## Abstract

Background. Previous reports of chronic kidney disease (CKD) prevalence in Thailand varied from 4.3% to 13.8%. However, there were methodological concerns with these reports in terms of generalization and the accuracy of estimation. This study was, therefore, conducted to determine CKD prevalence and its risk factors in Thai adult populations.

Methods. The population-based Thai Screening and Early Evaluation of Kidney Disease (SEEK) study was conducted with cross-sectional stratified-cluster sampling. Serum creatinine was analysed using the modified Jaffe method and then standardized with isotope dilution mass spectrometry.

Results. The study included 3,459 subjects were included in the study. The mean age was 45.2 years (SE = 0.8), and 54.5% were female. Six hundred and twenty-six subjects were identified as having CKD, which evidenced an overall CKD prevalence of 17.5% [95% confidence interval (95% CI) = 14.6–20.4%]. The CKD prevalence of Stages I, II, III and IV were 3.3% (95% CI = 2.5%, 4.1%), 5.6% (95% CI = 4.2%, 7.0%), 7.5% (95% CI = 6.2%, 8.8%) and 1.1% (95% CI = 0.7%, 1.5%), respectively. The prevalence of CKD was higher in Bangkok, the Northern and Northeastern regions than in the Central and Southern regions. Seven factors (i.e. age, gender, diabetes, hypertension, hyperuricaemia, history of kidney stones and the use of traditional medicines) were associated with CKD. Only 1.9% of the subjects were aware that they had CKD.

Conclusions. CKD prevalence in the Thai population is much higher than previously known and published. Early stages of CKD seem to be as common as later stages. However, albuminuria measurement was not confirmed and adjusting for persistent positive rates resulted in the prevalence of 14.4%. Furthermore, the awareness of CKD was quite low in the Thai population.

## Introduction

Chronic kidney disease (CKD) is an important risk factor for cardiovascular disease and mortality and can progress to end-stage renal disease (ESRD), requiring dialysis or transplantation. Compared to the CKD prevalence reported from the United States, China and Japan, reports from Thailand showed that CKD prevalence fluctuated from a much lower to a high rate of occurrence, i.e. ranged from +AH44.6% to 13.8% [1–3]. The low prevalence by the Royal Thai Air Force study would suggest that CKD is not a major public health problem in Thailand. On the other hand, methodological concerns with the Thai Air force study [1] raise doubts about the generalizability and accuracy of the current prevalence estimates. This is because the study enrolled only Thai Air Force members aged 19 to 65 years and screened for macroproteinuria. However, some of the low prevalence rates of CKD in Thailand that are reported in the literature seem at odds with the relatively high prevalence rates for treated ESRD of 220–286 per million population [4]. An accurate estimate of CKD magnitude is crucial for its ranking as a public health priority in Thailand. Furthermore, accurate data are essential for adequate allocation of educational resources and awareness programs, designing screening strategies and planning of nephrological resources for the care of CKD patients.

The objectives of the current study were to perform a cross-sectional survey to estimate CKD prevalence and to identify predictors of CKD using a standardized method for GFR estimation in representative Thai adult populations. We took advantage of methods utilized in other countries as part of the Global Screening and Early Evaluation of Kidney Disease (SEEK) program.

## Materials and methods

### Study design and subjects

A community-based cross-sectional survey study was conducted between August 2007 and June 2008. Thai male and female subjects with the following criteria were included: aged 18 or older, no menstruation period and no fever for at least a week before examination date and willingness to participate and provide a signed consent form. Subjects were excluded if blood or urine specimens were not taken. The study was approved by three Thai Institutional Review Boards (IRB), i.e. the IRB of the Faculty of Medicine at Ramathibodi Hospital, Mahidol University, the IRB of the Ministry of Public Health and as a part of the Global SEEK program by Partners Healthcare IRB in Boston, USA.

### Sampling method

Four regions of Thailand (i.e. Northern, Northeastern, Central and Southern) and Bangkok (metropolitan) were treated as strata. Stratified-cluster random sampling was applied to selected subjects. At the first stage, two to three provinces in each region were randomly selected. Each selected province was next classified as either an urban or rural area, and then one district from each area was randomly selected (i.e. the second-stage sampling unit). There were, in total, 10 provinces (i.e. Bangkok, Cholburi, Lopburi, Payau, Prae, Sakolnakorn, Nong-Bau Lamphu, Mahasarakam, Puket and Songkhla), and 20 districts were chosen for study sampling. Finally, subjects in each sampled district were randomly selected stratified by age group (i.e. 18–30, 31–40, 41–50, 51–60, 61–70 and >70 years) and gender. Data registry of health coverage of local hospitals were retrieved and used for this sampling.

### Sample size estimation

Sample size estimation was performed based on an estimation of the prevalence of CKD, which varied from 3% to 13.7% as per previous reports [1,2,5,6]. Type 1 error, confidence interval width and expected prevalence were set at 5%, +AH4±2.1% and 13.7%, respectively. A design effect for stratified-cluster sampling was set as three times higher than simple random sampling. If 3000 subjects were enrolled, the expected prevalence rate would lie between 11.7% and 16.0%. The percentage of missing data was set at 10%, and thus at least 3300 subjects were needed for sampling. The sample size for each district–province was calculated proportionally to their census population as of 2007 [7].

### Data collection and measurements

Data collection was performed in 10 provinces at 20 camp sites. Data collection teams at each site consisted of five to seven nurses and technicians and 15 interviewers led by a nephrologist. Teams were given orientation and informed of their responsibilities 1 day in advance of camp set-up. Subjects were interviewed by well-trained interviewers using standard questionnaires. Arterial blood pressure was measured twice after a rest for 15 min using mercury sphygmomanometers. The mean of the two measurements was used in the analysis. Physical examinations (e.g. weight, height, waist and hip circumferences) were performed by nurses. All subjects were instructed to maintain an 8-h, overnight fast before performing blood chemistry tests (e.g. plasma glucose, lipid profile, haemoglobin, uric acid) and urine collection the following day. Blood analyses were processed (i.e. centrifuged and separated) and transported at controlled temperatures (4–9°C) within 24 h to the Central Laboratory Department at Ramathibodi Hospital, Bangkok. Urine analysis and supernatant were performed at the local camp site.

All blood chemistry was measured using the Dimension RxL MAX analyser (Siemens Healthcare Diagnostics, USA). The Ramathibodi Central Laboratory Department was certified by the US Centers for Disease Control (National Heart, Lung and Blood Institute Lipid Standardization Program).

Urine albumin was measured using the immunoturbidimetry technique with the COBAS INTEGRA 700 analyser (Roche Diagnostics, USA).

### Creatinine gas chromatography/isotope dilution mass spectrometry traceability

Serum creatinine was measured using the modified Jaffe method, and three-levels were calibrated using CHEM I calibrators provided by the manufacturer.

In addition, serum creatinine was further standardized using the SRM-967 as a calibrator. Regression analysis was applied to construct a calibration equation between the isotope dilution mass spectrometry (IDMS) and the modified Jaffe methods. This equation, based on the conventional unit (CU) unit, was finally used to calibrate the whole modified Jaffe serum creatinine to the IDMS serum creatinine as shown below:

$IDMS=−0.0067+0.9525×MJ$

The glomerular filtration rate (GFR) was then calculated using the modification of diet in renal disease (MDRD) equation for the IDMS traceable serum creatinine values as follows [8]: estimated glomerular filtration rate (eGFR) (in millilitres per minute per 1.73 m2) = 175 × (serum creatinine)1.154 × (age)0.203 × (0.742 if female).

### Definitions

CKD staging was defined considering kidney function with or without kidney damage [9] as follows: CKD Stages I and II were defined as GFR >90 and 60–89 ml/min/1.73 m2, respectively, with haematuria and/or albumin–creatinine ratio 30 mg/g or greater, whereas Stages III, IV and V were defined as GFR 30–59, 15–29 and <15 ml/min/1.73 m2, regardless of kidney damage.

Haematuria was defined by the presence of more than five red blood cells per high-power ocular field in spun urine sediment. Microalbuminuria was defined as having an albumin–creatinine ratio of 30 to 300 mg/g without regard to gender.

The classification of subjects with hypertension, diabetes and high cholesterol was based on history, relevant medicines used, blood tests and physical examinations. For instance, subjects were classified as having diabetes if they had one of the following criteria: self-reported as being told by doctors that they had diabetes, if they were taking oral hypoglycaemic agents or fasting plasma glucose levels ≥126 mg/dl. Subjects were classified as having hypertension if they were told by doctors, taking antihypertensive drug(s) or had systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg. Anaemia was diagnosed if subjects had haemoglobin levels <11 g/dl.

### Statistical analysis

Data record forms were transported to the central data management centre, Section for Clinical Epidemiology and Biostatistics, Faculty of Medicine, Ramathibodi Hospital. Quality control programs were developed while doing data entry. EpiData version 3.1 was used for all databases.

Data were respectively described using the mean (±SE) and proportion (±SE) for continuous and categorical data. The prevalence of CKD was estimated according to the sampling methods. Three-stage sampling weight was applied to estimate prevalence using data from the Thai population (2007 census), Ministry of the Interior [7]. The weight was calculated by 1/[(probability of sampling provinces) × (probability of sampling districts) × (probability of sampling subjects)] in which probabilities of sampling were calculated as follows: The probability of sampling provinces was estimated by the number of sample provinces divided by the total number of provinces in that stratum (region). The probability of sampling districts was calculated by the numbers of sample districts divided by the total numbers of districts in the sample province. Finally, the probability of sampling subjects was estimated by using the number of subjects divided by the size of the population of the sample district. Overall CKD prevalence and gender-, age- and region-specific prevalence were estimated along with a 95% confidence interval (95% CI).

Factors associated with CKD were assessed using simple logistic regression for survey data analysis. Factors with P-values <0.15 were simultaneously included into the multivariate logistic model. Adjusted odds ratios (OR) and 95% CI were estimated. Goodness of fit of the model was assessed using chi-square test. All analyses were performed using STATA 10.1.

## Results

Three thousand, four hundred and fifty-nine subjects were included in the study. The characteristics of the subjects are described in Table 1. The mean age of subjects was 45.2 years (SE = 0.8), 54.5% were females and 36.1% had a BMI of 25 kg/m2 or higher. Mean waist–hip ratio (WHR) was 0.8 (SE = 0.01). For blood chemistry tests, the mean fasting plasma glucose, cholesterol level, uric acid and haemoglobin were 99.96 (SE = 0.73), 204.59 (SE = 1.41), 5.32 (SE = 0.03) and 13.36 (SE = 0.09), respectively. The prevalence of diabetes mellitus was 11.9%, whereas the prevalence of hypertension and high cholesterol were as high as 27.5% and 26.4%, respectively. Mean serum creatinine in males and females were 1.1 mg/dl (SE = 0.02) and 0.8 mg/dl (0.02), respectively. A history of taking non-steroidal anti-inflammatory drugs (NSAIDs) and traditional medicines was reported by 44.7% and 33.5%, respectively.

Table 1

General characteristics of subjects enrolled in the Thai SEEK study

Characteristics Number (n = 3459) % (SE)
Age, year, mean (SE)  45.19 (0.79)
Sex
Male 1569 45.46 (0.02)
Female 1890 54.54 (0.02)
Income, Baht
≤2000 264 6.71 (0.01)
2001–5000 1106 27.96 (0.04)
5001–10 000 935 28.10 (0.01)
10 001–15 000 455 14.29 (0.01)
>15 000 629 21.00 (0.04)
No income 62 1.93 (0.00)
Education
Primary 1985 55.69 (0.03)
Secondary 975 29.09 (0.02)
Diploma 148 4.65 (0.01)
Bachelor's degree 194 6.17 (0.01)
Master's degree 12 0.39 (0.00)
None 128 4.01 (0.01)
BMI, kg/m2, mean (SE)  24.03 (0.21)
<25 2250 63.86 (0.02)
25–29 924 27.26 (0.02)
≥30 285 8.88 (0.01)
WHR, mean (SE)  0.84 (0.01)
Smoke, cigarettes per day
0 2194 65.99 (0.02)
1–10 823 23.9 (1.02)
>10 331 10.09 (0.01)
Alcohol
Current 1596 45.19 (0.02)
Ever 488 13.77 (0.01)
Never 1360 41.04 (0.03)
Exercise
Yes 2057 59.85 (0.03)
No 1390 40.15 (0.03)
Work involve significant physical activity
Yes 2115 57.96 (0.05)
No 1296 42.04 (0.05)
Underlying disease
History of diabetes
Yes 331 8.88 (0.01)
No 3098 91.12 (0.01)
Fasting plasma glucose, mg/dl, mean (SE)  99.96 (0.73)
≥126 276 7.66 (0.01)
<126 3183 92.34 (0.01)
Diabetes
Yes 434 11.92 (0.01)
No 3025 88.08 (0.01)
History of hypertension
Yes 563 16.47 (0.01)
No 2887 83.53 (0.01)
SBP ≥140 or DBP ≥90 mmHg
Yes 676 19.74 (0.02)
No 2783 80.26 (0.02)
Hypertension
Yes 955 27.52 (0.02)
No 2504 72.48 (0.02)
History of high cholesterol
Yes 300 11.62 (0.03)
No 2574 88.38 (0.03)
Cholesterol level, mg/dl, mean (SE)  204.59 (1.41)
≥240 642 19.90 (0.02)
<240 2816 80.10 (0.02)
Abnormal cholesterol
Yes 851 26.39 (0.03)
No 2608 73.61 (0.03)
History of heart disease
Yes 118 3.37 (0.00)
No 3124 96.63 (0.00)
Cerebrovascular accident
Yes 44 1.35 (0.00)
No 3385 98.65 (0.00)
Anaemia
Yes 269 6.97 (0.01)
No 3129 93.03 (0.01)
History of kidney stone
Yes 169 5.04 (0.00)
No 3085 94.96 (0.00)
LDL, mg/dl
<130 1979 57.55 (0.02)
130–159 807 23.27 (0.01)
≥160 591 19.19 (0.02)
Serum creatinine, mg/dl, mean (SE)
Male 1569 1.07 (0.02)
Female 1890 0.83 (0.02)
Uric acid, mg/dl, mean (SE)  5.32 (0.03)
Haemoglobin, mg/dl, mean (SE)  13.36 (0.09)
NSAIDs
Yes 1577 44.71 (0.02)
No 1882 55.29 (0.02)
Yes 1143 33.49 (0.02)
No 2300 66.51 (0.02)
Family history
Diabetes
Yes 860 27.21 (0.02)
No 2316 72.79 (0.02)
Hypertension
Yes 1014 34.08 (0.01)
No 1976 65.92 (0.01)
High cholesterol
Yes 265 10.77 (0.02)
No 2358 89.23 (0.02)
Heart disease
Yes 382 12.80 (0.01)
No 2770 87.20 (0.01)
Cerebrovascular accident
Yes 125 4.14 (0.00)
No 3058 95.86 (0.00)
Anaemia
Yes 145 4.19 (0.00)
No 2807 95.81 (0.00)
Stone
Yes 289 8.15 (0.01)
No 2826 91.85 (0.01)
Renal disease
Yes 230 7.26 (0.00)
No 2820 92.74 (0.00)
Characteristics Number (n = 3459) % (SE)
Age, year, mean (SE)  45.19 (0.79)
Sex
Male 1569 45.46 (0.02)
Female 1890 54.54 (0.02)
Income, Baht
≤2000 264 6.71 (0.01)
2001–5000 1106 27.96 (0.04)
5001–10 000 935 28.10 (0.01)
10 001–15 000 455 14.29 (0.01)
>15 000 629 21.00 (0.04)
No income 62 1.93 (0.00)
Education
Primary 1985 55.69 (0.03)
Secondary 975 29.09 (0.02)
Diploma 148 4.65 (0.01)
Bachelor's degree 194 6.17 (0.01)
Master's degree 12 0.39 (0.00)
None 128 4.01 (0.01)
BMI, kg/m2, mean (SE)  24.03 (0.21)
<25 2250 63.86 (0.02)
25–29 924 27.26 (0.02)
≥30 285 8.88 (0.01)
WHR, mean (SE)  0.84 (0.01)
Smoke, cigarettes per day
0 2194 65.99 (0.02)
1–10 823 23.9 (1.02)
>10 331 10.09 (0.01)
Alcohol
Current 1596 45.19 (0.02)
Ever 488 13.77 (0.01)
Never 1360 41.04 (0.03)
Exercise
Yes 2057 59.85 (0.03)
No 1390 40.15 (0.03)
Work involve significant physical activity
Yes 2115 57.96 (0.05)
No 1296 42.04 (0.05)
Underlying disease
History of diabetes
Yes 331 8.88 (0.01)
No 3098 91.12 (0.01)
Fasting plasma glucose, mg/dl, mean (SE)  99.96 (0.73)
≥126 276 7.66 (0.01)
<126 3183 92.34 (0.01)
Diabetes
Yes 434 11.92 (0.01)
No 3025 88.08 (0.01)
History of hypertension
Yes 563 16.47 (0.01)
No 2887 83.53 (0.01)
SBP ≥140 or DBP ≥90 mmHg
Yes 676 19.74 (0.02)
No 2783 80.26 (0.02)
Hypertension
Yes 955 27.52 (0.02)
No 2504 72.48 (0.02)
History of high cholesterol
Yes 300 11.62 (0.03)
No 2574 88.38 (0.03)
Cholesterol level, mg/dl, mean (SE)  204.59 (1.41)
≥240 642 19.90 (0.02)
<240 2816 80.10 (0.02)
Abnormal cholesterol
Yes 851 26.39 (0.03)
No 2608 73.61 (0.03)
History of heart disease
Yes 118 3.37 (0.00)
No 3124 96.63 (0.00)
Cerebrovascular accident
Yes 44 1.35 (0.00)
No 3385 98.65 (0.00)
Anaemia
Yes 269 6.97 (0.01)
No 3129 93.03 (0.01)
History of kidney stone
Yes 169 5.04 (0.00)
No 3085 94.96 (0.00)
LDL, mg/dl
<130 1979 57.55 (0.02)
130–159 807 23.27 (0.01)
≥160 591 19.19 (0.02)
Serum creatinine, mg/dl, mean (SE)
Male 1569 1.07 (0.02)
Female 1890 0.83 (0.02)
Uric acid, mg/dl, mean (SE)  5.32 (0.03)
Haemoglobin, mg/dl, mean (SE)  13.36 (0.09)
NSAIDs
Yes 1577 44.71 (0.02)
No 1882 55.29 (0.02)
Yes 1143 33.49 (0.02)
No 2300 66.51 (0.02)
Family history
Diabetes
Yes 860 27.21 (0.02)
No 2316 72.79 (0.02)
Hypertension
Yes 1014 34.08 (0.01)
No 1976 65.92 (0.01)
High cholesterol
Yes 265 10.77 (0.02)
No 2358 89.23 (0.02)
Heart disease
Yes 382 12.80 (0.01)
No 2770 87.20 (0.01)
Cerebrovascular accident
Yes 125 4.14 (0.00)
No 3058 95.86 (0.00)
Anaemia
Yes 145 4.19 (0.00)
No 2807 95.81 (0.00)
Stone
Yes 289 8.15 (0.01)
No 2826 91.85 (0.01)
Renal disease
Yes 230 7.26 (0.00)
No 2820 92.74 (0.00)

SBP, systolic blood pressure; DBP, diastolic blood pressure.

### Prevalence of CKD

The prevalence of CKD was estimated according to gender (Table 2). Among 3459 subjects, 626 subjects were classified as having CKD and resulted in an overall CKD prevalence of 17.5% (95% CI = 14.6–20.4%). CKD Stages I and II (8.5% and 9.3% in males and females, respectively) were as high as the aggregates of Stage III, IV and V (7.8% and 9.3% in males and females).

Table 2

Estimation of CKD prevalence according to gender

Gender n CKD staging Overall
II III IV
No.a Prevalenceb (%) No. Prevalence (%) No. Prevalence (%) No. Prevalence (%) No. Prevalence (%)
Male 1569 51 2.6 100 5.8 104 6.9 15 0.9 270 16.3
(1.4, 3.8) (4.0, 7.6) (4.9, 8.9) (0.5, 1.3) (12.5, 20.0)
Female 1890 83 3.8 107 5.4 144 8.0 22 1.3 356 18.5
(2.8, 4.9) (3.5, 7.4) (6.0, 9.9) (0.6, 2.0) (14.8, 22.3)
Overall 3459 134 3.3 207 5.6 248 7.5 37 1.1 626 17.5
(2.5, 4.1)   (4.2, 7.0)  (6.2, 8.8)  (0.7, 01.5) (14.6, 20.4)
8.9 (6.8, 11.0) 8.6 (7.0, 10.3)
Gender n CKD staging Overall
II III IV
No.a Prevalenceb (%) No. Prevalence (%) No. Prevalence (%) No. Prevalence (%) No. Prevalence (%)
Male 1569 51 2.6 100 5.8 104 6.9 15 0.9 270 16.3
(1.4, 3.8) (4.0, 7.6) (4.9, 8.9) (0.5, 1.3) (12.5, 20.0)
Female 1890 83 3.8 107 5.4 144 8.0 22 1.3 356 18.5
(2.8, 4.9) (3.5, 7.4) (6.0, 9.9) (0.6, 2.0) (14.8, 22.3)
Overall 3459 134 3.3 207 5.6 248 7.5 37 1.1 626 17.5
(2.5, 4.1)   (4.2, 7.0)  (6.2, 8.8)  (0.7, 01.5) (14.6, 20.4)
8.9 (6.8, 11.0) 8.6 (7.0, 10.3)
a

Number of CKD patients from our samples.

b

Standardized prevalence to the national distribution for that gender.

### Age- and gender-specific prevalence

CKD prevalence was also estimated according to age and gender groups. We observed higher rates of CKD with increased age in both males and females (Figure 1). Extrapolating the CKD prevalence observed in our study to the Thai general population using the Thai Census Data in 2008[10], there were 15 065 000 adult males and 24 249 000 adult females in Thailand. Our estimates demonstrated that the number of male subjects with CKD aged <40, 40–59, 60–69 and ≥70 years were 392 000 (95% CI = 192 000–591 000), 884 000 (95% CI = 723 000–1 046 000), 416 000 (95% CI = 313 000–518 000) and 471 000 (95% CI = 257 000–686 000), respectively, and for females with the corresponding age ranges, 952 000 (95% CI = 556 000–1 348 000), 1 531 000 (95% CI = 1 233 000–1 830 000), 812 000 (95% CI = 494 000–1 131 000) and 971 000 (95% CI = 598 000–1 344 000), respectively.

Fig. 1

Age-specific CKD prevalence by sex.

Fig. 1

Age-specific CKD prevalence by sex.

### CKD prevalence by region

The prevalence of CKD was estimated by region (Figure 2). CKD was highest in Bangkok (23.9%, 95% CI = 22.1–25.8%), followed by the Northeastern (22.2%, 95% CI = 17.7–26.8%) and the Northern regions (20.4%, 95% CI = 18.7–22.1%). The prevalence in the Central and Southern regions were approximately 13% each.

Fig. 2

Prevalence of CKD by region.

Fig. 2

Prevalence of CKD by region.

### Factors associated with CKD

A univariate analysis was conducted to assess the association between CKD and risk factors. Subjects with CKD Stage I–V were aggregated and compared with normal subjects. The odds of having CKD were estimated for 16 factors: age, gender, body mass index (BMI), WHR, smoking, alcohol, exercise, involvement in a significant working activity, low-density lipoprotein (LDL), cholesterol, uric acid, diabetes, hypertension, history of kidney stone, use of traditional medicine and NSAIDs (Table 3). In addition, family history (i.e. father, mother and sibling) of chronic diseases (i.e. high cholesterol, diabetes, hypertension, heart disease and kidney stone) were also assessed (data were not shown).

Table 3

Assessing factors associated with CKD

Stages I–V Normal OR (95% CI) P-value OR (95% CI) P-value
Number Number
Age, year
≥70 139 22.26 128 4.08 14.83 (8.46, 25.99) <0.001 7.34 (4.18, 12.90) <0.001
60–69 148 22.85 255 9.40 6.60 (4.22, 10.33) <0.001 3.63 (2.26, 5.86) 0.001
40–59 237 39.19 1227 43.85 2.43 (1.77, 3.33) 0.001 1.71 (1.16, 2.52) 0.017
<40 102 15.70 1223 42.67
History of kidney stone
Yes 74 11.30 95 3.72 3.30 (2.09, 5.21) 0.001 2.72 (1.80, 4.12) 0.002
No 516 88.70 2569 96.28
Diabetes
Yes 183 28.48 251 8.40 4.34 (2.87, 6.55) <0.001 2.72 (1.57, 4.73) 0.005
No 443 71.52 2582 91.60
Hypertension
Yes 329 53.60 626 21.99 4.10 (2.94, 5.72) <0.001 1.96 (1.44, 2.67) 0.002
No 297 46.40 2207 78.01
Uric acid, mg/dl
>5.61 331 55.03 938 35.09 2.68 (1.79, 4.01) 0.001 2.87 (1.77, 4.64) 0.002
4.40–5.61 166 26.58 960 33.49 1.36 (0.89, 2.07) 0.123 1.50 (0.92, 2.46) 0.087
<4.40 129 18.39 935 31.42
Yes 263 42.65 880 31.55 1.61 (1.38, 1.89) 0.001 1.20 (1.02, 1.42) 0.035
No 361 57.35 1939 68.45
Sex
Female 356 57.77 1534 53.86 1.17 (0.85, 1.61) 0.253 1.70 (1.18, 2.43) 0.013
Male 270 42.23 1299 46.14
BMI, kg/m2
≥30 65 11.71 220 8.29 1.59 (1.15, 2.20) 0.014
25–29.9 191 30.45 733 26.58 1.29 (1.01, 1.65) 0.045
<25 370 57.84 1880 65.13
Waist/hip
Male Female
≥0.96 ≥0.90 117 18.57 237 7.37 2.87 (1.70, 4.83) 0.004
<0.96 <0.90 509 81.43 2595 92.63
Smoking, cigarette per day
1–10 157 23.53 666 25.73 1.10 (0.78, 1.53) 0.191
>10 53 10.46 278 8.34 0.80 (0.54, 1.18) 0.516
0 391 66.01 1803 65.93
Alcohol consumption
Yes 326 49.51 1758 60.97 1.59 (1.02, 2.49) 0.044
No 299 50.49 1061 39.03
Exercise
Yes 380 61.87 1677 59.42 1.11 (0.94, 1.30) 0.164
No 242 38.13 1148 40.58
Work involve significant activity
Yes 323 49.18 1792 59.85 1.54 (1.13, 2.11) 0.016
No 297 50.82 999 40.15
Abnormal cholesterol
Yes 203 34.31 648 24.71 1.59 (1.21, 2.09) 0.007
No 423 65.69 2185 75.29
LDL, mg/dl
≥160 116 20.89 475 18.83 1.11 (0.78, 1.56) 0.490
130–159 134 21.39 673 23.66 0.90 (0.73, 1.11) 0.263
<130 359 57.72 1620 57.51
NSAIDs
Yes 308 48.33 1269 43.94 1.19 (0.83, 1.72) 0.266
No 318 51.67 1564 56.06
Stages I–V Normal OR (95% CI) P-value OR (95% CI) P-value
Number Number
Age, year
≥70 139 22.26 128 4.08 14.83 (8.46, 25.99) <0.001 7.34 (4.18, 12.90) <0.001
60–69 148 22.85 255 9.40 6.60 (4.22, 10.33) <0.001 3.63 (2.26, 5.86) 0.001
40–59 237 39.19 1227 43.85 2.43 (1.77, 3.33) 0.001 1.71 (1.16, 2.52) 0.017
<40 102 15.70 1223 42.67
History of kidney stone
Yes 74 11.30 95 3.72 3.30 (2.09, 5.21) 0.001 2.72 (1.80, 4.12) 0.002
No 516 88.70 2569 96.28
Diabetes
Yes 183 28.48 251 8.40 4.34 (2.87, 6.55) <0.001 2.72 (1.57, 4.73) 0.005
No 443 71.52 2582 91.60
Hypertension
Yes 329 53.60 626 21.99 4.10 (2.94, 5.72) <0.001 1.96 (1.44, 2.67) 0.002
No 297 46.40 2207 78.01
Uric acid, mg/dl
>5.61 331 55.03 938 35.09 2.68 (1.79, 4.01) 0.001 2.87 (1.77, 4.64) 0.002
4.40–5.61 166 26.58 960 33.49 1.36 (0.89, 2.07) 0.123 1.50 (0.92, 2.46) 0.087
<4.40 129 18.39 935 31.42
Yes 263 42.65 880 31.55 1.61 (1.38, 1.89) 0.001 1.20 (1.02, 1.42) 0.035
No 361 57.35 1939 68.45
Sex
Female 356 57.77 1534 53.86 1.17 (0.85, 1.61) 0.253 1.70 (1.18, 2.43) 0.013
Male 270 42.23 1299 46.14
BMI, kg/m2
≥30 65 11.71 220 8.29 1.59 (1.15, 2.20) 0.014
25–29.9 191 30.45 733 26.58 1.29 (1.01, 1.65) 0.045
<25 370 57.84 1880 65.13
Waist/hip
Male Female
≥0.96 ≥0.90 117 18.57 237 7.37 2.87 (1.70, 4.83) 0.004
<0.96 <0.90 509 81.43 2595 92.63
Smoking, cigarette per day
1–10 157 23.53 666 25.73 1.10 (0.78, 1.53) 0.191
>10 53 10.46 278 8.34 0.80 (0.54, 1.18) 0.516
0 391 66.01 1803 65.93
Alcohol consumption
Yes 326 49.51 1758 60.97 1.59 (1.02, 2.49) 0.044
No 299 50.49 1061 39.03
Exercise
Yes 380 61.87 1677 59.42 1.11 (0.94, 1.30) 0.164
No 242 38.13 1148 40.58
Work involve significant activity
Yes 323 49.18 1792 59.85 1.54 (1.13, 2.11) 0.016
No 297 50.82 999 40.15
Abnormal cholesterol
Yes 203 34.31 648 24.71 1.59 (1.21, 2.09) 0.007
No 423 65.69 2185 75.29
LDL, mg/dl
≥160 116 20.89 475 18.83 1.11 (0.78, 1.56) 0.490
130–159 134 21.39 673 23.66 0.90 (0.73, 1.11) 0.263
<130 359 57.72 1620 57.51
NSAIDs
Yes 308 48.33 1269 43.94 1.19 (0.83, 1.72) 0.266
No 318 51.67 1564 56.06

All of these factors with the exception of gender, LDL, smoking, exercise and NSAIDs were associated with CKD in the univariate analysis. Therefore, those 11 factors were considered in the multiple logistic model. Since BMI and WHR were highly correlated (r = 0.78, P ≤ 0.001), including them together in the same model would result in multicollinearity. Because WHR was better at explaining the prevalence of CKD compared with the BMI (F test = 26.8, df (1,5), P = 0.0035 for WHR; F test = 7.7, df (2,4), P = 0.0429 for BMI), it was chosen in the multivariate model. After adjustment, only seven variables: age, gender, hypertension, diabetes, high uric acid, use of traditional medicine and history of kidney stone were shown to be significant predictors of CKD.

## Discussion

The main finding of this study is that 17.5% of a representative cross-section of the Thai population was observed to have CKD. CKD prevalence in Thailand was estimated based on a community-based cross-sectional study design with stratified-cluster random sampling. The prevalence of CKD Stages I, II, III and IV were observed as 3.3%, 5.6%, 7.5% and 1.1%, respectively. Prevalence was not much different between males and females, but there was a trend toward prevalence with increasing age. The prevalence of CKD was higher in Bangkok, the Northern and Northeastern regions compared to the outlying Central and Southern regions. While the prevalence of CKD was remarkably high in Thailand, awareness of CKD in the Thai general population was quite low: only 1.9% were aware that they had CKD. Predictors of CKD not only included, as expected, older age, diabetes, hypertension, hyperuricaemia and gender but also a history of using traditional medicines and a history of kidney stones.

Our reporting of the prevalence of CKD was similar to previous studies conducted in the Beijing population aged 40 years or older, which found that the prevalence of Stages I and II was about 8% [11]. There was, however, one study in the Thai population which found very low Stages I and II prevalence, i.e. only 1.5% [1]. This could be due to the fact that the study only selected members of the Royal Thai Air Force who were predominately well-educated and relatively diabetes-free (only 8% prevalence). In addition, the degree of Stages I–II prevalence might have been underestimated since it was diagnosed based on urine protein, regardless of microalbuminuria and haematuria. Our finding of the prevalence of Stage III or higher (8.6%) was somewhat higher than the study in Taiwan (6.9%) [5] and previous reports in the Thai population (3.1% [1] and 6.8% [2]), but quite close to the pooled prevalence in Asian studies using meta-analysis (9.3% [12]). However, the figure was lower than the finding by the InterASIA Study (8.6% versus 13.2% [3]). Differences might be due to younger subjects in our study (≥ 18 versus ≥ 35 years) and creatinine measurements (IDMS versus Jaffe).

The CKD prevalence rates varied by region, i.e. highest in Bangkok (23.9%), followed by the Northeastern (22.2%) and North regions (20.4%) (Figure 2). This might be explained by the fact that, in Thailand, Bangkok has the highest prevalence of diabetes (19.0%), compared to the other regions (11–13%). In addition, although the overall prevalence of kidney stones was much lower than previously reported (16% in rural areas of northeast Thailand [13]), there appeared to be geographical variation in self-reporting of kidney stone disease. The Northeastern and North regions had higher prevalence of kidney stone (i.e. 6.4% and 8.4%, respectively) compared to the other regions (2.2–2.6%).

While the prevalence of CKD was remarkably high in Thailand, awareness of CKD in the Thai general population was quite low. Only 1.9% were aware that they had CKD. This could be due to awareness of CKD as a public health problem being new for the Thai population. In addition, previous reports of prevalence, which mostly considered Stage III or higher, appear to be underestimates and may have created a lack of concern by governmental and health organizations about the magnitude of the problem. Under-diagnosis might also be another reason for a lower awareness of CKD. General practitioners in Thailand routinely use serum creatinine to assess kidney function because it is widely available in general hospitals across the country, but do not use an eGFR prediction equation such as either the MDRD or Cockcroft–Gault equation. Using serum creatinine to estimate kidney function would underestimate the diagnosis of CKD, particularly in women and the elderly [13]. Adding eGFR along with serum creatinine in routinely reporting laboratory results should be considered. General practitioners should also become more familiar with the National Kidney Foundation Practice Guidelines in order to correctly diagnose the CKD [9]. Implementation of the guidelines should, therefore, be performed by all general practitioners and laboratories. Education should be encouraged and prevention programs should be launched in order to delay the onset of higher stages of CKD in subjects at risk within the population. Although our study could not confirm the diagnosis of CKD, the magnitude of this problem should be a signal to doctors, health care providers and policy makers to consider more aggressive means of seeking out patients with early stages of CKD.

The public health situation in Thailand differs from the situation in many westernized nations. In most western nations, the onset of end-stage kidney failure allows for dialytic support or transplantation. In most parts of the developing world, renal replacement therapy is unavailable because of resource constraints and ESRD is uniformly fatal. The economic effects are also likely to be profound. In most developing countries, strategies targeting early detection of kidney disease draw on government resources. In developing countries, however, World Health Organization (WHO) data indicate that governmental spending on health care is limited to 0.4% to 4% of the gross national product, compared with 10% to 16% in developed nations. Our paper emphasizes the magnitude of the problem of kidney disease.

### Risk factors

We observed that older age, diabetes, hypertension, hyperuricaemia and gender were associated with CKD, which was consistent with previous studies from developed countries [14–17]. Additional risk factors included history of using traditional medicines and kidney stones. Thirty percent of our studied subjects reported a history of using traditional medicines. Different traditional medicine forms were used, such as boiled medicine, powdered medicine, Chinese traditional medicine and small black tablets (called Luke Klon in Thai). These medicines have been quite popular in the general population because they are less regulated and easily accessible in terms of cost and place (e.g. general drug store, temple or even a grocery). Two reasons for using traditional medicines are to maintain well-being and relieve pain. This finding was similar to one Chinese study that reported an association from the use of Chinese herbs and CKD [18]. Two mechanisms of nephrotoxicity from the use of these traditional medicines were proposed. Firstly, they may be combined with various herbal plants (e.g. Aristolochia species, Securidaca longepedunculata, Euphorbia matabelensis, Crotalaria laburnifolia and Callilepsis laureola [19–21]) that cause renal damage. Secondly, unregulated compounds may be contaminated with drugs or heavy metals [22–24]. Evidence suggests that some Asian herbal medicines contain toxic heavy metals or undeclared prescription drugs due to either intentional alteration for medicinal purposes or accidental contamination. The role of kidney stones associated with CKD is still unclear, although this was consistent with previous studies [25,26]. Patients with kidney stones should have frequent, careful monitoring to check for CKD.

### Methodological issues

Our study has both strengths and weaknesses. We selected subjects using a three-stage stratified-cluster random sampling and thus we believe that our study should be representative of the entire adult Thai population. Our analyses were performed according to survey sampling in which the probabilities of subject selection for the three stages were taken into account. We standardized the modified Jaffe creatinine to IDMS standards, aiming for a standardized and validated method for calculation of kidney function.

Our study also had limitations. The diagnosis of CKD was based on single measures of proteinuria and serum creatinine. An important but unexpected finding of our study was that CKD Stages I and II accounted for nearly a half of the total cases of CKD. These early stages are known as silent and asymptomatic stages [29]. Previous studies [8,20] showed that persistent positive rates for microalbuminuria were only 50.9–53% and 66.6–75% for CKD Stages I and II, respectively. Correcting our results with the persistent rates would, respectively, yield CKD Stages I, II and overall prevalence rates of about 1.7%, 4.1% and 14.4%. Another limitation is that we used the MDRD GFR prediction equation. This equation has not been validated for the Thai population and this could have accounted for the lack of precision in estimating GFR. Applying the CKD-EPI equation [30], which is regarded as more accurate in estimating GFR for higher value than the MDRD, one would identify only approximately 2% more CKD Stage I than the MDRD could. The prevalence for CKD-EPI Stages I, II, III and IV–V were 5.0%, 4.3%, 5.5% and 1.0%, respectively, with the overall prevalence of 15.9% (95% CI = 12.1–19.7%). Finally, although we applied stratified-cluster sampling to select subjects across the country, it is possible that our sampling methodology was not representative of the Thai population. We considered this possibility because the mean age of our sample (mean = 45.3, SD = 15.4) was approximately 4 years older than that of the general population with the same age range (mean = 41.2, SD = 15.7). We performed a sensitivity analysis to test whether this altered our results. We calculated age-specific standardized CKD prevalence in the Thai population using the same age groups as the reference population. This yielded the overall CKD prevalence of 15.0% (95% CI = 11.7–18.4%), which was not much different compared to our reported CKD prevalence (i.e. 17.5%, 95% CI = 14.6–20.4%). We had also re-calculated the prevalence by assuming that subjects were approximately 4 years younger: the adjusted CKD prevalence was 16.9% (95% CI = 13.6–20.4%), which was also similar to our reported CKD prevalence rate for Thailand.

In summary, CKD prevalence in the Thai population is much higher than previously published (17.5% versus 4.6–13.8%). The early stages of CKD are as common as later stages. However, albuminuria measurement was not confirmed, and adjusting for persistent positive rates resulted in the prevalence of 14.4%. There is some geographical variation in the prevalence of CKD in Thailand. Among predictors of CKD, exposure to traditional medicines seems to be important. Finally, there is low awareness of CKD in the general population. Our recommendation is that, in light of the high rate of CKD, a screening strategy for the early recognition of CKD should be launched to prevent further progression of the disease. This should be initiated by targeting high-risk populations. A clinical prediction score model should be further developed to aid in identifying high-risk populations. Furthermore, considering the high prevalence of predisposing factors like hypertension (27.5%), diabetes (11.9%) and smoking (20.5%) in the screened population, education strategies for awareness of CKD among those high-risk populations should be developed. Also, continuing education for primary health care providers who routinely take care of patients with these diseases should be offered to instruct the providers to test for serum creatinine, estimate GFR and perform simple urine testing for proteinuria for the early detection of ESRD.

The Thai-SEEK group members: Anchalee Chitthamma, Ph.D., Division of Clinical Chemistry, Department of Pathology, Faculty of Medicine, Ramathibodi Hospital; Chaiyot, MD., Chon Buri Hospital; Osot Nerapusee, Pharmacist, Medical Affairs Director, Thai Janssen-Cilag Ltd; Talerngsak Kajanabuch, MD, Faculty of Medicine, Chulalongkorn University; Porntip Chatchaipan, M.Sc., Section for Clinical Epidemiology and Biostatistics, Faculty of Medicine, Ramathibodi Hospital; Sasivimol Rattanasiri, Ph.D., Section for Clinical Epidemiology and Biostatistics, Faculty of Medicine, Ramathibodi Hospital; Thananda Trakarnvanich, MD, Renal Unit, Department of Medicine, Bangkok Metropolitan Administration Medical College and Vajira Hospital; Vuddhidej Ophascharoensuk, MD, Faculty of Medicine, Chiang-Mai University Hospital; Warangkana Pichaiwong, MD, Division of Nephrology, Department of Medicine, Faculty of Medicine, Rajvithee Hospital, Bangkok, Thailand; and The Nephrology Society of Thailand and Thai Janssen-Cilag Ltd. This study was partly granted by Thai Janssen-Cilag Ltd and National Health Security Office of Thailand. A.I. has received a travel fund from Pfizer. A.T. has received honorarium from this project. A.S. has received research grants from Johnson and Johnson, Watson, Roche and Amgen.

Conflict of interest statement. None declared.

The original version was incorrect. The equation of page 2 has been corrected.

## References

1
Chittinandana
A
Chailimpamontree
W
Chaloeiphap
P
Prevalence of chronic kidney disease in Thai adult population
J Med Assoc Thai
,
2006
, vol.
89
(pg.
S112
-
S120
)
2
Domrongkitchaiporn
S
Sritara
P
Kitiyakara
C
, et al.  .
Risk factors for development of decreased kidney function in a Southeast Asian population: a 12-year cohort study
J Am Soc Nephrol
,
2005
, vol.
16
(pg.
791
-
799
)
3
Perkovic
V
Cass
A
Patel
AA
, et al.  .
High prevalence of chronic kidney disease in Thailand
Kidney Int
,
2008
, vol.
73
(pg.
473
-
479
)
4

International comparisons of ESRD. http://www.usrds.org/2008/pdf/V2_12_2008.pdf (23 December 2008, date last accessed)
5
Hsu
CC
Hwang
SJ
Wen
CP
, et al.  .
High prevalence and low awareness of CKD in Taiwan: a study on the relationship between serum creatinine and awareness from a nationally representative survey
Am J Kidney Dis
,
2006
, vol.
48
(pg.
727
-
738
)
6
Tanaka
H
Shiohira
Y
Uezu
Y
, et al.  .
Metabolic syndrome and chronic kidney disease in Okinawa, Japan
Kidney Int
,
2006
, vol.
69
(pg.
369
-
374
)
7

Department of Provincial Administration, Ministry of Interior. http://www.dopa.go.th/ (2008, date last accessed)
8
Coresh
J
Selvin
E
Stevens
LA
, et al.  .
Prevalence of chronic kidney disease in the United States
JAMA
,
2007
, vol.
298
(pg.
2038
-
2047
)
9
Levey
AS
Eckardt
KU
Tsukamoto
Y
, et al.  .
Definition and classification of chronic kidney disease: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO)
Kidney Int
,
2005
, vol.
67
(pg.
2089
-
2100
)
10

Office of the National Economic and Social Development Board, Office of the Prime Minister. http://www.nesdb.go.th/Default.aspx?tabid=96 (20 November 2008, date last accessed)
11
Li
ZY
Xu
GB
Xia
TA
, et al.  .
Prevalence of chronic kidney disease in a middle and old-aged population of Beijing
Clin Chim Acta
,
2006
, vol.
366
(pg.
209
-
215
)
12
Anothaisintawee
T
Rattanasiri
S
Ingsathit
A
, et al.  .
Prevalence of chronic kidney disease: a systematic review and meta-analysis
Clin Nephrol
,
2009
, vol.
71
(pg.
244
-
254
)
13
Yanagawa
M
Kawamura
J
Onishi
T
, et al.  .
Incidence of urolithiasis in northeast Thailand
Int J Urol
,
1997
, vol.
4
(pg.
537
-
540
)
14
Yiengprugsawan
V
Lim
LL
Carmichael
GA
, et al.  .
Tracking and decomposing health and disease inequality in Thailand
Ann Epidemiol
,
2009
, vol.
19
(pg.
800
-
807
)
15
Perrone
RD
NE
Levey
AS
Serum creatinine as an index of renal function: new insights into old concepts
Clin Chem
,
1992
, vol.
38
(pg.
1933
-
1953
)
16
Chonchol
M
Shlipak
MG
Katz
R
, et al.  .
Relationship of uric acid with progression of kidney disease
Am J Kidney Dis
,
2007
, vol.
50
(pg.
239
-
247
)
17
Elsayed
EF
Sarnak
MJ
Tighiouart
H
, et al.  .
Waist-to-hip ratio, body mass index, and subsequent kidney disease and death
Am J Kidney Dis
,
2008
, vol.
52
(pg.
29
-
38
)
18
Fox
CS
Larson
MG
Leip
EP
, et al.  .
Predictors of new-onset kidney disease in a community-based population
JAMA
,
2004
, vol.
291
(pg.
844
-
850
)
19
Haroun
MK
Jaar
BG
Hoffman
SC
, et al.  .
Risk factors for chronic kidney disease: a prospective study of 23,534 men and women in Washington County, Maryland
J Am Soc Nephrol
,
2003
, vol.
14
(pg.
2934
-
2941
)
20
Zhang
L
Zhang
P
Wang
F
, et al.  .
Prevalence and factors associated with CKD: a population study from Beijing
Am J Kidney Dis
,
2008
, vol.
51
(pg.
373
-
384
)
21
Cosyns
JP
Aristolochic acid and ‘Chinese herbs nephropathy’: a review of the evidence to date
Drug Saf
,
2003
, vol.
26
(pg.
33
-
48
)
22
Isnard Bagnis
C
Deray
G
Baumelou
A
, et al.  .
Herbs and the kidney
Am J Kidney Dis
,
2004
, vol.
44
(pg.
1
-
11
)
23
Seedat
YK
Acute renal failure in the black population of South Africa
Int J Artif Organs
,
1993
, vol.
16
(pg.
801
-
802
)
24
Diamond
JR
Pallone
TL
Acute interstitial nephritis following use of tung shueh pills
Am J Kidney Dis
,
1994
, vol.
24
(pg.
219
-
221
)
25
Ernst
E
Toxic heavy metals and undeclared drugs in Asian herbal medicines
Trends Pharmacol Sci
,
2002
, vol.
23
(pg.
136
-
139
)
26
Mao
JJ
Desai
K
Metal content in Ayurvedic medicines
JAMA
,
2009
, vol.
301
pg.
271

27
Gillen
DL
Worcester
EM
Coe
FL
Decreased renal function among adults with a history of nephrolithiasis: a study of NHANES III
Kidney Int
,
2005
, vol.
67
(pg.
685
-
690
)
28
Worcester
EM
Parks
JH
Evan
AP
, et al.  .
Renal function in patients with nephrolithiasis
J Urol
,
2006
, vol.
176
(pg.
600
-
603
discussion 3
29
Kinchen
KS
J
Fink
N
, et al.  .
The timing of specialist evaluation in chronic kidney disease and mortality
Ann Intern Med
,
2002
, vol.
137
(pg.
479
-
486
)
30
Levey
AS
Stevens
LA
Schmid
CH
, et al.  .
A new equation to estimate glomerular filtration rate
Ann Intern Med
,
2009
, vol.
150
(pg.
604
-
612
)