Associations of Waist Circumference, Socioeconomic, Environmental, and Behavioral Factors with Chronic Kidney Disease in Normal Weight, Overweight, and Obese People

Background: Chronic kidney disease (CKD) places a heavy burden on the healthcare system worldwide. The risk factors may vary by body adiposity. We aimed to investigate the associations of socioeconomic, environmental, and behavioral factors with CKD in different groups of body mass indexes (BMI). Methods: A case-control study was conducted in 3280 participants (1048 CKD and 2232 non-CKD) from seven hospitals and nearby communities from May 2012 to August 2015. Personal characteristics, anthropometrics, environmental exposures, and health−related behaviors were assessed using a structured questionnaire. The logistic regression models were utilized for analysis. Results: Older age (odd ratio, OR = 2.85; p < 0.001), being men (OR = 4.23; p < 0.001), smoking (OR = 3.36; p < 0.001), stable income (OR = 0.33; p < 0.001), higher education (OR = 0.37~0.38; p < 0.001), and daily adequate water intake (OR = 0.64; p = 0.010) were associated with CKD in normal weight people. Older age (OR = 2.49; p < 0.001), being men (OR = 3.36; p < 0.001), education (OR = 0.44, p = 0.004), hypertension (OR = 2.93; p<0.001), diabetes (OR = 1.83; p = 0.004), and using traditional Chinese medicine (OR = 2.03, p = 0.014) were associated with CKD in overweight people. Older age (OR = 2.71; p < 0.001), being men (OR = 2.69; p < 0.001), hypertension (OR = 2.93; p < 0.001), diabetes (OR = 1.94; p = 0.001) were associated with CKD in obese people. Conclusions: The associated factors of CKD varied by different groups of BMI. These findings may help to develop potential interventions to manage CKD.


Introduction
Chronic kidney disease (CKD) is a major public health issue and significantly contributes to morbidity and mortality globally [1]. CKD places a heavy burden on the healthcare system in every country [2]. The global prevalence of CKD was 10-15% in the general populations [3][4][5], and about 36% in high-risk populations [5]. Taiwan experiences a high prevalence of CKD [6][7][8], and is one of leading countries with highest prevalence of end-stage renal disease [9]. Moreover, awareness of the disease and its risk factors is low in general populations in 12 countries from six world regions [5], and in Taiwan as well [7,10].
It is the consensus that the optimal management of CKD includes cardiovascular risk reduction, treatment of albuminuria, avoidance of potential nephrotoxins, and adjustments to drug dosing [11]. However, factors like socioeconomic, environmental, and behavioral factors have been believed to be important in management of CKD, but have not yet been delineated. From an international perspective, a multidisciplinary, multi−sector, and multifaceted action plan is needed to develop a strategy to prevent and manage CKD in order to combat its burden and complications [4,12]. In Taiwan, the multidisciplinary care helps to slow down the renal progression and improve survival rate [13]. In addition, the risk factors of CKD in people with normal weight, overweight, and obese might be varied. Therefore, we aimed to explore the association between personal characteristics, socioeconomic, environmental factors, and health-related behaviors with CKD in normal, overweight, and obese people.

Sampling and Sample Size
Participants recruited were Taiwanese citizens aged 18 years or older, who understand Mandarin Chinese. We excluded participants who were diagnosed as CKD stage 5 or end-stage renal disease (ESRD), or who lacked renal function parameters. Cases were patients with chronic kidney disease stage 1 to stage 4. Controls were non-CKD patients, relatives of CKD patients, healthy people who underwent annual health checks, and in the communities around the selected hospitals. The final sample for analysis was 3280 participants with N = 1048 cases, and N = 2232 controls ( Figure 1).

Instruments and Assessments
Participants' characteristics: Participants were asked to provide their age (years), gender (male vs. female), blood type (O, A, B, AB), marital status (single, married/cohabited, separated/divorced/widow), education attainment (elementary and below, junior high school, senior high school, college/university and above), occupation (unemployed, employed), monthly household income (unstable income, stable income), and medical history (hypertension, diabetes mellitus, hyperlipidemia).
Environmental exposures: Participants were asked to provide information regarding past expose to environmental risk factors such as pesticides, animals (keeping pets or poultry, slaughtering or selling chicken, ducks, etc.), chemicals (asbestos, heavy metal lead, mercury, polychlorinated biphenyls, etc.), and organic solvents or compounds (e.g., gasoline, derivatives, plywood adhesives, paints, methane, rosin, etc.).
Health-related behaviors: The behaviors assessed included daily intake of water (<1.5 L/day vs. ≥1.5 L/day), using traditional Chinese medicine products (yes vs. no), smoking tobacco (yes vs. no), drinking alcohol (yes vs. no), and chewing betel nut (yes vs. no). Betel quid and areca nut, called betel nut, is widely used as plant chewing gum in Asian communities [17]. It was summarized that betel nut chewing was associated with increased CKD risk [18]. To evaluate physical activity level, the short version of the International Physical Activity Questionnaire (IPAQ), a widely used method, was utilized [19]. Patients were asked to provide information on their time spent (days per week, and minutes per day) on different levels of physical intensity (vigorous, moderate, walking, and sitting). The overall physical activity score was given by calculating the sum of minutes spent on activities at different levels of vigorous, moderate, walking, and sitting over the last seven days multiplied by 8.0, 4.0, and 3.3, 1.0, respectively [19]. The common method using metabolic equivalent task scored in minute per week (named as MET-min/wk) was used to represent the physical activity [20].

Data Collection Procedures
Interviewers were seven research assistants from seven hospitals. They received 4 h of training for data collection. Patients and healthy people participated in the study voluntarily and were checked by research assistants for eligibility. The face-to-face interviews were conducted in hospitals or in community activity stations. After interviews, blood specimens were collected by registered nurses and then were analyzed in hospital laboratories to determine renal function parameters. The same laboratory criteria and protocol were adopted for all hospitals involved. The diagnosis of CKD was based on the hospital laboratory results. All the misclassification of patients from interviews or lack of renal function parameters were excluded from final sample. The study procedure was recapped in Figure 1.

Ethical Approvals
The study protocol was approved by the ethical committee of Taipei Medical University Joint Institutional Review Board of Taipei Medical University (TMU-JIRB No. 201204036). All participants signed the written informed consent form before their participation.

Data Analysis
The studied variables were described as mean and standard deviation, frequency and percentage, median and interquartile range. The independent-samples T-test and one-way ANOVA test, and the Mann Whitney test were used to compare the distribution of studied variables between non-CKD and CKD groups, appropriately. Finally, the bivariate logistic regression model was used to examine the associations of participants' characteristics, medical history, health-related behaviors with CKD in the overall sample, in the group with normal weight, and in the overweight/obese group. Variables selected into the multivariate logistic regression model were those that showed the association with CKD at p < 0.20 from the bivariate model [23]. In order to avoid multicollinearity in the multivariate analysis, the Spearman correlation was utilized to check the correlations between selected variables. Analysis was conducted using the IBM SPSS software version 20.0 for Windows (IBM Corp., Armonk, NY, USA). The significant level was set at p-value < 0.05.

Results
The average age of the study population was 60.5 ± 13.2 years, 47.3% were men, 29.2% were overweight, and 24.6% were obesity. The prevalence of CKD was varied by different categories of age, gender, blood type, BMI, WC, marital status, education, occupation, household income, history of diseases, exposure to pesticide and animal, daily intake of plain water, smoking, chewing betel nut, and physical activities (Table 1).  * Data were presented as mean ± SD, frequency and percentage, median (interquartile range) and p-values were calculated using independent sample t-test, Chi-square test, and Mann Whitney test, appropriately.
In the overweight group, results of simple regression analysis showed that age, gender, WC, marital status, education, occupation, income, hypertension, diabetes, hyperlipidemia, and smoking were significantly associated with CKD ( Table 2). Moderate correlations were existed between gender and smoking (r = 0.46), smoking and chewing betel nut (r = 0.45) (Table S1). Gender, chewing betel nut, and other factors were analyzed in the multivariate analysis. We found that the factors showing significantly positive association with CKD were age ≥65 years ( Table 3).

Discussion
The current study showed that people aged ≥65 years had a higher likelihood of CKD. Age was found to be a risk factor of CKD in previous studies [24][25][26]. Additionally, our study showed that men have a higher probability of having CKD than women. Note that, across all regions, it has been shown that men were closely associated with the occurrence of CKD [27]; by contrast, in other parts of the world such as in China, Switzerland, and Iran, women had a higher risk of CKD than men [24,28,29]. In China, authors analyzed people with eGFR ≥ 60.0 mL/min/1.73 m 2 [24], while we analyzed people with CKD at stage 1 to stage 4, the different selection criteria might affect the results. In Switzerland and Iran, the different genetic background might contribute to this contradiction [30]. Based on this evidence, we conclude that gender should be taken into account when developing strategical approaches to prevent and treat CKD and its complications [31]. Moreover, we found that people with higher education and stable income had a lower likelihood of having CKD, a finding consistent with previous studies showing that low socioeconomic status (low education, occupation, and income) was strongly associated with a higher risk of having CKD [32,33].
In the multivariate model, being married or cohabited had a lower likelihood of having CKD as compared with being unmarried in the overall sample. This could be explained by the previous evidence that being unmarried or living alone was associated with higher systolic blood pressure [34], a high prevalence of hypertension [35], and lower hypertension awareness and control rate [36,37]. Hypertension has been proven to be a strong predictor of CKD [24]. In addition, married people had lower prevalence of diabetes and smoking [38]. Diabetes [39,40], and smoking [26] were recognized as important predictors of CKD. Taken together, married or cohabited people had lower blood pressure, higher hypertension awareness and control, and lower prevalence of diabetes and smoking, which, in turn, led to a lower risk of CKD as compared with unmarried people.
Our study illustrated that people with hypertension and diabetes had a higher likelihood of CKD. It has been proven that hypertension is a cause and complication of CKD [24,[41][42][43]. In addition, diabetes was one of the most common causes of CKD [39,40,43]. On the other hand, insulin resistance was independently associated with cardiovascular risks in ESRD patients [44]. Therefore, there is a consensus that better management of hypertension, and diabetes is required to improve CKD patient care [45][46][47].
Our data showed that, general overweight and obesity was positively associated with a higher likelihood of CKD in the overall sample. The previous evidence showed that obesity measured by different adiposity indices (percentage body fat, visceral fat index, body mass index, waist circumference, and waist-to-height ratio) was significantly associated with increased risk of having CKD [48]. In addition, results of bivariate analysis showed that elevated WC was significantly associated with higher prevalence of CKD in the overall sample, in overweight, and in obese individuals as well. This was partly supported by the evidence that the effect of obesity in triggering CKD was only significant in the obese population [49].
Our study showed that people exposed to animals had a higher CKD occurrence. This might be explained by environmental factors also being a common cause of CKD [39]. In addition, people who had a daily intake of plain water ≥1.5 L of plain water had a lower likelihood of developing CKD. Water is an essential nutrient for body hydration [50]. A previous study showed that increasing daily water consumption slowed down the progression of CKD [51]. Moreover, overweight people using traditional Chinese medicine (TCM) products had a higher rate of CKD as compared to those did not use. A number of studies found the adverse effect of TCM on CKD [52]. However, the evidence is insufficient, the practice of TCM in varied, rigorous, well−designed studies is required to provide evidence-based practices to improve the safety and efficacy [52].
Current smoking was associated with a higher prevalence of CKD in the current study. Smoking has been recognized as a risk factor for CKD in both men and women [26], and smoking increased the risk for cardiovascular events, ESRD, and mortality in CKD patients [53,54]. Alcohol consumption was not associated with CKD in our study. This could be explained that the quality of alcohol was not measured in the current study. The previous study showed that alcohol intake was associated with incident CKD [55]. It is suggested that limited alcohol consumption could minimize the health risks [56] and delay CKD progression [57].
This study had some limitations. Firstly, a causal relationship cannot be drawn from a cross-sectional design. However, the findings from this study might be useful for future epidemiological studies or public health programs. Secondly, study indicators were subjectively assessed using a structured questionnaire. However, considering that (1) interviewers were well-trained for data collection, and (2) Taiwanese people have high self-awareness in personal health under universal health coverage policy. These factors may reduce the interview bias. Future studies with a larger sample and better study design are suggested to comprehensively explore the risk factors and assure the causality.

Conclusions
In normal weight people, older age, being men, and smoking were associated with a higher prevalence of CKD, while stable income and drinking enough daily plain water were associated with a lower prevalence of CKD. In overweight people, older age, being men, history of hypertension, diabetes, using TCM products, and smoking were associated with a higher prevalence of CKD, while higher education status was found to be a protective factor of CKD. In obese people, older age, being men, history of hypertension, diabetes, and smoking were associated with a higher rate of CKD.
The evidence obtained from this study may help to improve multifaceted strategical interventions to prevent and manage CKD.
Supplementary Materials: The following are available online at http://www.mdpi.com/1660-4601/16/24/5093/s1, Table S1. The Spearman correlation among the variables associated with chronic kidney disease at p < 0.20. The funder had no role in the decision to collect data, data analysis, or reporting of the results.