration (hour per time) × frequency (times per week) [19]. Therefore, according to the categories of PA, each category of vigorous activity, moderate activity and walking was calculated with the corresponding MET values of 8, 4, and 3.3 respectively. The total value of MET was obtained by adding value of these three categories. Finally, based on the standard scoring criteria of IPAQ, three levels of PA were determined: light (physically inactive), moderate and vigorous [16]. The individual total ST every day in the past week was estimated by using the question “About how many hours in each 24-h day do you usually spend on sitting? Same as some recent studies [21], daily ST was classified by four groups: <4 h/d, 4 to 6 h/d, 6 to 8 h/d, and ≥ 8 h/d.
2.3 Definition of HUA
Blood samples were obtained from each individual after at least 8 h of overnight fasting to measure multiple biochemical indicators. Serum uric acid level was measured by ROCHE Cobas C501 automatic biochemical analyzer with enzymatic colorimetric method. HUA was defined as serum uric acid level > 7.0 mg/dL (417 µmol/L) in males and serum uric acid level > 6.0 mg/dL (357 µmol/L) in females among Chinese population [17].
2.4 Potential covariates
Potential confounders were similar to our previous study of HUA [17], including demographic covariates (age and gender), socioeconomic covariates (marital status, average monthly individual income and education level) and lifestyle factors (smoking, alcohol drinking, dietary, etc.), which were collected through face-to-face interview by trained research staff using a standardized questionnaire. Socioeconomic covariates included education level (“primary school or below”, “middle school and high school or above”), marital status (“married/ living together” or “divorced/widowed/separated/ unmarried”), average monthly income (“<500 RMB”, “500–1000 RMB” or “>1000 RMB”). Smoking status was classified into current, former, and never groups. Drinking status also was classified into current, former, and never groups. The four-cluster dietary patterns were obtained in the current study by factor analysis using the previous standard principal component analysis method [22]. Body weight and height were measured with the standard methods and body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared. Blood pressure was measured 3 times on the right arm at heart level sitting position. Obesity (yes/no), hypertension (yes/no), type 2 diabetes mellitus (T2DM) (yes/no) and dyslipidemia (yes/no) also were defined as previous descriptions [17].
2.5 Statistical analysis
The differences in basic characteristics between HUA and non-HUA participants were tested using the Student's t-test and the chi-square test. Continuous variables and categorical variables were expressed as mean ± standard deviations (SD) and counts and percentage, respectively.
The associations between PA or ST and serum uric acid level were analyzed using multiple linear regression models and presented as β values with 95% confidence intervals (CI), where the light PA and the lowest sitting (< 4 h/day) served as the reference categories, respectively. Multivariable-adjusted logistic regression models also applied to assess associations of PA or ST and prevalence of HUA, results are presented as odds ratio (OR) with corresponding 95% CI. Similarly, the effects of MET-hour/day and per hour increased in ST on serum uric acid level or prevalence of HUA both were estimated. All potential covariates were selected based on the existing literature and three models were performed: model 1 was adjusted for PA and ST level where applicable; model 2 included ST and PA as well as age, gender, education level, marital status, average monthly income, smoking status, drinking status, dietary pattern and model 3: model 2 plus adjustment for obesity, T2DM, hypertension and dyslipidemia status. In addition, a cross-product term was incorporated into logistic regression models to evaluate the statistical significance of interaction of PA and ST on HUA. Restricted cubic spline regression was used to explore the dose–response relationship between continuous PA-MET hour and ST (h/day) and HUA.
Meantime, stratified analyses were performed to estimate the effects of PA on HUA in different ST groups. Joint association between PA and ST also was examined the by deriving a combined variable with 12 groups, where the combined light PA and lowest ST (< 4 h/day) served as the reference group. Furthermore, generalized linear models also were employed to visualize the interactive effect of ST (h/day) and PA (MET-hour/day) on HUA. Effect estimates of MET-hour/day were plotted with their 95% CI as a function of increasing ST (h/day) in an interaction plot.
All analyses were conducted using IBM SPSS V.19.0 and R 3.5.0. A two tail of P value < 0.05 was considered statistically significance.