The Prevalence of Hyperuricemia and Its Correlates among Adults in China: Results from CNHS 2015–2017

This study aimed to investigate the prevalence of hyperuricemia (HUA) and associated risk factors in Chinese adults aged 18 to 59 years. All the data were collected from the China Nutrition and Health Surveillance during the period 2015–2017, which adopted a stratified, multistage, random sampling method on a national scale. A total of 52,627 participants aged 18 years or older were included in this study. The definition of hyperuricemia was 420 μmol/L for male and 360 μmol/L for female. The Rao–Scott chi-square test was used to compare the differences in prevalence between or among the subgroups. A weighted two-level multivariate survey-logistic regression was used to detect the correlations between HUA and demographic, physical, lifestyle and metabolic factors. The total prevalence of HUA was 15.1%, and that was higher in male, current smokers, higher BMI and less physical activities subgroups, and also in noninfectious chronic diseases (NCDs) subgroups. The subgroups of non-vegetarian diet, insufficient vegetable intakes and excessive red meat and alcohol intakes had significantly higher HUA prevalence. After introducing all the variables in the survey-logistic regression, gender, age, BMI, physically active, hypertension, diabetes mellitus, bean and nut intake, vegetable intake, red meat intake, alcohol consumption and vegetarian were associated with HUA. Among the significant variables, age and physical active served as a protective factor, and BMI showed to be a risk factor for HUA. Hypertension and dyslipidemia could increase the risk for HUA, while diabetes mellitus was shown a negative association with it. For dietary factors, vegetarian diet, sufficient beans and nuts and vegetables intake could lower the risk of HUA, but more alcohol could increase the risk of HUA. Dietary factor played a key role in HUA. It suggested that the intervention of dietary factor should receive more attention to ameliorate the high prevalence of HUA in China.


Introduction
As the final product of purine metabolism [1], uric acid (UA) is produced in the liver from degradation of dietary and endogenously synthesized purine compounds. It had been confirmed that the organ systems could be negatively affected by the elevated serum uric acid (SUA) level [2], for the potential effects of UA, such as pro-oxidant, antioxidant, and proinflammatory. However, for the shortage of urate oxidase, UA cannot be degraded to allantoin in the metabolic process of human being. If the producing rate of UA exceed the eliminating rate via kidney, UA could be accumulated and result in high concentration [3]. Hyperuricemia (HUA) is a status of high SUA concentration, which may lead to urate crystals being deposited in joints, tendons, and other tissues, and could also elevate the risk of gout or other comorbidities development [4,5]. Besides gout, the relationships between HUA and dyslipidemia [6], cardiovascular disease (CVD) [7], metabolic syndrome (MetS) [4], insulin resistance [8], hypertension [9] or renal disease [10] have been observed in previous studies.
In China, HUA was also showed that might be a potential risk to public health. In 2015-2016, the total HUA prevalence in China was 11.1% and then elevated to 14.0% in 2018-2019 [11]. For the broad territory and multi-ethnic populations, the prevalence of HUA in China was diverse. For regions, Huang et al. [12] found that the HUA prevalence of urban adults in southwestern China was 13.5%, and it was higher than the HUA prevalence of Henan rural population (12.6%) [13]. The fluctuations in HUA prevalence that have been shown above might suggest that the affect factors for HUA in different regions or populations are quite varied. In previous studies, the associations of HUA with age, gender, socioeconomic conditions, and marital status have been observed [14][15][16]. Additionally, some chronic diseases, for example, hypertension, hyperlipidemia, overweight or obesity were also considered to have the effects on an increased risk for HUA [13,17,18]. For food sources, the purine-rich diet, such as seafoods, meats, and drinking have been considered to be the risk for HUA [19]. Dietary pattens, for instance, the Healthy Cardiometabolic Diets and the Dietary Approaches to Stop Hypertension (DASH) Diet, contributed to the development of HUA [20], and that has also been verified in an animal model [21].
However, while the previous studies have found some of the risk factors for HUA, few of them considered the combinations of demographic information, chronic disease status and diet intaking conditions, and the representativeness of population was also a limitation to them. The current study aims to describe the prevalence of HUA using the data of the China Nutrition and Health Surveillance (CNHS) 2015-2017 [22], a national representative survey, and demonstrate the risk factors for HUA in Chinese adults.

Data Source and Participants
Data of the study were collected from CNHS 2015-2017, which adopted a stratified, multistage, random sampling method to recruit representative participants from 31 provincial-level administrative divisions (PLADs), including provinces, autonomous regions, and municipalities in mainland China [22]. A total of 52,627 participants aged 18 years or older were included in this study.

Data Collection and Measurements
Data were collected from the participants of CNHS 2015-2017 via four parts, including standard questionnaire interview, physical examination, dietary survey, and laboratory test. Questionnaire interview was used to collect the demographic characteristics and lifestyle of participants including the family and personal information by face-to-face interviews. Physical examination was conducted to obtain the height, weight, and blood pressure. Height was measured by the tool of height and sitting height meter (TZG, Suzhou, China), weight was measured by electronic weighing scale (G&G TC-200K, Changshu, China) and blood pressure was measured by Omron HBP-1300 electronic sphygmomanometer (Dalian, China). The consumptions of vegetables, fruits, milk, alcohol, red meat, bean and nut were collected using the method of Food Frequency Questionnaire (FFQ).
Laboratory test was used to detect the levels of serum UA (SUA). Overnight fasting cubital venous blood samples were collected in blood collection tubes (SSTTM II Advance, Becton, Dickinson and Company, Franklin Lakes, NJ, USA) by local CDC staffs and clinicians. After collection, the blood samples were left standing for at least 30 min. Sera were separated from the whole blood by centrifugation at 3000 rpm for 15 min at the survey-site, and then stored at −80 • C until tested in the laboratory. HbA1c (glycosylated hemoglobin α-1-c) was assessed by high-performance liquid chromatography (Premier Hb9210 trinity biotech, Ireland). Fasting blood glucose (FPG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglycerides (TG) were measured by a Hitachi 7600 automatic biochemical analyzer (Hitachi Inc., Tokyo, Japan) with reagents obtained from Wako Pure Chemical Industries Ltd. (Tokyo, Japan) [23] at the central laboratory of the Beijing CDC and National Institute for Nutrition and Health, China CDC. SUA levels were measured using the uricase-HMMPS method on a Roche analyzer (Cobas 8000-c702, Roche Diagnostics, Switzerland).

Quality Control
All implementations of the surveillance were under strict quality control and supervision. Staff of the surveillance should be unified, trained and qualified by the China CDC project team. The face-to-face questionnaire interviews were administered by the qualified investigators, and then questionnaires were checked and verified by inspectors. Physical measurements were operated following the unified manuals or procedures, and the equipment were unified and well calibrated. The blood samples were measured in the designated laboratories using a standard protocol according to the manufacturer's instructions and reference samples. Data were double input and cleaned up by uniform standard.

Definition of HUA
The definition of HUA was based on the concentration of UA in serum. HUA was diagnosed according to the clinical diagnostic criteria. The cut-off SUA level was 420 µmol/L for male and 360 µmol/L for female [24].

Dietary Intake Assessment
Dietary intake was assessed by 7 kinds of food sources such as bean and nut, vegetable, fruit, milk, red meat, and alcohol. Among these indices, bean and nut, vegetable, fruit, and milk intakes were assessed by the criteria of Chinese Dietary Guidelines (2022): (1) Bean and nut intake was defined as sufficient if the daily average intake was ≥25 g; (2) Vegetable intake was defined as sufficient if the daily average intake was ≥300 g; (3) Fruit intake was defined as sufficient if the daily average intake was ≥200 g; (4) Milk intake was defined as sufficient if the daily average intake was ≥300 g; (5) Red meat intake was categorized as insufficient (daily average red meat intake <18 g), moderate (18 g to 27 g) or excessive (>27 g) [25]. (6) Alcohol consumption was categorized by international guide for monitoring alcohol consumption and related harm (WHO), such as never consumed (consume < 1 g of alcohol per day for both male and female), low risk consumed (consume 1 to 40 g of alcohol per day for male or consume 1 to 20 g of alcohol per day for female), medium risk consumed (consume 41 to 60 g of alcohol per day for male or consume 21 to 40 g of alcohol per day for female) or high and very high risk consumed (consume ≥61 g of alcohol per day for male or consume ≥41 g of alcohol per day for female).

Statistical Analysis
SAS version 9.4 software (SAS Institute Inc., Cary, NC, USA) was used for all data cleaning and analysis. The demographic structures of subjects were presented as percentages. The weighted prevalence of HUA was presented as ratios and 95% confidence intervals (95% CI), and Rao-Scott chi-square test was used to compare the difference between or among the subgroups. PROC SURVEYLOGISTIC procedure was conducted to detect the influencing factors of HUA, and the results were presented as odds ratio (OR) and 95% CI by comparing with the reference in each model. Model 1 was a crude model, the covariables including gender, residence location, area of the country, age, education level and household income. Model 2 was adjusted for body mass index (BMI), smoking, physical active, hypertension, diabetes mellitus, dyslipidemia. Model 3 was further adjusted for the diet covariables, including bean and nut intake, vegetable intake, fruit intake, milk intake, red meat intake, alcohol consumption and vegetarian. A two-sided p value < 0.05 was considered to indicate statistical significance.

Ethics Statements
Written informed consent was obtained from all participants and their guardians. The study protocol was approved by the Ethical Review Committee of the Chinese Center for Disease Control and Prevention (No. 201519-B).

Basic Characteristics
As shown in Table 1, a total of 52,627 participants (46.41% for males and 53.59% for females) were finally included in current study. Participants who were mid-aged and elderly, had lower household income and normal weight, lived in urban areas and eastern China composed a higher proportion. Compared between different genders, males tended to have higher education level, more physical activities, but more former and current smokers. For those who had hypertension or dyslipidemia, males composed higher proportion, while for diabetes mellitus, females composed higher.

Dietary Intakes
Among whole of the participants, more participants ate a non-vegetarian diet, had less alcohol intakes, but had sufficient intakes of beans and nuts, insufficient intakes of vegetables, fruits and milk, and excessive intakes of red meats. Additionally, except for females tended to have healthier intakes of the above kinds of foods. Further information is available in Table 1.

Prevalence of HUA among Participants
Weighted prevalence of HUA in different categories was shown in Table 2. According to different categories, those who were male, younger, current smokers, and lived in urban areas and eastern China, had higher education level, moderate household income, higher BMI, less physical activities had significantly higher prevalence of HUA. Meanwhile, compared with hypertension or dyslipidemia patients, those who were free of these noninfectious chronic diseases (NCDs) had lower significantly prevalence of HUA. In consideration of food intakes, participants who ate non-vegetarian diet, had insufficient intakes of vegetables, excessive intakes of red meats and alcohol had significantly higher prevalence of HUA.

Discussion
This study was based on the data from CNHS 2015-2017, and the results were deemed to be national representative. In our study, we observed the HUA prevalence, and further analyzed the influencing factors for HUA. Because of the resulting limitations in previous studies, such as the smaller sample size, region presentiveness or the special population characteristics, the results of this study might be more able to reflect the HUA prevalence status and verify the influencing factors for HUA in Chinese adult population.
The prevalence of HUA was different among populations or regions. In the present study, the overall prevalence of HUA in Chinese adult population was 15.1%, and the prevalence in male (21.2%) was significantly higher than female (8.5%). Data of the National Health and Nutrition Examination Survey (NHANES) showed that the HUA prevalence in the United States in 2015-2016 was 20.1% in total population (20.2% in men and 20.0% in women) [31]. The HUA prevalence in the general Italian population was 11.9% [32]. Additionally, a systematic review demonstrated that the prevalence of HUA in Australia ranged between 10.5% and 16.6% in Caucasian or an Australian representative population [33]. In the neighboring countries, the results drawn from the study that was conducted in the Bangkok population showed that the prevalence of HUA in the study population was 24.4%, and that was significantly more common in male (59%) than female (11%) [34]. The Korean National Health and Nutrition Examination Survey (KNHANES) revealed that the prevalence of HUA in the general Korean population was 11.4% (17.0% in male and 5.9% in female) [35]. From the results shown above, the HUA prevalence in China was lower than that in the US and higher than that in Italy. Moreover, the HUA prevalence of female was nearly equal to male in the US, but that was obviously higher than the Chinese female group. Among the neighboring countries, the prevalence trends of HUA in China were consistent with others. Although the prevalence of HUA in previous domestic studies were fluctuating, the results of our study were consistent with the commonly higher HUA prevalence in male population. In our study, the HUA prevalence in urban and rural was 17.2% and 12.8%, respectively, with both being higher than the results of Yang's study conducted in Jinan and lower than the results of Zhang's study conducted in Sichuan province. A study [36] conducted in a China rural population showed that the prevalence of hypertension, dyslipidemia in the HUA group were both higher than that in the non-HUA group. Our results were consistent with these. In our study, the prevalence of HUA was significantly higher in hypertension and dyslipidemia populations, and a negative result was observed in the diabetes mellitus population. The current study showed that the HUA prevalence in the insufficient physical active population was significantly higher than in the sufficient physical active population. However, in Yang's [30] and Wang's [37] study, the differences between the insufficient physical active population and sufficient physical active population were not observed. With regard to diet intakes, no positive results of the HUA prevalence were observed in bean and nut intake, fruit intake and milk intake analyses. However, significant differences were detected in vegetable intake, red meat intake and alcohol consumption analyses, being similar to Yang's and Huang's studies [12,38]. To our knowledge, our study is the first of its kind to introduce the vegetarian factor, and the prevalence of HUA in the vegetarian group was significantly lower than in the non-vegetarian group.
Although the risk factors for HUA had been detected in previous studies, the effects on HUA were different among them. In the present study, the risk factors for HUA were detected by three survey-logistic regression analysis models. In Model 1, covariates merely included demographic factors. A positive association was observed between the factors of gender, residence location and age and HUA. In Model 2, after adjustment for BMI, smoking, physical active, hypertension, diabetes mellitus and dyslipidemia, the effect of residence location was nullified. Smoking was not observed to have any significant association with HUA. In Model 3, after further adjustment for diet covariables including bean and nut intake, vegetable intake, fruit intake, milk intake, red meat intake, alcohol consumption and vegetarian, the effects of fruit and milk intake were not observed. Among demographic factors, gender and age have been confirmed in nearly all the relevant studies. The underlying biological mechanism could be interpreted as the estrogen concentration fluctuating. Guan's study [39] showed that in Chinese post-menopausal women population, the HUA prevalence increased remarkably with age increasing. The similar results were also observed in western populations [40,41]. E2 estradiol was inversely associated with the uric acid level [42]. The potential biological mechanisms of E2 affecting the SUA levels may include renal clearance, secretion or reabsorption [43]. That could be the causes of different SUA levels between genders. BMI had been confirmed as a risk factor associated with HUA [44], and that was duplicated in our study. In the results, the prevalence of HUA in obese group was nearly threefold higher than that in the normal group. However, BMI changing may be caused by multiple factors, and more studies should be conducted in further investigations. In the present study, sufficient physical activity was presented as the protective factor for HUA, which was consistent with the results of Cui's study [17]. The mechanism might be that a lack of exercise may reduce insulin sensitivity and increase urine volume, thus resulting in an elevated SUA level [45]. The associations between hypertension, dyslipidemia and diabetes mellitus and HUA have been detected in previous studies. Yang et al. found that hypertension, hypercholesterolemia and hypertriglyceridemia were the risk factors for HUA [38]. In Zhu's study, diabetes mellitus was detected as the risk factor for HUA, but that was only observed in the women group. That might be due to the sex-specific physiologic impact of HUA [46]. In our findings, hypertension and dyslipidemia were detected as the risk factor for HUA in our study, which was consistent with previous studies, but the contrary was true for diabetes mellitus. Although the associations between HUA and metabolic diseases have been observed, the potential mechanisms should be further studied.
Diet may play an important role in the development of HUA. The associations between dietary factors and HUA have been detected in several studies. Purine, as one of the most important sources of UA in food components, has been observed in previous studies. A study conducted in The Health Professionals Follow-up Study found that higher consumption levels of purine-rich foods, such as meat and seafood, were associated with an increased risk of gout, and dietary patterns rich in animal foods were positively associated with a higher risk of HUA, whereas the vegetable pattern was negatively associated [47]. Villegas et al. reported that both meat and seafood intake have been associated with higher prevalence of HUA, but no association between total protein intake and HUA was observed [48]. In our study, the higher consumption levels of red meat were positively associated with HUA, and the negative associations between bean and nut intake and vegetable intake and HUA were observed. The vegetarian was also detected as the protective factor for HUA in our results. No associations between fruit and milk intake and HUA were observed in the current study, which was consistent with previous studies. According to a previous study, protein-rich foods might enhance UA production due to the higher quantity of purines. However, this also could promote the urinary urate excretion and lead to a reduction in the concentration of SUA [49]. Higher unsaturated fatty acid and phytochemical-rich plant intakes are the main contents of the principal food sources of the vegetarian diet. By virtue of xanthine oxidase inhibition, phytochemicals have strong effects on reducing SUA levels while inhibiting UA generation in purine metabolism [50]. The relationship between alcohol and HUA was also observed in our study. In comparison with the never consumption group, the OR in the medium consumption group was 1.640 (95 CI%:1.093-2.459) and dropped to 1.390 (95 CI%: 1.117-1.730) in the high-and very high-risk consumption group. This trend was consistent with Choi's study [51]. One of the potential mechanisms could be the conversion of alcohol to lactic acid, which reduces renal uric acid excretion by competitively inhibiting uric acid secretion by the proximal tubule [52]. However, a mendelian randomization study demonstrated that alcohol consumption had no causal effect on the development of HUA [51]. It suggested that, apart from other factors, genetic backgrounds may potentially influence the HUA prevalence among the different ethnic groups.
As a national representative survey, this study was first time to describe the prevalence of HUA in the Chinese adults population. According to our results, the associations between HUA and gender, age, physically active, hypertension and dietary factors were verified. Because of the population mobility and economic development, the influence strength of geographical, education and household income were vanished. To the best of our knowledge, the present study was the first of its kind to introduce vegetarian as an independent influencing factor into the analysis. However, some limitations of this study should be noticed. Firstly, the present study was cross-sectional research, and thus, the causal relationship between HUA and influence factors may not be well elucidated. Secondly, cancer status, genetic and ethnic factors were not considered in this study, and thus, the effects of cancer and genes on HUA might be ignored.

Conclusions
HUA is a public health problem in the Chinese adult population, especially in male and urban residents, and it should be paid more attention by the relevant government sectors. Dietary factors, including bean and nut intake, vegetable intake, red meat intake, and alcohol consumption, were significantly associated with HUA. Moreover, vegetarian diet could be a protective factor for HUA. Comparing with other influencing factors of HUA, dietary factors were intervened more feasibly. Besides focusing on a certain type of food item, an optimal dietary pattern could bring about better HUA intervention effects. The outcomes of this study could also provide some notions for the prevention of other chronic diseases and be beneficial for large-scale populations. For the aim of diet conversion, policy making and targeted health education should be taken into account.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are non-public.